The Peter Attia Drive - #331 ‒ Optimizing endurance performance： metrics, nutrition, lactate, and more insights from elite performers ｜ Olav Aleksander Bu (Pt. 2)

00:00:00.000 Hey, everyone. Welcome to the Drive podcast. I'm your host, Peter Atiyah. This podcast,

00:00:16.540 my website, and my weekly newsletter all focus on the goal of translating the science of longevity

00:00:21.520 into something accessible for everyone. Our goal is to provide the best content in health and

00:00:26.720 wellness, and we've established a great team of analysts to make this happen. It is extremely

00:00:31.660 important to me to provide all of this content without relying on paid ads. To do this, our work

00:00:36.960 is made entirely possible by our members, and in return, we offer exclusive member-only content

00:00:42.700 and benefits above and beyond what is available for free. If you want to take your knowledge of

00:00:47.940 this space to the next level, it's our goal to ensure members get back much more than the price

00:00:53.200 of the subscription. If you want to learn more about the benefits of our premium membership,

00:00:58.040 head over to peteratiyahmd.com forward slash subscribe. My guest this week is Olav Alexander

00:01:06.580 Boo. Olav was a guest in March of 2024, and at the time of that conversation, I realized we hadn't got

00:01:12.840 through the majority of what I wanted to speak about, so it was inevitable that I would have him

00:01:17.280 back. Olav is an endurance coach, exercise scientist, engineer, and physiologist. He is the

00:01:22.980 head of performance for Norway Triathlon and is best known for coaching two of the world's top

00:01:28.520 triathletes, Christian Blumenfeld and Gustav Iden. In this episode, we review his work and his approach

00:01:37.380 to coaching and the way that he relies very heavily on data. We talk about and define various performance

00:01:44.120 metrics like FTP, functional threshold power, critical power, anaerobic threshold, lactate

00:01:49.940 threshold, VO2 max, and the importance of consistent protocols when testing these performance metrics

00:01:56.180 and how they can vary depending on an athlete's training. We discuss differences in training

00:02:02.280 methodologies across sports and how different sports and activities influence power, pace, and endurance.

00:02:08.680 We look at the significance of nutrition in endurance sports, how athletes train to

00:02:14.040 properly fuel themselves for races, and why this is so different from what has been done

00:02:18.840 historically. In fact, we really got into this difference in carbohydrate metabolism.

00:02:23.520 Just like the first time Olav and I spoke, this is a discussion that can be quite complex at some

00:02:28.240 points. We do get a little bit into the weeds, but the truth of it is because of the nature of what

00:02:33.320 we're talking about, it's very difficult to talk about these things meaningfully and superficially.

00:02:38.040 Patience is always appreciated and the rewards are always there if you're able to stick with it.

00:02:42.560 Without further delay, please enjoy my conversation with Olav Aleksandr Buu.

00:02:51.780 Olav, thank you very much for coming to Austin on your way. I guess you're on your way to Arizona?

00:02:57.780 Yes, on my way to Flagstaff. Also, thank you very much for having me again.

00:03:01.760 Yeah. Well, as I mentioned last time we spoke, I had a lot of notes. We got through,

00:03:06.760 I think, one-eighth of them. There's a lot to cover today. Of course, as we've been sitting here for

00:03:11.340 the last 15 minutes or so, we've already kind of started the podcast, unfortunately. And I want

00:03:15.980 to go right to where we just left off, but I'm going to resist the urge. Last time we spoke really

00:03:21.240 about sort of the most nuanced ins and outs of cardiorespiratory fitness. Maybe just for the

00:03:26.840 person who didn't catch that episode, can you give the one-minute version of what you do and why

00:03:33.640 you're certainly one of the few people that would be poised to talk about what we're going to talk

00:03:37.560 about today? I'm not very good at the pitches. My background is from engineering, and it means

00:03:42.940 also that this principle have guided me quite a lot through my journey in endurance sports or in

00:03:48.060 sports in general. We embarked on a journey 10 years ago, more than 10 years ago, I would say now 15

00:03:53.780 years ago, where we started to do what I would call extreme in-depth and longitudinal studies on

00:04:00.280 two of the arguably fittest athletes in the whole world. Yeah, I think that pretty much resembles it.

00:04:05.620 Obviously, a large part of that also involves technology development, simply because we are at

00:04:10.480 the edge of basically what we have available information on, research on. And that means

00:04:16.600 that even in some cases, we have to develop technology to allow us to even progress the

00:04:21.820 understanding of getting a more granular understanding of why things are the way they are.

00:04:27.240 Yeah. So in many ways, you're kind of an applied scientist.

00:04:29.680 Yes.

00:04:29.960 And your laboratory is both a CPET lab and then a racing environment where most of the athletes you

00:04:36.760 work with are triathletes, correct? Yes and no. I would say that it's actually quite spread. It's

00:04:41.820 a mixture between triathletes, cyclists, runners, track and field, to even sailors, which is, let's

00:04:48.400 say, on the explosive end of the domain, actually, and not endurance, but yeah.

00:04:52.160 Yeah. I feel like we're going to use a lot of terms today. We're going to probably throw out the

00:04:57.160 word anaerobic threshold, lactate threshold, VO2 max, FTP. So I just want to make sure everybody

00:05:02.240 kind of understands those things. So let's just take them one at a time. Can you define FTP or

00:05:07.600 functional threshold power for folks? There is a couple of definitions of this already,

00:05:11.920 which is, I have to say that it's bad when you have good terminologies, but they start to get diluted.

00:05:17.460 But the original definition of FTP by the authors, I think, was Andy Kogan. And that was that

00:05:24.320 basically, you first actually have to do a five-minute all-out effort. And then basically,

00:05:28.600 there's a short pause in between there. And then you go to a 20-minute all-out effort.

00:05:33.460 And then you subtract 5% from that to find your FTP. So typically, it would be your 20-minute all-out

00:05:38.920 minus 5%. And the reason for that is to try to get a ballpark-ish idea of what your, let's say,

00:05:45.460 sustainable output is possible to do over an hour. But as we have learned over the years,

00:05:51.200 is that one, we figure out that this is not as accurate always, because there are a couple of

00:05:55.960 other things in there. And also, unfortunately, there have become different ways to doing it.

00:06:00.700 Like some people, they just do a warm-up and then they do a 20-minute all-out and then they

00:06:03.920 subtract 5%. And that's already different. Yeah. We used to sometimes do 20 minutes and

00:06:08.360 subtract 10%. So we would do a gentle warm-up for an hour, then do 20 minutes and then subtract 10%.

00:06:15.460 So yes. But I guess the spirit of FTP, which is maybe what we want people to think about and not

00:06:20.900 get mired in the details, is it really approximates an energy zone that is more than just an all-out,

00:06:29.140 but clearly less than what you could hold indefinitely. And it's directionally about

00:06:33.920 the highest output you could have for an hour. And there's different ways to approximate it, of course.

00:06:38.980 Maybe more importantly to that point is that as long as you do something consistently,

00:06:43.600 you keep the same protocol, you could say, okay, this one thing is the original thinking and that

00:06:48.700 went into devising this kind of protocol. But I would say more importantly is rather to stay true

00:06:54.120 to the principle of how you do it. So as long as you do it the same way each time, this is more

00:06:59.200 important. And then how does that differ from another term that is used interchangeably, but I

00:07:05.260 believe erroneously, which is critical power? So critical power is something that you normally more

00:07:10.480 extract from that you do multiple all-out efforts. And then you apply reverse extrapolation to this

00:07:16.980 to basically figure out what is your critical power. So it's a more of a little bit more advanced

00:07:23.080 mathematical approach to it. Typically, you would say that there are different concepts to this.

00:07:27.760 I personally have to say that I like the critical power approach a little bit better.

00:07:32.440 And what is it trying to approximate?

00:07:33.980 So critical power is basically where you try to divide something into two zones. That's a

00:07:41.200 cross oversimplification, but you distinguish between, let's say, a non-severe and a severe state,

00:07:47.120 or basically where you are, again, also in the same way as FTP, trying to figure out what is the power

00:07:53.000 you are capable of staying at for a prolonged time, and basically where you go into a territory where

00:08:00.020 small changes has a huge consequence on the duration that you're capable of holding it.

00:08:06.900 And where does critical power typically lie in relation to FTP?

00:08:10.620 Again, this depends a little bit on how you test FTP, but I would say that how maybe FTP have been

00:08:16.340 used over the last, or let's say a little bit more deviated from how the authors originally devised it,

00:08:22.240 I would say that it actually is not too different. And normally you would say that critical or

00:08:27.480 functional threshold power would lie slightly lower in power output than critical power.

00:08:34.060 But again, it depends very much on, let's say, also in critical power there are today,

00:08:37.880 I don't know how many different definitions there are of critical power and how you should do the

00:08:41.720 protocol there. But it means that normally critical power, if you look at it from a metabolic perspective,

00:08:47.000 it sits somewhere between your maximum lactate steady state, or call it anaerobic threshold,

00:08:52.180 and VU2 max. So typically more close to VU2 steady state or so on, but that's introducing another term

00:08:59.660 that it just doesn't bring clarity. We'll define those in a second. So many of these metrics,

00:09:05.280 FTP, critical power, are easiest to think about in terms of cycling, because we use power meters.

00:09:11.680 Do the same concepts still apply in swimming and running?

00:09:14.860 Yes, I would say. I think for the sake of simplicity, it sometimes helps to refer to,

00:09:23.920 like say, one condensed number. So you use, for example, you can say critical pace,

00:09:27.720 for example, in running and swimming. And the way of testing it is not too different to what you

00:09:33.020 would do in cycling. But one thing that actually happens in the process is that you actually have

00:09:38.080 access to more granular information, and you are dumping down information a little bit when you

00:09:42.340 call it FTP or you call it critical power. Because more importantly, I would say, is that when you

00:09:48.140 present a critical power number only, or an FTP number only, you have taken two-dimensional

00:09:54.640 information and made it one-dimensional, meaning that you only look at the y-axis and you take away

00:09:59.720 the information that is quite critical, I would say, and that is how long are you able to sustain

00:10:03.760 something. So for example, for critical power testing, I would say, again, also when you first do

00:10:08.480 critical power testing, it's more interesting to know, let's say, how fast or how far you're able

00:10:14.200 to get over one minute, for example, five minute and 15 minutes, and then actually see what happens

00:10:19.080 in the next time you do it for the one minute, the five minute and 15 minutes, because this is

00:10:22.840 actually quite crucial to understand also the balance or what's happening with the different

00:10:27.260 balances in the body. Also FTP, most people think of this as a 20-minute power, so it's even a little

00:10:33.660 bit more condensed or even simpler information. But as we will probably come back to, is that when

00:10:40.020 you're training, there are two things that normally can happen. One is that you increase your general

00:10:43.880 capacity, meaning that both power and capacity, one or two, are increased. The other thing that

00:10:50.160 happens there is that at some point, we are all time limited, and this is where we need to pivot.

00:10:54.320 So we need to pivot or let's say, prioritize what is more important for us. Is it the more explosive

00:10:58.820 or speed or is it more endurance? And then it becomes more interesting for how you guide the

00:11:04.340 training to understand what's happening between this different power or pace and durations.

00:11:10.540 Okay. You mentioned two other terms there, anaerobic threshold. Let's start with that. How do we define

00:11:15.200 it? Probably in the same way today that we say something we just say and call, and now in the

00:11:19.440 world of chat GPT and everything, we just call something AI. Basically an extremely broad term that try

00:11:25.780 to encompass more or less the exact same thing that we already talked about. Basically, let's say the

00:11:31.020 difference between where something is steady and where it goes over to unsteady, or basically for

00:11:37.200 where you can hold something for a longer duration and where it goes to shorter duration. It's a

00:11:42.240 misconception to think that anaerobic threshold means basically when you become anaerobic, because

00:11:47.300 that's not the case. Yeah. It's a continuum, of course. It's not a switch. Yeah. Yeah.

00:11:51.480 Where does anaerobic threshold typically, if we're just limiting this to make the discussion easier

00:11:57.700 in talking about, say, cyclists with a power meter, where does anaerobic threshold tend to sit relative

00:12:02.480 to FTP? So again, obviously, because this is a more, but if we then, let's say, bring it over more

00:12:08.200 to lactate as a tool for trying to figure out where this is. And again, this opens up a whole new world of

00:12:16.180 definitions, but to try to answer it simply, I would say that typically anaerobic threshold,

00:12:20.980 when you use lactate, and let's say that you use, for example, gold standards or maximum lactate

00:12:25.400 steady state, this will normally sit below the critical power of FTP, so at the lower power.

00:12:30.820 So for keeping track, we've got critical power, FTP, AT.

00:12:35.520 Yeah. Also, what's important here is that when we say anaerobic threshold is that when we talk about

00:12:39.740 the differences here, the differences here depend a little bit on what kind of athlete you are,

00:12:43.780 whether you are a high-power athlete or endurance athlete. So if you're a high-power athlete,

00:12:48.160 typically you'll see that there are larger differences between this, so the percentage

00:12:52.320 difference between this will be larger. Our endurance athletes, it will be smaller,

00:12:55.600 but you could basically call these minute differences. We are talking percent differences.

00:12:59.940 It's not like these are going to be 10% apart in terms of power or anything like this. We talk about

00:13:04.880 percentages from a couple of percents to maybe five, worst case, maybe 10.

00:13:10.400 Okay. And then lactate threshold, just adding more definitions.

00:13:12.960 Yes. So lactate threshold, I would say the lactate threshold has probably more of a result of

00:13:18.240 limitations in the measurement method, because we know that lactate is something you produce

00:13:23.940 all the time, even when you are sleeping. So the difference here is more where you're looking at

00:13:28.940 where you find a first infliction point on a lactate curve when you do as an increase in intensity.

00:13:34.840 So one thing that is maybe easy to get wrong here is that it looks like when you do increasing,

00:13:41.040 so if you start low enough power or low enough pace, it looks like, oh, there is no increase in

00:13:45.420 lactate production. The problem is the way we measure. We don't measure lactate in the muscle,

00:13:50.260 we measure in the bloodstream. And for it to basically be reflected on the instrument,

00:13:54.620 there needs to be a large enough lactate production in the muscles where you're not able necessarily to

00:13:59.320 metabolize the lactate immediately. And it starts to get transported out into the bloodstream,

00:14:04.580 and it starts to reflect also as an increase in lactate there. So the difference is also is that

00:14:09.560 depending on protocol, this will change. So that's also a little bit of the challenges when you also

00:14:14.920 start to mix something external with something internal as well.

00:14:19.280 And so if we did a lactate protocol, a speed or power escalation, right? I think we talked about this

00:14:26.960 in the last podcast, which was kind of the way we used to do it in swimming is you'd think we would

00:14:30.960 do 200-yard swims. So you do a 200-yard swim at a very modest pace. Come back, check the lactate,

00:14:37.120 rest, do it again, five seconds faster per 200. Go and do it again. Lactate, da-da-da-da-da. And you

00:14:43.280 get a plot. So pace on the x-axis, lactate on the y-axis. And the curve is very distinct. It is very,

00:14:50.440 very flat, and then it is not. And we would draw these tangent lines between the two, and then that

00:14:55.480 point was the lactate threshold. Where would that typically lie? Assuming we did it on a bicycle,

00:15:02.380 so it was really easy to do the power checks. First of all, how long a duration would you have

00:15:07.280 an athlete do this if you were doing it on an ergometer? Would you say we're going to do

00:15:12.000 three-minute efforts or something like that? Is that appropriate to generate the lactate performance

00:15:16.260 curve? It depends, again, how you're going to utilize it. If you are going to use it for actively,

00:15:21.640 let's say, controlling the intensity outdoors, for example, using a lactate meter,

00:15:25.520 then I would say it really doesn't matter. Even taking into account that there's some lag

00:15:29.700 between what's happening in the muscle and what's happening in the blood?

00:15:32.320 Exactly. Because the thing here is that here you're looking for more, not necessarily,

00:15:36.760 here also it's important just for me to say initially that when you find, for example,

00:15:40.720 a lactate concentration of LT1 and LT2, for example, or basically AT and LT, for example,

00:15:46.080 or AT, LT2, but we basically say lactate turn point two and lactate turn point one,

00:15:51.040 so the first one and the second one. And then when you do this,

00:15:54.380 it's also a misconception to think that your LT2 is always a constant value because this will be

00:16:00.380 influenced by many factors. We measure a concentration in the blood and the blood

00:16:03.940 is influenced by hydration and all things like this. So for example, if you have a change in

00:16:07.520 hematocrit, we won't come into that, but you can easily from the beginning of the session

00:16:11.520 towards the end of the session, see changes in hematocrit of more than 10%.

00:16:15.640 As you get dehydrated.

00:16:16.860 Yes. So this will already influence the lactate concentration, but to use it as a guiding

00:16:21.260 principle, it really doesn't matter how long your steps are. Basically, what you're looking

00:16:25.380 for is to find a concentration value, and then you go out into the field and you're trying to

00:16:29.500 figure out where this is. You basically said, okay, if you figured this out to be for an endurance

00:16:33.940 assay, for sake of simplicity, we say one millimole and two and a half millimoles. So one millimole at

00:16:38.960 the lactate threshold or LT1, and then two and a half of LT2 or anaerobic threshold, AT.

00:16:44.360 We usually saw them a little bit higher than that. Actually, I didn't really differentiate

00:16:48.260 between LT1 and LT2. I kind of looked at the single inflection point just using a two line.

00:16:54.620 Usually people were kind of in the three to four range was where that inflection point.

00:16:59.300 You're saying that corresponds to LT2?

00:17:01.140 Yes. So, and this also comes back to a little bit what kind of athlete you are.

00:17:05.200 Yeah. This was mostly swimming that we were doing.

00:17:07.740 I found swimmers, by the way, had the highest lactate capacity.

00:17:11.400 Basically for how much lactate-

00:17:12.520 How much lactate they would produce and tolerate.

00:17:14.540 And tolerate, yeah.

00:17:15.200 Yeah. Any idea? I mean, it could be just a small sample size.

00:17:18.240 Very often, it depends on what kind of swim is. If you look at the 100 meter swimmer, for example.

00:17:22.300 So call it two to 400 breaststroke, butterfly, individual medley. I never saw higher numbers

00:17:28.120 of lactate in myself or other swimmers than in those. I always assumed it was two things. It was

00:17:33.820 individual medley, you're using every muscle in the body. It's not like

00:17:37.660 cycling or running. You're hemorrhaging lactate into the system. And then secondly, at that distance,

00:17:43.460 two to 400, I mean, you're really in the pain train of, you're clearly not able to do this

00:17:49.180 fully anaerobically. So you're sort of maximizing aerobic and then topping up anaerobic. But that was

00:17:54.200 sort of why I sort of assumed, I mean, literally I measured on several elite swimmers, lactate's over

00:17:59.320 20 millimole.

00:18:00.040 I think very often it helps to think that metabolism is basically the same. The body is

00:18:07.040 the same, no matter what kind of sport we do. And that's why sometimes it makes it more universally

00:18:11.680 easy to understand. If we think of what kind of sport we do as a function of intensity and duration.

00:18:17.400 So as long as you find a sport, so let's say that you compare a sport that has a certain

00:18:22.560 intensity and duration. So let's say, for example, if it's a 200 meter or 400 meter, you could say that

00:18:27.240 then you are typically in the range of, let's say for the fastest swimmers, typical, let's say two to

00:18:31.840 four minutes, for example. If you take a 1500 meter runner and you look also at the lactic concentration

00:18:37.060 of them, you will also find actually pretty much the same values, let's say among also tracking

00:18:41.440 field athletes that does 1500 meters or 800 meters as well. As the moment you start to go a little bit

00:18:47.200 longer, then you will actually start to see that this actually is not the case anymore.

00:18:51.220 And this is also where it helps to also distinguish between lactic concentration,

00:18:55.020 let's say the highest value you're looking for, and also lactic production. Because lactic

00:18:59.280 production, it's even more complicated. We can't really measure lactic production,

00:19:03.760 then we need to integrate for time. So you have to do, let's say a pre-measurement,

00:19:07.300 a post-measurement, and then also obviously here there are, again, many weaknesses to an approach

00:19:12.180 like this, but you have to derive it based, let's say more on calculations than really that you can

00:19:19.420 measure it directly. How high you can get in lactic concentration is also valuable because it obviously

00:19:24.120 tells you something about how much you are able to buffer in your body too. So again, what we see

00:19:29.700 coming back to the main question of how do you use lactate and then basically for the sake of where

00:19:34.780 you do a profile and testing, this is where I would say that for endurance athletes like,

00:19:39.920 for example, marathoners, triathletes, and others, which has a very, very long duration,

00:19:44.580 and they don't have very much top speed. Here it's actually not uncommon even to see where the

00:19:49.240 second infliction points even validated by maximum lactose steady state that actually can

00:19:52.980 even be below two millimoles. Let's just translate that into English for people. One of the challenges

00:19:59.260 of talking with you is I enjoy it so much, and I sometimes forget that there are other people

00:20:03.260 listening who might not care nearly as much as you and I do about some of this minutia, but

00:20:08.160 it's just, it's, I can't get enough of this stuff. So let's make sure people understand what you just

00:20:13.720 said. You just said that when you look at the most elite endurance athletes, these people have

00:20:19.860 either some combination of such low lactate production and or such high lactate clearance that

00:20:27.820 frankly, the need to buffer it becomes secondary. They're producing such a high degree of power.

00:20:33.840 They're generating so much work with so much aerobic efficiency that their steady state lactate

00:20:39.680 doesn't even need to hit two millimole, which of course you can sit at two millimole all day long

00:20:44.380 and not notice it from an acid base standpoint. Is that a fair assessment of what you said or

00:20:49.000 implication of what you said maybe? Yeah. Yeah. Yeah. Because I think for those that are a little bit

00:20:53.420 more into the details, I think also again here is important to come back to that it's a lactate

00:20:59.000 concentration. And one thing that we also do know in elite athletes is that elite athletes has a

00:21:04.100 higher blood volume, larger plasma volume and other things as well. So this means basically that the

00:21:08.200 absolute lactate could still be higher. Exactly. So lactate production can be higher,

00:21:11.880 but basically it shows up as a smaller concentration in the body. But what we distinguish between is

00:21:17.160 basically where comes the point where you are continuing at the same pace or power, but now lactate

00:21:22.740 doesn't stay steady anymore. So if you measure after same pace, same power, but after five minutes,

00:21:27.640 after 10 minutes and so on, basically then you'll start to see that lactate still continues to rise.

00:21:32.420 It doesn't stay stable anymore. This I think is probably one of the few concepts of where you

00:21:37.560 can say that, okay, if you test it this way, it becomes very easy to translate. Let's say the

00:21:42.580 communication between people, because at the moment you start to talk about anaerobic threshold,

00:21:46.500 this is more like, okay, how do you define it? Is this 20-minute power? Is this 60-minute power?

00:21:50.500 Or what kind of duration, what kind of protocol, and all these kinds of things.

00:21:54.440 But that's one. And the second point also is that what happens, because obviously when you are an

00:22:00.360 endurance athlete, like a triathlete or a marathoner, for example, is that you need to have

00:22:04.580 a sustainable energy supply for the duration of the event. And obviously the faster you can go of

00:22:10.280 your potential for that duration, the more beneficial it is. So that means also that your utilization,

00:22:16.040 maybe come back to this a little bit, because VO2max is one of the terms we're going to define.

00:22:19.320 So utilization normally, just to take that term then, or let's say partially, let's define

00:22:24.960 utilization. Utilization is basically, normally you can do it as whatever you want. You can do

00:22:29.840 it as a race-based, oxygen consumption at race-based versus your VO2max. But very often in scientific

00:22:34.520 literature, we define this as, for example, your oxygen consumption at maximum active steady state

00:22:40.520 versus VO2max. So for example, if it was, then it could be that your maximum active steady state,

00:22:46.520 you see the oxygen consumption there is, for example, 80% off your VO2max.

00:22:51.880 Is that what you would typically see in an elite athlete?

00:22:54.360 That's low.

00:22:55.240 That's low.

00:22:55.940 That's low. For a marathoner, you would go even higher. So for a marathoner,

00:22:59.240 like the elite marathoners, you're talking about 94, 95% of the VO2max. For Christian,

00:23:05.040 Gusto, when they do Ironmans, you're also pretty much around there.

00:23:07.920 Let me just make sure we get people to understand that, because that is, as you know, now that I

00:23:12.520 have a VO2 master, I'm testing my VO2max all the time. And I'm also testing my VO2 at submax

00:23:18.900 levels all the time as well, and looking at how that corresponds to lactate levels.

00:23:22.560 Yes.

00:23:23.000 So why don't you define VO2max, and then we're going to come back to this point,

00:23:26.300 because I want to make sure people can internalize and maybe even one day experience what you just

00:23:31.780 described.

00:23:32.880 So VO2max is normally defined as the maximum oxygen your body is able to consume over a

00:23:40.960 minute, a full minute. But this is also something that is debated. It's less of, let's say,

00:23:48.420 protocol influence. You can still go out and do, for example, efforts out in the field,

00:23:54.660 and you will be able to produce higher numbers than what you would be able to do on a standard

00:23:59.100 graded exercise protocol.

00:24:00.760 That's been the case for me. You know, I've shared my data with you. So I like to do it on

00:24:05.940 my bike on a hill. Take a hill that's going to be not too steep, 6% grade. You're in the saddle the

00:24:12.320 whole time, and you're in the big chain ring, and it is go for broke and make sure you're dead at the

00:24:18.340 top. Four minutes of climbing, and that produces a VO2max. There's going to be a level, usually near the

00:24:24.200 very top of the hill in the last minute where I reach the maximum volume of oxygen that I'm consuming.

00:24:32.340 Now, indoors, I, for some reason, hate doing it on a stationary bike. So instead, I do it on a

00:24:37.960 Stairmaster. So now I'm just sprinting upstairs doing the same thing. I get a comparable number,

00:24:43.580 but it's a little bit lower. Now, what I haven't done yet, and I was going to try to do it before we

00:24:47.580 met today, but I ran out of time. I was traveling last weekend. I have a prediction, which is, I think

00:24:54.200 I would have a higher VO2max running on a treadmill, even though I'm very inefficient running, because

00:24:59.460 my heart rate is always higher when I run than when I cycle. Do you think that would be the case?

00:25:05.240 Normally, I would say that should be the case, but it is also where there can also be other

00:25:13.080 factors also involved in it. One of the places where this is something that have quite passed me

00:25:20.300 is that, for example, Formula One drivers, they sit in for almost, let's say, one and a half hour

00:25:25.100 with extremely high heart rates. But having looked at that telemetry very closely, a lot of that is

00:25:32.120 driven by the G-forces. I mean, the first time I looked at an F1 driver's telemetry, I was like,

00:25:37.240 there's a mistake. These are just errors. These aren't real numbers, because I had never seen

00:25:42.240 such rapid changes from low to high to high to low. And then I realized they were all corresponding to

00:25:50.080 either very, very strong break points or very, very fast corners. Obviously, the only thing that's

00:25:56.500 common to that is something about 5G on the body is dramatically doing that. Now, it would be very

00:26:02.840 interesting if we could do it, I don't know how you would do it, would be to measure VO2 and see if it

00:26:08.600 has a commensurate change. What would you hypothesize?

00:26:11.940 So this is, I think also when you are doing these, so it could also be like jet fighters as well.

00:26:18.080 I would imagine that's even more pronounced for them. And we could actually measure that because

00:26:21.780 they're already in a mask.

00:26:22.600 Yes, exactly. So you could say that in dogfighting, for sure, then I think you will actually be pretty

00:26:27.640 much maybe same as Formula One. But I think that if you're just pushing really high G-forces

00:26:33.560 in one direction, then it would maybe not be as high, even if it's a prolonged one.

00:26:38.900 The reason for that is because I think that the G-forces, obviously, you're trying to counteract

00:26:42.520 that. So you're mobilizing every bit of muscles that are in your body in order to stay exactly

00:26:47.600 where you need to stay.

00:26:48.440 But there's a slight difference, which maybe you're about to get to, so I apologize if you are.

00:26:53.180 There's got to be also a part of that heart rate that is coming from the compromised Venus return

00:26:57.640 to the heart due to either the Valsalva or the actual impeding of the inferior vena cava.

00:27:05.260 I would think at a high enough G-force, part of that high heart rate is stroke volume has

00:27:10.940 gone way down because preload is way down. Again, sorry, I should make sure everyone understands

00:27:16.260 what we're saying. When you fill the heart less with blood, you don't stretch the muscles

00:27:21.720 out. That's called preload. So the heart needs to be preloaded with lots of blood volume

00:27:26.140 so it can get a good squeeze. And anything that prevents that, either dehydration or literally

00:27:31.220 forces that are preventing the blood returning to the heart, could make that high heart rate

00:27:36.220 really a product of tiny, tiny little ineffective beats.

00:27:39.580 Yes. That's also, we've done some studies on that.

00:27:43.280 In high G-force athletes?

00:27:44.720 No, actually in similar, because it's quite interesting to see that what happens also with

00:27:49.200 the muscles. So when you do work, you produce a certain amount of power. Obviously, there are forces

00:27:54.120 involved and there is velocity involved. These forces require a certain amount of recruitment

00:27:59.060 of muscle fibers. And these muscle fibers, at a certain point, will actually start to cause

00:28:03.440 vasoconstrictions. So they actually start squeezing off the blood supply. They normally act also as a

00:28:08.520 pump. They actually help pumping blood around in the body, promoting preloading. But the interesting

00:28:13.660 thing is that we do see that where you also get in the area of anaerobic threshold is actually

00:28:22.360 where you are starting to come to the point where you're squeezing off the blood flow as well in

00:28:25.860 the muscles. And this happens around ballpark-ish. Obviously, there is a fairly large ranger,

00:28:31.280 but ballpark-ish around 30% of your 1RM.

00:28:34.360 You should never experience that on a bike, right? Because in theory, it's hard to imagine you could

00:28:41.420 ever come close to 30% of a 1RM force on a pedal stroke. Could you?

00:28:47.280 That's true.

00:28:47.720 Maybe there's an extreme moment you're at a 16% grade or something like that, I guess. It's possible,

00:28:53.900 huh?

00:28:54.260 Yeah, for sustained efforts. For short efforts, you can get very, very close. But for sustained ones,

00:28:59.260 yes, I would agree. But coming back to running versus the bike, which was, let's say, the place

00:29:05.060 where it started. Normally, I would say that, yes, in running, you would at least a sub-maximal

00:29:10.980 efforts have a higher view to max. The only thing is that we see that in people that are, let's say,

00:29:16.360 somewhat balanced trained, so they spend some time on the bike, they spend some time on running,

00:29:20.520 then normally there are more muscles involved for a longer duration in cycling than it is in running.

00:29:25.840 So when you test Christian or Gustav, if you were to have them fresh on two separate days,

00:29:31.560 and on one day you put them on the bike, and on one day you put them on the treadmill,

00:29:37.100 and you have them do a VO2 max test, what would be the approximate difference you would see in them?

00:29:41.100 Actually, when I started working with them, then it was close to a significant difference between

00:29:46.460 cycling and running, and even more in swimming.

00:29:49.340 Higher in swimming?

00:29:50.280 No. Lowest in swimming, actually. And that comes also back to one other thing.

00:29:54.120 One thing is also when we talk about preloading and how this also affects it, 0.88

00:29:58.140 I actually were stupid enough to teach Christian and Gustav this a long time ago,

00:30:02.240 and that is if they wanted to have a very high view to max. Obviously, if you're trained, 0.99

00:30:06.460 you will see this immediately on the breathing patterns. But what you can do to create an

00:30:10.960 artificially high view to max is that you basically, when you come pretty close to your

00:30:14.700 all-out effort, if you try to restrict your breathing, let's say for a short, short time

00:30:19.060 there, you will create a depth, and this depth will actually boost the numbers even higher.

00:30:23.360 This is the VO2 max hack for everybody watching this. How do I boost my, because my hack for boosting

00:30:29.740 VO2 max is weight loss. Just figure out a way to lose five pounds in the week leading up to the test,

00:30:35.920 then you're, goes, I've been obviously knowing how to train for it. I mean, if you train in those

00:30:40.440 intervals. So you're saying if you restrict breathing, as you get close to that failure point,

00:30:45.700 each breath you take becomes that much more of an oxygen explosion.

00:30:49.940 And even there, you can even train yourself also to the point where you're not talking about

00:30:54.100 marginal differences in your VO2. You can spike the numbers up to extreme numbers,

00:30:59.380 if you practice this a little bit. That aside, I think it's important here to remember that

00:31:03.940 that's not equivalent to getting fitter. It's just a way of basically manipulating something

00:31:08.320 in the body, and you're cheating a protocol. And it's not your, let's say, VO2 max.

00:31:12.900 But for those listening to this who have a bet with their friends about who's going to have

00:31:16.280 the highest VO2 max, this becomes their little trick.

00:31:18.820 Yes, then definitely you can do this, and you will get some really nice numbers.

00:31:22.840 So that's true.

00:31:23.460 How much of a boost would you get? Like, let's say you're a person who just

00:31:26.060 doing the protocol normally is 60. With no other change other than this breathing trick,

00:31:30.620 what would you take that to? One thing I want to add here just before answering that,

00:31:34.920 VO2 max for me, in the same way as power, you can look at it in relative terms and absolute terms

00:31:39.840 as well. So obviously losing weight. Yeah. So let's say it's five liters based on the weight that comes

00:31:45.080 out to 60. Yeah. I would say that in this case here, it comes very much down to technique,

00:31:50.320 but if you practice this a little bit that you can boost it above 70.

00:31:53.040 What? Not a problem.

00:31:56.960 Wow. I thought you were going to say 64, 65.

00:32:00.540 No, no. I've seen some crazy numbers. I don't even have to go into the raw dollar. You just

00:32:04.540 see it on the screen and say, okay. But wait a minute though. Doesn't that sort of fly in the

00:32:08.520 face of our definition of VO2 max, which is it's the highest oxygen consumption you sustain for one

00:32:14.880 minute? Because I know when I do my test, I think I'm always checking for 60 second average,

00:32:21.000 and I'm not ever just looking at the peak. It's very noisy data. So in this breathing technique,

00:32:26.880 don't you just get a really big spike that is otherwise noise in the one minute best effort?

00:32:32.720 No. So it would be a little bit more prolonged than that if you create an oxygen depth in your

00:32:37.360 body. But one thing that I think also here is important too, because this is also one of the

00:32:42.020 things that is a little bit challenging when you read a lot of research too, is that very often we

00:32:46.640 take that research for granted. Well, it's good. There is a lot of things around it that we don't

00:32:51.880 have access to or we don't know. And also one thing that also is sometimes a challenge is for

00:32:57.320 people to understand also whether the technology they are using, one, is it good for what you really

00:33:02.520 are doing here? So you can get a lot of machines that can measure some kind of oxygen uptake or

00:33:07.420 ratios and other things, but it doesn't mean that it necessarily is a good instrument for what you

00:33:11.760 want to do. And the second part of it also is that we have to understand where are we measuring?

00:33:16.940 Obviously in this case, when we do oxygen measurement, we measure it on the exhaust and

00:33:21.420 not in the muscles. And that's also an important difference here, because if we go to the point,

00:33:27.560 and I think we're going to run out of time if we continue on this path here, but if basically

00:33:32.120 you look at cellular or basically cell respiration, the numbers you will see much higher. I remember

00:33:38.180 this from our last discussion. We're not limited at the cell.

00:33:42.260 No. Again, here, I think it comes down to just staying true to some principles that can guide us.

00:33:47.860 The good thing with VU2 max testing is that there, the generally accepted standard for doing this

00:33:53.400 is a graded exercise test. And you go through that and it will produce fairly comparable numbers

00:33:59.980 for most people to relate to. There will be some differences.

00:34:03.540 What are some of the mistakes you want people to look out for? Because again,

00:34:06.300 people listening to this podcast are no stranger to VU2 max. We talk about it as if you were going

00:34:10.820 to sort of say, I need to know 10 parameters of my body for optimal health. VU2 max is on that list

00:34:17.080 every day of the week. So even if you are listening to this and you don't care about triathlons or

00:34:23.220 cycling or swimming, but you just have the desire to live as long as you can. And as well as you can,

00:34:28.580 you have to know your VU2 max and you need to optimize it and make it frankly, as high as you can

00:34:32.920 with whatever time you're willing to devote to training. That said, most people aren't going

00:34:36.840 to go out and buy a VU2 master and they're not going to geek out and do this stuff on their own.

00:34:40.440 They're going to rely on going to a place where the VU2 max is measured in a stationary fashion,

00:34:44.460 either on a treadmill or on a bike. What are some things that those folks need to be aware of when

00:34:48.420 they go to a center to test this as far as mistakes that people make? So for example,

00:34:52.780 I've seen many people go and do a test and I talked to them after and I go,

00:34:55.920 how exhausted were you at the end of the test? And they say, oh, it was okay. I could have kept

00:35:00.460 going and I'm realizing, okay, you didn't do a maximal test. Other mistakes I've seen are people

00:35:04.680 who don't have a long enough warmup time. The total test took five minutes and I'm like, well,

00:35:09.100 you didn't come close to sufficiently warming up. So are there any other things that you would just

00:35:12.840 have people have in the back of their mind when they call a place up to say, hey, I want to come

00:35:16.140 and do a VU2 max test. But before I give you my hundred bucks, I want to understand the protocol.

00:35:21.140 Of course, the question becomes also to what extent you're able to influence this as well,

00:35:24.680 because there are some practices that they want to dominate in a very rigorous way. They even

00:35:28.760 almost prescribe what you should do the days leading before and another where you just come

00:35:32.840 in and they just put you through a test and then you get what you get.

00:35:36.560 So what should folks look for? If you were sending somebody to a public place where they could go and

00:35:41.800 do this, they're not coming into your lab where you're going to administer the test.

00:35:45.160 What would you want them to be looking for?

00:35:47.560 What I would normally do is the one I would try to create as a standardized protocol from time to time.

00:35:54.120 And that means not only the testing itself, but also what happens a little bit before

00:35:57.660 without it, it becomes too invasive into people's lives, like standardizing what you do the days

00:36:02.300 before and all these kinds of things. I think this is impractical for most people. If this is a very

00:36:07.040 important metric for you, like one of the key metrics you use in order to understand how have

00:36:12.120 the previous period influenced you and you are looking for marginal changes, then it becomes very

00:36:17.280 important also how you standardize the days before. But I would say for a normal person, I would say that

00:36:22.280 I would keep a standardized protocol. So I would come in there, let's say at least trying to eat

00:36:27.300 pretty much the same every time I go there. And that means making sure you are well fueled as if

00:36:32.800 you were going to do important training session. That means normally more carbohydrates, make sure

00:36:38.720 you're well hydrated. And obviously also that you are to the extent you can influence it or you can,

00:36:44.020 you should be able to influence it, but make sure that you also had a proper sleep so you don't come in

00:36:47.980 like completely tired and things like this. Reasonably well rested, not trained the day

00:36:53.040 before, would you say, or like training the day before? This is where I would say it's more

00:36:57.360 important to use your experience and to have enough understanding of what makes you set up for a

00:37:02.940 good test the day after. So for some people, exercise is irregular and there are more time in

00:37:07.900 between. Obviously, if you're going to do a little bit harder exercise the day before or anything like

00:37:11.400 this, you'll come into the session and you even feel sore in the muscles. This is not ideal

00:37:15.420 necessarily for the ability to mobilize. For the athletes that are well trained, then it's

00:37:20.780 completely different. Then it's more like, okay, what would I do the day before a competition or

00:37:25.040 anything like this? It doesn't have to be an elite level, but it's more like you think a little bit

00:37:28.380 like, what do I want to do in order to make sure that I've gone through the different stuff and I

00:37:32.260 feel ready for the day after. So on the day of the test itself, one is important, timing of the day.

00:37:37.760 This is important to try to keep somewhat consistent, I would say, because of the circadian rhythm.

00:37:42.780 We know that, for example, your temperature levels in your body, this fluctuates. This is maybe one of

00:37:47.380 the easiest observable differences in a human, basically, as an indirect measure of the circadian

00:37:52.760 rhythm, where we basically see that you have a minimum point that happens typically in the morning

00:37:56.880 just before you wake up, and then you have a maximum point typically in the afternoon.

00:38:01.000 And there is, of course, plenty of research suggesting that aerobic exercises are best done

00:38:07.800 normally a little bit in the afternoon when you are typically also around the highest

00:38:12.060 core temperature or natural core temperature. The minimum point is easy to measure. So for

00:38:16.740 people that has the core, like now it's become very popular, like in the Tour de France and all

00:38:20.460 the places, triathlons, you see these people, they run around with this little temperature device on

00:38:24.460 their body. And it's very easy to observe if you use it 24-7. This is very easy to observe,

00:38:29.680 basically, when your minimum point is. The maximum point is influence, let's say it can be masked

00:38:34.640 by noise and other activities and other things you do, so it's a little bit harder to see where it is.

00:38:38.420 But let's say you put them 12 hours apart as a guiding tool. But now I'm already getting to the

00:38:43.100 point where it gets a little bit, let's say, more detail, but I would just say, keep the timing of

00:38:47.600 the day the same. So it really doesn't matter if you do it in afternoon or you do it in the morning,

00:38:51.580 as long as you keep it consistent. Then from there, I would do a proper warmup. That's important.

00:38:57.280 How long do you think that should be for a person who's like modestly trained? Let's assume we're

00:39:00.280 not doing this on someone completely untrained, but someone who's recreationally trained versus

00:39:04.220 relatively elite. So again, here I would do a warmup much in the same way I would do a training

00:39:09.540 session. So I would normally stick to a standardized warmup protocol where you start out maybe, for

00:39:14.020 example, six minutes, very easy. This can even be walking. Then I would do, or basically pedaling,

00:39:18.880 soft pedaling. Then I would do six minutes with a little bit more effort, but it typically should

00:39:23.140 be more something that you would be comfortable doing as your longest sessions that you do. So if you

00:39:28.480 are out and you do, I say the longest sessions you do is an hour. Obviously, if they're all out,

00:39:33.060 then it would be a little bit too hard, but then you should more aim for something I could do for

00:39:36.080 like, this would be an easy session for me, but I could do it for some while. You could,

00:39:39.740 for some people say, this is your marathon, let's say pace or where you do more of a longer rides.

00:39:45.260 And then I would do a short effort around three minutes that sits somewhere where at this time,

00:39:50.760 I wouldn't look too much on my power meter or my heart rate or anything, but more by feeling

00:39:55.380 where you feel, okay, this is where my normally where my treasure would be. And then I would probably

00:40:00.020 also do, let's say a couple of short, short efforts that are, let's say progressively towards

00:40:06.180 view to max. So let's say one, two, three, maybe two to three efforts that are somewhere between

00:40:11.280 10, 15 seconds with equal rest in between. And the reason for this is that it's easy to think,

00:40:16.360 ah, but don't you exhaust yourself now before you do the view to max then? Well, no, you don't.

00:40:21.120 The little fatigue or let's say the little fatigue that you possibly would induce with this

00:40:25.220 is easily offset by other factors that will much more contribute to your view to max when you start

00:40:31.280 doing this. So that means also that after I've done this one, I would just again, go back to very

00:40:35.540 short, easy efforts. Let's say you get a total warmup of 20 minutes, something like this in

00:40:39.840 between there, take a toilet break or anything like this, but don't keep this rest too long,

00:40:43.560 because obviously now you're going to cool down again, if you don't go to your view to max test now.

00:40:47.980 So that means also like have a small sip of water, something like this, if you want to do that.

00:40:52.220 Also, before you start your warmup, it can be good things. If you have something

00:40:54.880 carbohydrate based, you consume a little bit of that. Some people would say, ah,

00:40:59.940 I don't like this because it could influence. Yeah. It can influence the RQ.

00:41:03.280 Yeah. But then again, this is something I would say that a sub maximal, yes, maximal,

00:41:07.560 it won't do that very much because the closer you get to your metabolic steady state or maximal

00:41:12.860 metabolic steady state and then above basically things start to homogenize. Yeah.

00:41:16.960 We tend to do these separately now. So we don't use the day of the VO2 max test to do the fuel

00:41:24.160 partitioning test. So we want to measure fat oxidation on a totally different day. We'll

00:41:28.680 usually separate by a day even. So you'll do your VO2 max test on Monday and we'll do it this kind

00:41:33.460 of way where we can agree completely like you should have all the carbs in the world. We actually

00:41:37.380 like to standardize the fat oxidation test to a fasting test. That way it's just always the same.

00:41:42.460 You're going to do it in the morning, you're fasted. It's a sub maximal effort. We're not

00:41:46.040 pushing you to maximum because we're just pushing to see what your maximum fat oxidation is.

00:41:49.680 This is a good way to open up, let's say a little bit the door to how we, because again,

00:41:54.860 we are at the edge cases where basically we have to do the research ourselves.

00:41:57.980 One thing we have to consider is that when we do gas exchange measurements, for example,

00:42:02.060 to understand a little bit substrate metabolism.

00:42:04.600 What's your preferred in-lab test? Do you use Parvo?

00:42:07.580 No, we use a device which is still on all device. There are two devices I like to use.

00:42:13.800 One is the Jager Oxycon Pro with a mixing chamber and this is important. The Parvo has a mixing

00:42:19.240 chamber as well, but we have the Oxycon Jager Pro. When I've been testing in the US, for example,

00:42:23.720 on elite athletes, Olympic athletes, then we have used the Parvo simply because that's the one that

00:42:28.020 has been here. Europe is more used to Jager. We don't use the, maybe I shouldn't say that,

00:42:33.420 we don't use the Vintu CPX, which is the successor of Jager Oxycon Pro. This is one of the devices.

00:42:38.660 We use also the AEI Moxes, but this is more of our, call it for specially interested people.

00:42:45.440 We use also the Vue to Master actually quite significantly and more and more of the testing

00:42:49.440 actually have gone over to the Vue to Master. We have been fortunate enough to work with them

00:42:54.120 for some while. So we have the CO2 version as well, but this is where it has been in alpha mode and it

00:42:59.500 requires also completely different skills than what you would normally require. But the reason also

00:43:05.180 why we do that is again, also for one of the reasons you experienced is that you see that

00:43:09.160 your Vue to Max is maybe a little bit different when you are out and cycling and do the efforts

00:43:12.800 and it's not inside the laboratory. Because also we have to remember that Vue to Max is on the one

00:43:18.040 side, we say it's the highest oxygen consumption normally as they normalize over one minute.

00:43:22.120 That is a wrestle how much muscles that are involved in the work there. So obviously modality

00:43:26.560 will influence this quite a bit as well.

00:43:29.220 Which is why I was surprised when you said the guys were lower in swimming than cycling and running,

00:43:33.560 even though I realized that their efficiency in swimming is relative to the world's best,

00:43:39.420 less than it is in cycling and running, where they would be closer to the world's elite.

00:43:43.800 But I would have just thought more muscles would have meant higher O2 consumption.

00:43:47.440 Let me shed some light on it because that was basically when I started working with Christian

00:43:50.480 and Gustav. Where they are today is that basically there are virtually no difference between

00:43:54.480 swimming, cycling, and running.

00:43:55.920 Same Vue to Max.

00:43:57.080 More or less on the three disciplines, but it has again to do exactly like you said,

00:44:01.020 you can manipulate your Vue to Max by that you are targeting, let's say, typically shorter

00:44:05.900 duration efforts. And so you start recruiting more, learn to recruit a little bit more,

00:44:10.280 and also train, let's say, fibers that you are not normally used to use so much.

00:44:14.720 And this is also something that is highly plastic, much more plastic than we ever have thought

00:44:19.120 it was before. To the point where we have leading researchers in Norway that have solely

00:44:24.760 focused on Vue to Max. And let's say they went into schools where they've done this for decades,

00:44:28.820 but where we even sat down and we looked at the data and thought something has to be wrong.

00:44:33.440 And that comes actually from the period when we switched from Tokyo Olympics and then on to

00:44:37.140 Ironman. And basically we saw a decline in Vue to Max. For me, not unexpected, but more,

00:44:44.080 there hasn't been data on this. So it was actually quite nice to measure.

00:44:46.760 What is the decline? How much?

00:44:48.040 So the decline in absolute, or let's say in relative Vue to Max, and so it makes more sense

00:44:52.920 for people, is that Christian and Gustav, typically the highest numbers we have had on them

00:44:57.000 are very, very high. Highest numbers ever measured in history.

00:45:00.880 I mean, 90?

00:45:02.260 Well, Christian, we've had measurements with him actually invalidated, but on him, we have

00:45:08.040 exceeded in absolute terms 7.7, 7.8 liters per minute in oxygen uptake.

00:45:15.120 At what, 75 kilos?

00:45:16.480 No. So at that time he was around 80 kilograms. So just let's say around 100 milliliter per minute

00:45:22.220 per kilogram in oxygen uptake. So Gustav is a little bit smaller, so he has a little bit

00:45:28.380 lower oxygen uptake than what Christian have, but he has the highest he clocked in around,

00:45:33.600 let's say 94 milliliter per minute per kilogram, something like this at the highest. But I would

00:45:37.560 say that this is not beneficial for what they are doing.

00:45:40.480 Of course. I think anybody who's done this type of testing and done this kind of work,

00:45:44.780 they'll have an appreciation for what those numbers mean, which is to say, I'm trying to come up with

00:45:48.960 an analogy for another sport to explain what that would be. That would sort of be like, I don't know

00:45:54.320 how to compare it to non-endurance sports. It's like saying in an NBA basketball game, occasionally

00:45:59.980 they score a hundred points. It doesn't compute. It doesn't really compute that they could utilize

00:46:06.300 that much oxygen.

00:46:07.060 Yeah. So the interesting thing there and the good thing about it is that this was something we

00:46:11.100 reproduced over three months.

00:46:14.000 But the training at that time was presumably geared specifically towards VO2 maximization.

00:46:19.140 It was short distance. It was interval based.

00:46:22.080 Well, actually, yes, it was. But this is also where one place where we resonate a lot. VO2 max

00:46:28.520 is the single best metric we probably have for anything that is related to human health and performance.

00:46:36.060 But it is also where we understand that there's a little bit more nuanced than that. I'm not

00:46:40.180 talking about it as pure as a predictor for your race performance, because that's a different domain.

00:46:44.740 But where you could also say that you could still be healthier with a little bit lower

00:46:48.140 VO2 max. And now I'm not talking because about that you put yourself through some stress and

00:46:52.520 other things like this, so this becomes negative longer term. VO2 max is a one-dimensional

00:46:56.960 also unit. Again, we talk about something on a y-axis. Yes, we normalize it as a milliliter per

00:47:02.140 minute. So you could say that it has a second access to it, an x-axis or time access to it as

00:47:06.680 well. But we don't say how long- Over what duration. We don't get into capacity.

00:47:11.200 Exactly. And this means also that you can trade off some of your VO2 max. And I would say that

00:47:16.720 the very, very best predictor is capacity. Capacity is maybe the single best one for everything. But

00:47:22.340 the problem is that testing for capacity is such a brute force endeavor that it's not practical.

00:47:27.580 It is really not practical. That's why VO2 max is, I would say, is the single best predictor of

00:47:33.080 everything measured in a practical way. But for Christian, what was really nice when we did this,

00:47:39.140 this is not only actually where we measured. And I can say that we are at the edge where there are

00:47:44.820 things that we don't understand. There are still things we are researching. And like you said,

00:47:49.060 introductory, I'm in the applied world. And in the applied world means basically much more

00:47:53.980 experimentation and understanding what happens here with some individuals. My sample size doesn't

00:47:59.680 come from the population. It comes from the sheer amount of data that I gather on these apps. So

00:48:05.400 it's not like a single view to max test. It is so many view to max tests that there's probably no

00:48:10.060 other human in history where there have been true this testing protocols over time, where we can

00:48:15.580 correlate it to also a range of other metrics as well, both internally, but also external metrics.

00:48:21.980 And that means that, for example, here, to put this into context, the good thing is that when we do

00:48:26.380 testing, I normally not only have Christian into the lab, I have also minimum of, let's say,

00:48:30.620 one or two athletes as well into the lab. And our protocols are quite extensive. So this is also

00:48:35.700 one thing that we already talked about, that protocols can have an influence on testing. And what you do

00:48:40.840 before the protocol will have an influence on your V2 max. That's exactly why we're discussing,

00:48:44.760 should you warm up or should you not warm up? Yes, you should, because it will normally give you,

00:48:48.920 let's say, a better result when you are warmed up. But that means also that the protocol will also

00:48:53.520 have, can also then have an influence on your V2 max measurements. And we normally do the V2 max

00:48:58.760 measurements at the end of our test, which is, you would say, ah, but that's normal. Yes, well,

00:49:03.480 we also do that simply because it also more normally simulates what they're also going to see

00:49:08.960 on racing. So normally what happens in racing is that you normally also will go closer to,

00:49:13.520 let's say, complete exhaustion when you get towards the end of the race. And this is why it's

00:49:17.260 interesting to test it there. But we don't only test it once, we even repeat it. So we do two

00:49:21.980 V2 max tests just 10 minutes apart. And even this year, we did even three tests, let's say,

00:49:27.140 with less or 10 minutes apart as well. Interesting thing here is that what we see is that you normally

00:49:31.900 don't get your highest V2 max on the first one, even though you're exhausted. We even see that even

00:49:35.680 on the second one. More interestingly to also to serve here is that then you would think that,

00:49:40.200 I say, intuitively you can say that, okay, you would maybe also then see a higher carbon

00:49:43.960 dioxide production as well. You don't. That's actually lowered. That has to, let's say,

00:49:48.520 call it some priming effects more in general terms. But also what we do see is that there

00:49:53.100 are some substrates as well that actually do influence on micronutrients also that actually

00:49:57.720 also can help boost your V2 max as well. Such as?

00:50:01.520 Long story short, beetroots obviously have been something that people have found quite interesting

00:50:06.520 over a long time.

00:50:07.720 Beetroots?

00:50:08.340 Beetroot, yeah. So they use beetroot concentrates. And the main thinking behind is that when you eat

00:50:13.820 nitrate, which beetroot is normally rich in, and the body converts this to nitric oxide,

00:50:19.640 this helps for vasodilation. So vasodilation, you can think of this almost like a plumbing

00:50:23.820 in our body. We already said that from our previous conversation that basically cellular respiration

00:50:28.300 is not the limitation to UVU to max. So there are other things that are limiting factor.

00:50:32.980 That means also that, for example, your cardiovascular system, meaning also actually your ability to

00:50:38.520 transport blood around in your body is going to be important. For example, one of the reasons

00:50:42.860 why when you then use some supplements that are highly enriched in nitrate, so you get

00:50:50.160 a nitric oxide boost, is that this was actually, it's almost like plumbing. You're opening up

00:50:55.220 the plumbing and it allows your blood to circulate faster throughout your body. This is hard to

00:50:59.680 reproduce in elite athletes. Like in amateur athletes, we normally see that this has a positive

00:51:03.400 effect.

00:51:03.800 How much of an effect? 5%? Something measurable?

00:51:07.240 Let's say something like that, and then take that with a grain of salt.

00:51:10.200 These are not people that are out there literally eating endless beetroots you could

00:51:13.360 buy as a supplement.

00:51:14.660 Yeah, concentrate.

00:51:15.960 Now, would it be anything that increases nitric oxide?

00:51:18.320 So that's the thing, because one thing that we don't see this, we don't see these effects in

00:51:22.900 elites, for example. And this is obviously one of the benefits when we do more longitudinal

00:51:27.020 studies with such, let's say, granularity and in-depth measurements we do.

00:51:31.300 And one of the things we see there is that when they use beetroot concentrate, nitrates

00:51:36.480 are considered an ergogenic aid. When you do all these measurements, you leave out the

00:51:41.380 guesswork and we see, well, over all the testing we do, we have not seen a real effect out of

00:51:47.200 this, at least not on VU2max, maybe other places, but it doesn't seem to have any effect

00:51:50.960 on VU2max.

00:51:51.480 Because they're already optimized in that regard.

00:51:53.420 Probably, or maybe there are other limiting factors instead. We were approached by a company

00:51:58.040 that is called Plasmaid, and they actually focus more on the other part of it. Let's

00:52:02.460 say the catalyst that actually help, because you have to convert nitrate into nitric oxide.

00:52:07.400 And there's a cost to this. So what they did instead is that they made from pine bark extracts

00:52:12.480 instead, they made adaptogen, or they extract an adaptogen that helps catalyze this process.

00:52:17.800 And the interesting thing with this adaptogen is that, again, now we are a place we can just

00:52:22.760 look at the observations and repeated observations over time, and we can't necessarily explain

00:52:27.820 100% yet, so it becomes more speculations. But from those, what we can only speculate

00:52:32.960 or hypothesize is that what happens here when they use this Plasmaid is that, I remember

00:52:38.880 when we got this first presented, I was thinking like, okay, we've tested nitric oxide or nitrate.

00:52:44.200 It really didn't give any help, so why should this really give a help, or basically make a

00:52:48.140 difference? The interesting thing is that Christian and Gustav were quite positive, so they said,

00:52:52.820 okay, fine, we're going to use this. We were just going to test it. We had a bunch of them

00:52:56.300 laying around there, and they said, okay, let's try it. I said, okay, fine. Gustav was the first

00:53:00.440 one to say immediately that he felt it did something with his respiration. Also, it's a little bit the

00:53:06.140 placebo as well, like, okay, how much of this is placebo? How much of this is real? You get something

00:53:10.320 new and say, oh, looking for something in your body. And Christian also, but he observed a little bit

00:53:15.840 like a different effect to him. He felt more like he normally felt quicker ready for the next effort

00:53:22.040 using this. But I would say that these record measurements that we did came in the tail of

00:53:27.960 when we started using this. And that was a little bit of shock because what we also saw at the same

00:53:32.100 time was that the efficiency, so the biochemical efficiency in the body also went down. The

00:53:36.840 interesting thing was the RQ was heavily, on the CO2 side, was slightly reduced. The VO2 were heavily

00:53:43.420 increased. Tell folks what RQ is and what the reduction would imply. At the end, normally you

00:53:48.580 would say that for a VO2 max test to be valid, one of the criteria is that you should meet 1.1

00:53:53.260 in RQ. Meaning you have 10% more production of CO2 than consumption of oxygen. Yes. In elite

00:54:00.780 athletes, this is sometimes hard. In the best trained elite athletes, bringing them to 1.1 during

00:54:06.340 the test can actually be quite challenging. So if you have a short warmup protocol, they are not well

00:54:10.560 warmed up and all these kind of things, then you would easily be able to exceed this. But if you

00:54:14.000 do it more in a simulation like we did, or you have longer protocols before, you normally see that

00:54:18.100 this is a little bit suppressed. So let's say that Christian normally on a VO2 max test, or let's say

00:54:22.500 at the end of the VO2 max effort, he would basically click in at, let's say, 1.05. Then after we started

00:54:28.180 using these supplements, we could see suddenly not 1.03 or 1.02. We basically dropped it by 10 points.

00:54:35.280 So basically to 0.95 instead. Preserving the VO2. No. CO2 was basically not as much changed.

00:54:42.880 So the VO2 was increasing. Yes. So VO2 went up quite a bit. The good thing here is that now I

00:54:48.220 have two other athletes obviously in the lab at the same time. So I have one athlete that comes

00:54:51.760 in before Christian. So I already do dynamic calibration of the machines before. We also

00:54:56.060 have a lung simulator just because we are working with edge cases. So we need to independently validate

00:55:00.480 the machines as well. So this means basically we have a large gas tank with a reference gas beyond

00:55:05.220 what is used for just the calibration. But we have a separate system where basically we have a lung

00:55:09.440 simulator where we feed in, for example, six liters of oxygen, six liters of CO2, and we know this is

00:55:14.240 the exact amount in the bottle, gravimetrically calibrated. And then if we don't get this out of

00:55:18.640 the machine, then there's something wrong with the machine independently of the calibration.

00:55:22.340 But the good thing here is that despite this, first athlete in, numbers are where you expect them to be

00:55:27.440 based on the training we did before. Christian comes in, you start to see some numbers that

00:55:31.260 are crazy and you start, did I do something wrong here in the calibration process or anything like

00:55:35.300 this? But then also when you validate it afterwards as well, you basically know this is, the numbers

00:55:40.060 are good. And then this we did then over, this was in December, January, and February over nine tests,

00:55:46.000 more or less. We saw a little bit over time that actually VU2Max started to come down a little bit

00:55:50.760 because what this indicated also was that his biochemical efficiency was not optimal either. What that

00:55:57.180 means again is that if you look at basically his oxygen consumption versus his power, so the power,

00:56:03.760 power was higher as well, but the ratio was not the same as it was before. VU2 had gone up much more

00:56:09.860 than the power, enough. Suggesting efficiency is going down. Yes, efficiency is going down. And this

00:56:15.440 was then efficiency going down, well, can we somehow get an understanding of this? And this is, of course,

00:56:21.020 on this test, they actually are using a temperature pill. So we put a pill up their ass. 1.00

00:56:24.600 And also at the same time, we have multiple of core sensors around in the body to also measure 0.99

00:56:29.720 the temperature of the body as well. And where we basically see the same thing. We basically see an

00:56:33.760 increased heat production for the same power output, basically indicating the same. You can

00:56:38.060 basically think of that. The reason why you have gas exchanges is because we call this indirect

00:56:41.900 calamity. It's not useful oxygen. It's actually the worst thing you could possibly have because

00:56:46.520 you're also utilizing more fuel. So now you've created a scenario where for the same amount of power,

00:56:51.980 not just you need more oxygen. Oxygen is free. You're going to need more glucose.

00:56:56.840 Well, actually not more glucose because since the RQ here is basically also now in favor,

00:57:01.860 you actually are able to oxidize more fats. Oh, that's interesting. Yeah. So you're saying

00:57:07.180 because the RQ came in the same, you're going to hold carbohydrate metabolism constant and you're

00:57:13.480 actually increasing fat oxidation. Yes.

00:57:16.700 And that works out perfectly in the stoichiometry?

00:57:18.700 Pretty close. It's counterintuitive.

00:57:22.200 This is a rabbit hole I would like to go down into, but-

00:57:24.560 We'll delve gently. We'll delve very gently for the four people that are still listening to this.

00:57:28.820 Yeah. Because also the difficulty with this is that if you just look up purely the RQ of where

00:57:33.640 basically the threshold sits as well for elite athletes or for any athlete that you have in the

00:57:37.260 laboratory. From some literature, you would just say, okay, if you have an RQ, let's say around

00:57:40.580 talking about thresholds, you'll find several places in research literature where they basically just

00:57:45.160 use an RQ of 0.96 as a proxy for your anaerobic threshold, for example. But this is something

00:57:51.560 that you also see is differing. If you basically take an athlete in a training for short course or

00:57:55.320 basically for sprinting, you will basically see that then basically you will have an RQ that is

00:57:58.940 higher than 0.96. If you go to extreme endurance athletes, this actually gets closer to 0.94,

00:58:04.300 93, 92 even. So the implication of this is that if you then measure the lactate threshold,

00:58:09.360 you will actually come into a zone already at 0.92, 0.93, where you actually are starting just

00:58:14.920 as a function of time now, you will get a higher lactate concentration in your body.

00:58:18.880 And if you take that lactate concentration now and you calculate the volume of lactate that is

00:58:23.520 available in your body, you'll actually see that the volume of lactate now becomes a significant

00:58:27.800 contributor to actually your energy production. But you have to know plasma volume pretty well to make

00:58:33.580 that calculation, right? Yes. So plasma volume, but this is a good thing. We have our own machine

00:58:37.700 for measuring carbon monoxide rebreather. So we do regularly also testing of basically their blood

00:58:43.500 volume, plasma volume, and hemoglobin mass as well when we do this testing as well. And of course,

00:58:48.540 you want to know also something about the water content in the body as well. There are different

00:58:52.880 ways. The gold standard is probably double label water, and we've done this as well. But the point is

00:58:57.940 that to give you an example, you can easily at 0.92, 0.93, go from, let's say, if you consider

00:59:04.060 normal lactate levels, or let's say, if we then take it into volumes, it would mean that the energy

00:59:09.180 contribution from lactate could be, for example, five. But if you just stay long enough there,

00:59:14.780 basically, you would say that, okay, if these were the levels, you're talking about now 13%, 14%

00:59:19.120 energy contribution from lactate. And the problem with this is that all the tables we have today,

00:59:23.760 which basically look as a ratio where you use RQ to say something about the ratio between

00:59:27.820 carbohydrates and fats, we normally say that, well, this is good up to an RER of one, but above one,

00:59:33.040 we don't do it anymore, because exactly lactate and other things becomes a too large contributor to

00:59:37.600 it. So this already tells us that we have to be also cautious when we look at RQ and just go

00:59:43.720 crudely into a table and just say, okay, this is your fat metabolism, this is your carbohydrate

00:59:47.800 metabolism, because it is... Do you feel we can do that safely up to one?

00:59:51.100 No. Not even? No, no, no, no. Not even. I would say that for elite athletes,

00:59:55.360 if you're an endurance athlete, I would already start to be skeptical around in the low 90s.

01:00:00.460 Hmm. Very interesting. Sorry, one other question there. What RQ are you typically seeing for maximum

01:00:07.380 fat oxidation? Not obviously percent of fat oxidation, which is low, but in absolute grams per

01:00:13.540 minute, max fat oxidation at an elite athlete who's on a high carb diet, you're typically seeing that at

01:00:20.200 what RQ? That's a longer time since I basically really paid a lot of attention to that.

01:00:26.400 Because I would have to imagine that that corresponds very closely to their race pace.

01:00:32.280 No, because here it also gets a little bit more complicated simply because one of the things we

01:00:36.700 do, because we also do isotope tracing also of substrates as well. We've used quite a lot of carbon

01:00:42.220 13 or 13 carbon over the last years. And we added that as a tracer to glucose and fructose,

01:00:47.600 for example, to look at how much of the exogenous carbohydrates, to use a more common term, how much

01:00:53.500 of the carbohydrates that you ingest that you're able to utilize. Because obviously you have-

01:00:57.140 You have your own glycogen and your own fatty acid.

01:00:59.200 Yeah. And then you want to understand how much- And this is also where exactly when you look at

01:01:04.300 even VO2 max, this starts also to become where you understand that in our context, this is a little

01:01:10.980 bit of less valuable, less valuable purely. Because if you took Christian, for example, or Gustav, and

01:01:17.560 you test it just before race. So you did exactly the protocol we just talked about. You did the

01:01:22.380 step-toe warm-up, a short effort just to make your body ready. Then you did a great exercise test to

01:01:27.040 VO2 max. Okay, fine. You went out, you did your racing. And then basically at the moment you come over

01:01:32.200 the finish line, you test this again, or even half an hour afterwards. It would matter a little bit,

01:01:37.260 but not that much. You will basically see the race pace is actually now coming to the point where it's

01:01:41.080 extremely close. You like your utilization from the beginning of the race to the end of the race

01:01:45.700 have completely changed, significantly changed. Sorry, you're saying fuel utilization?

01:01:50.580 Yes. Not only your fuel utilization, but also actually would also imply fuel utilization,

01:01:55.500 but also your threshold sits now so much closer to, yeah, yeah.

01:01:59.200 This is hilarious. It was about an hour ago that we got off onto this tangent in response to what you

01:02:05.500 said, which was at the most elite level, their race pace is above 90% of their VO2 max.

01:02:13.960 For elite marathoners, yes. And also for Christian and Gustav, when they do Ironman racing, this is

01:02:18.280 getting, let's say, in those domains. We can't push it all the way because it has some implications.

01:02:22.240 Again, this is just impossible for me to wrap my head around. This means that someone whose VO2 max is

01:02:28.020 five liters per minute, if they're elite, which means they're obviously pretty light,

01:02:33.640 they're going to spend an entire Ironman at 4.5 liters per minute. 0.98

01:02:39.660 No. The thing is that what will happen here is that this will actually come down. So your VO2 max

01:02:43.720 that they don't have-

01:02:44.180 Ah, the VO2 max is declining and therefore it's, okay, that's what I wanted to make sure of because

01:02:48.680 I was like, how in the hell can they hold 4.5 liters per minute? Or in their case, 7.2 liters per minute.

01:02:54.920 So this is the difference. This is the interesting thing because this is the difference. The

01:02:58.300 consequence of aiming for Ironman is that you want to have minimal decline in this.

01:03:02.720 Yes. And you need a lower VO2 max.

01:03:04.760 Because it has to do with priority and training. You can't prioritize having like a high one minute

01:03:09.400 power or five minute power simply because it's too far away from specificity of what you really need

01:03:14.140 there. So if you start building good sessions with where you basically are looking to increase your

01:03:18.720 one minute power, five minute power, that's obviously going to have a cost for the whole week that

01:03:22.640 you are doing of training. So let's talk about that now in the context of the Olympics. So you talk

01:03:27.900 about an athlete like Christian or Gustav where in theory, they want to be able to go between three

01:03:32.720 distances in theory, Olympic distance, half Ironman, Ironman. Yes. Let's remind people what the

01:03:39.380 distance is. Olympic is 1.5 kilometer swim, 40 kilometer bike, 10 kilometer run. Elites are doing

01:03:47.700 this in an hour 45 ish. Yeah. Okay. Half Ironman. Well, let's just do Ironman. 2.4 mile swim. So

01:03:55.480 4K swim. 3.8K. Yep. Yep. Yep. Yep. 112 bike. So 180K bike marathon run 42K. The best in the world

01:04:05.000 are doing this in seven ish, seven and a half. What are they doing it in? Christian is the record holder

01:04:11.280 still. He actually did that on his debut and it still stands. What was the time? Seven? 7.21.

01:04:16.920 Seven and a half to eight hours. And then half Ironman is you take one of those last distances

01:04:20.960 and you cut it in half. So just under two kilometer swim, 90K bike and call it a 21K run.

01:04:27.380 So roughly twice the Olympic distance in some regards, less in others. And again, they're doing

01:04:32.920 this in three and a half hours. So these are three very different events. Is it possible to be elite

01:04:40.640 in all of them? This is also quite interesting because one of the main differences I would say

01:04:45.860 that differs them, call them as, or compared to the specialists in the sports is actually not

01:04:51.680 their metabolism. If you look at their view to max on running, cycling, and swimming, you will actually

01:04:58.180 see that there are equally or higher than their peers in those sports. But their efficiency is

01:05:03.660 not the same. This comes probably most down to the fact that they have to do three different sports.

01:05:09.500 You don't get the same time of just pure stimulus, mechanical stimulus from doing something and

01:05:14.200 optimizing it because you have to change. And it also has some priorities also as well. In running,

01:05:19.300 you could say you don't want to compromise on your leg stiffness at all. In cycling, this is a little

01:05:23.840 bit more beneficial to do this, for example. So here you have to strike a balance between the three.

01:05:27.860 Which makes it very complex. But in swimming, they are higher than the highest view to max that

01:05:33.060 are measured on elite swimmers in the world. But the difference is also I had the bronze medalist.

01:05:36.500 The efficiency is unbelievable.

01:05:37.880 It is so poor. It is so poor. It's actually funny. I think we talked about this last time. I had one of

01:05:44.120 the best swimmers in the world in the flume and tested him. And he's a big guy, 195, 100 kilograms,

01:05:50.080 close to 100 kilograms, muscles all over. And then Christian into the flume at the same, 0.99

01:05:55.340 basically same velocity. I don't remember exactly what this velocity is now, but basically this big

01:06:00.800 guy. And the flume is an endless pool?

01:06:03.260 Yes.

01:06:03.560 It's a stationary swimming pool?

01:06:05.160 The difference here is that endless pool, very often people think of it as a counter current,

01:06:09.660 where there's a lot of turbulence. Here we are talking about something which is virtually

01:06:13.140 laminar flow that goes through. So it's a big canal, which is circulating the water on the outside,

01:06:17.580 and it comes back in the front. And you have even honeycomb structures in the front and the back.

01:06:21.240 So it truly replicates swimming stationary. So it's the closest thing we have to an aerotunnel

01:06:26.060 for swimming.

01:06:26.520 Same. Yeah.

01:06:27.260 Yeah.

01:06:27.500 Does it allow you to use dye in the water or anything like that? Bubbles, I assume you use to-

01:06:32.580 You can, yeah. Yeah. The thing here is that when you look at these two guys now swimming-

01:06:37.120 So you put them at the same speed?

01:06:38.320 Same speed.

01:06:39.000 Okay.

01:06:39.220 The lead swimmer, he is even retired. First of all, Christian, eight kilograms, this guy,

01:06:44.920 close to 100 kilograms. And that's not because of fat. He is lean. He is well-trained still.

01:06:50.240 All things equal, he should be much higher in his VO2 at that moment in time. He's simply

01:06:54.440 supporting more muscle.

01:06:55.460 Much more muscles involved and bigger proportions. He's utilizing 25% less oxygen than Christian. 0.84

01:07:01.540 In an absolute term.

01:07:02.560 In absolute terms, yeah.

01:07:03.980 On a relative basis, a third less oxygen.

01:07:06.400 Yeah. Which is still crazy. Yeah.

01:07:09.760 Here's my thesis on swimming. Of all the three big endurance sports, I feel like swimming has the

01:07:15.340 most potential for radical change in performance based on drag avoidance. All of these sports,

01:07:23.640 but especially swimming and cycling, come down to propulsion versus drag. Running is less so because

01:07:29.260 these speeds aren't high enough relative to air. Obviously, swimming, just so people listening

01:07:33.600 understand why I would say that, swimming is much slower than running, but the density of water is,

01:07:37.960 what is it, 1,300 times that of air. So you don't need a high V squared to get a lot of drag just

01:07:43.780 based on the density of the medium you're in, which is water. So in that sense, cycling and

01:07:48.700 especially swimming really come down to this ability to avoid drag. And that's why in a time

01:07:52.860 trial, position matters so much on a bike and how can you generate power, even if you have to

01:07:56.980 compromise your power there. So I feel like in swimming, there could be like a massive breakthrough

01:08:02.580 if a technique emerged that reduced output or power or forward propulsion by 10%, but reduced drag by

01:08:10.760 20%. You know what I mean? Again, I'm so far from the sport, I don't know. But to me, I would really

01:08:17.340 be curious as to that. Because again, remember, this happened in cycling in the 80s, where prior to

01:08:23.140 that, nobody was paying attention to bike position. And then Francesco Moser comes along in 1984,

01:08:28.340 smashes the one-hour record with all this crazy aero equipment. And then of course, you got into

01:08:34.000 Boardman and all these guys getting into more and more crazy aero positions where their actual power

01:08:38.720 went down relative to what they could have been in a less kinked position. But of course, 0.72

01:08:43.240 their speed went up because the CDA goes down. And I just wonder, do you ever think about that in

01:08:47.440 swimming? Is there some major disruption where we just have this dramatic change in technique

01:08:53.280 to have a bigger positive impact on frontal surface area than the negative effect it might have on

01:09:00.100 propulsion? Yeah. So this then touches up on one of my favorite areas and that's biofeedback.

01:09:07.120 In cycling, we have extremely good tools for biofeedback today. You have a power meter

01:09:11.640 and you have a GPS. So just these two combined, for example- Speed and power perfectly.

01:09:15.200 Yes. Like we talked about last time when you were out riding 200 watts constant,

01:09:19.080 and basically you see you're getting faster. So you have this direct biofeedback because

01:09:23.160 as long as you ride enough, you will basically start to get a very good feeling for when you sit

01:09:28.480 at a certain power and you start doing different things. Suddenly you creep up like half a kilometer

01:09:32.320 power or what? Like for a person, when they start cycling or they're cycling a little bit like,

01:09:36.380 for them, it doesn't become interesting enough yet because there are other things that is basically

01:09:40.580 more challenging for them to master. But for people that do a lot of biking like yourself,

01:09:45.780 it's where you exactly start to pay attention to those, let's say even half a kilometer per hour,

01:09:50.440 maybe even you're getting below that as well, where you start to really pay attention to this.

01:09:54.440 And then over time, simply because wind and other factors makes a difference too.

01:09:58.160 Running, this is also a place where we have really good biofeedback tools. You have your watch,

01:10:02.100 you're running and you can have a look at it. And now we have also really good power meters in

01:10:06.080 running as well. How does that work?

01:10:07.720 So most of these, there are some of these power meters in running today that requires an insole.

01:10:12.480 So you put basically an insole inside and it measured the forces.

01:10:14.820 So it's a force plate inside your shoe.

01:10:16.820 Yes. And then you have other ones that are more motion capture devices. So they basically,

01:10:21.180 you rather input your body weight. And since when you're touching on the ground with one foot,

01:10:25.880 you're basically carrying your whole weight there. So as long as you have a good enough motion

01:10:29.880 capture device that are capable of capture the three-dimensional accelerations,

01:10:34.320 you can then basically also say some, well, you know, the force because you basically have to

01:10:37.780 carry your weight.

01:10:39.080 So you can directly measure this with a force plate insole, or you could indirectly do it

01:10:43.740 with motion capture. But I assume the motion capture only works on a treadmill.

01:10:46.480 No. So the motion captures today, they are, it becomes so small. They're basically a small

01:10:50.000 device that you attach to a shoe and they are, when we validate these.

01:10:53.160 Are these commercially available products?

01:10:54.780 Yes.

01:10:54.880 Yes. If you go to a laboratory and you basically test this, you will basically see that. So we

01:11:00.280 have been for a long time in working together with Stride. I started working with them when

01:11:04.040 they were in beta stage and we've gone through there, but they are so accurate today that when

01:11:08.720 you measure on an athlete.

01:11:11.400 First of all, do you have any idea how much you just ruined my wife's life? 0.96

01:11:14.820 Like, do you have any, so how many minutes do you think after this podcast, am I going to be

01:11:20.380 on my computer ordering these devices, jamming them on her shoes? You know, she's running the 0.99

01:11:25.260 Boston marathon next year and I'm convinced she's actually going to run it faster than she did her 0.56

01:11:30.560 first Boston marathon 20 years ago. Cause she actually now works with a running coach and her 0.92

01:11:35.380 qualifying time this year was only one minute slower than her qualifying time 19 years earlier.

01:11:41.580 And again, it's just because she's more structured in her training, not because she listens to a word

01:11:46.620 I say she doesn't, but she's like finally at least agreeing to use heart rate and velocity

01:11:52.320 for tempo training. She's not listening to this podcast. Obviously one of her friends will

01:11:55.980 probably hear this and tell her to listen, but we're going to implement power training for her 1.00

01:11:59.300 running and she's going to curse me all day long because she doesn't want data. But how is every

01:12:04.860 runner not doing this? I think this has to do with a lot with tradition. And also that even when

01:12:09.840 power meters were introduced, it's easy to look back and say like, why didn't we have this before?

01:12:13.760 Or even how power, there's so much information we can extract from a single power meter today,

01:12:18.520 which is, I say, beyond people's comprehension. Still, we only use the power number that is there

01:12:23.840 and we use it even in a one dimensional context, FTP, for example, or critical power.

01:12:28.300 The amount of information you can get out of this is crazy because we are not going to talk about

01:12:32.740 this today. We still kind of use normalized power and other... No.

01:12:37.140 We never use it? No. We only go by raw numbers.

01:12:39.760 I only work by raw numbers all the time. So I don't condense it down to a single metric. I use

01:12:44.260 only raw numbers. But the implication here is that even when you look at studies that are looking at

01:12:48.680 gross efficiency, for example, in the old days, or even today, one thing you consider is a net

01:12:53.200 mechanical power. So the interface between basically looking at your gross efficiency

01:12:56.980 is where you take VO2 and we're already there. Most people understand that, okay, you have a difference

01:13:01.700 between net oxygen consumption and gross oxygen consumption. You would ideally like to have the net

01:13:06.040 oxygen consumption. There are other things there. But in cycling, actually, also what we only consider

01:13:10.760 there is the net mechanical power. You don't consider the gross mechanical power. On your

01:13:15.220 biochemical efficiency, you should not look at the net mechanical power. You should look at the gross

01:13:19.020 mechanical power. But that's even before we start to get into vectors, power vectors, force vectors,

01:13:23.360 and other things, even in a three dimensional plane as you do cycling. Because this is something we

01:13:27.520 can extract from the power meters today.

01:13:29.440 So when running, you have more degrees of freedom. There's more inefficiency. There's

01:13:34.780 probably a lower relationship, a more strained relationship between gross and net power.

01:13:40.400 This is maybe one of the reasons for why it is not as widely adopted in running as it is in cycling,

01:13:46.160 because it is still debated what really is running power. How do you really quantify it? And this is

01:13:51.840 something that I even discussed with the team at Stride. So when we are having our regular calls and we

01:13:56.320 are diving into the topic, even we also sometime have our different opinions on how this really

01:14:02.740 should be looked at. Because if you want to translate this into running, then you have to

01:14:06.220 look at only at the propulsive power. Could you imagine seeing this? Have they put these in Olympic

01:14:11.000 sprinters? The power numbers there? No. Commercially available, this has its limitations. Because since

01:14:15.980 you do motion capture, you need to do a little bit of filtering. You can't take out because it's going

01:14:20.140 to be so much noise. Exactly because as you said, you have so many degrees of freedom. If you're going to

01:14:24.860 output all the vectors that basically you were able to measure with strain gauge-based power meters on

01:14:29.660 a bike today, people wouldn't be able to utilize. That's why you condense it down to a single number

01:14:33.540 that is there for most people. If you had that force plate chip in the insole, you could at least

01:14:38.980 capture force normalized to weight with each step, right? Yes. So what you can do, and this is

01:14:44.520 validated. So if you, for example, run on a track, a track which has force plates, or you run on

01:14:49.160 specialized treadmills that has force plates integrated in them, you will basically see that the curves are

01:14:53.660 the same. So this is the way, let's say, this is the external validation of the device is good, both in terms

01:14:58.080 of that it captures the force curves, but also here, when you have motion capture devices, you can also

01:15:02.540 capture the footpath as well. So this is something we can visualize in 3D today, after the event, when you

01:15:07.660 are doing running. We can see what happened there, fresh, fatigued, throughout the race, when it's technical

01:15:12.600 and other things. But I think the reason why this is not as widely adopted is because in science, this is still

01:15:18.380 debated on how do we really capture and quantify mechanical power for running. Stride have taken a

01:15:25.100 smart approach to this, to make it commercially viable, and that is that they output it as a

01:15:30.160 metabolic power. So if you actually went on a treadmill, and you went running, and you looked

01:15:34.640 at it versus your oxygen consumption, you will basically see that this matches perfectly with

01:15:38.420 cycling. That's without having a bicycle near you at all. So you can say that, well, this is what you

01:15:42.500 expect. So when you have a certain power, then this would have a certain metabolic cost.

01:15:45.920 I don't like it because I don't like modeled numbers. I want to have raw numbers. So in my case,

01:15:52.060 I'm extracting the net and the gross mechanical power, or the positive and the negative mechanical

01:15:56.220 power, because these are the components I want to have, because I have VO2 master, I have metabolic

01:16:00.740 devices. I don't need a metabolic equivalent. I want to have the raw because I'm using as an interface

01:16:06.060 to gauge the difference. Like for example, on a Formula One car, what really matters in the end there,

01:16:11.100 how fast can you go around the track for the full event with a certain amount of fuel? Because

01:16:15.780 there's a limitation to how much fuel you can have in your car. That's the true input and the true

01:16:19.320 output. Your engine output and this kind of thing is actually secondary. It's easy to think,

01:16:23.620 oh, we won't have the biggest engine, but big engine without efficiency is still bad. The engine,

01:16:28.260 it becomes more of a device to measure. Let's say you look at the power and you look at how efficiently

01:16:32.860 are you able to translate this fuel out into speed over, let's say, the event or velocity.

01:16:38.120 For me, having access to the net mechanical power and gross mechanical power, and then you have the

01:16:42.360 metabolic devices, this allow me to, on the one side, look at when we change something in running.

01:16:46.540 So let's say you change, for example, your shoes, or you change your training. You can now have a much

01:16:52.180 more granular understanding of, am I influencing the biomechanical part of the training, or is it

01:16:56.540 the biochemical part of the training? But in order to do so, you have to have that interface that

01:17:00.700 distinguishes between gross mechanical power and net mechanical power. And the interface between gross

01:17:05.660 mechanical power and net mechanical power is only the, let's say, work efficiency. How efficient are you

01:17:10.360 working, but not metabolizing or moving? Because in the end, you have to take this power and be able

01:17:15.980 to output into velocity. If you're looking at a cyclist, for example, can you make a statement

01:17:20.180 using something so simple? So if you had 10 different cyclists and you did a VO2 max test on all of them,

01:17:26.560 and each of them, you end up getting a number, which is going to be, let's normalize it to weight.

01:17:31.580 So you line them up in rank order from the lowest to the highest in milliliters per minute per kilogram.

01:17:37.220 And then you look at what power they were at when they achieved that VO2 max, and you normalize that

01:17:43.480 to weight to watts per kilogram. It's not going to be a one-to-one match.

01:17:48.140 The interesting thing is that when you are getting close to VO2 max, then it's getting close to a

01:17:52.060 one-to-one match actually. But a submaximal, it doesn't. A submaximal you will normally see in

01:17:56.380 running simply because it's a weight bearing sport. Just to make sure I understand that, are you saying

01:18:00.440 that the rank order will be identical for VO2 max and watts per kilo at VO2 max in cycling?

01:18:08.180 In running and cycling at maximal effort, so basically at VO2 max, pretty much the same, yeah.

01:18:13.380 But I feel like my personal power at VO2 max is lower than many other people's at a comparable VO2 max.

01:18:21.980 Like, I just feel like I'm very inefficient. I have a higher VO2 max than a PVO2 max,

01:18:26.320 if that makes sense. Yeah. And that's because probably there is a lot of anaerobic contribution

01:18:30.740 in there as well that basically supplies that gap that you're observing there too.

01:18:34.660 Oh, you're saying I'm anaerobically not trained enough and that's why my power, yeah, yeah. I'm

01:18:40.340 too aerobic, not anaerobic enough. Yeah.

01:18:42.500 Great point. That's a very interesting point.

01:18:44.640 So the thing here is for Christian Gustav obviously being, or we're trying to keep exactly the training

01:18:49.220 as balanced as possible between the three sports, there you will normally see at maximal efforts

01:18:53.600 that actually that the power also is pretty much the same. The mechanical power is pretty much

01:18:57.640 at VO2 max, but as you go at submaximal effort, then that's where you start to observe the big

01:19:01.760 differences. Because in cycling, you would see that as you go down in intensity, basically because

01:19:09.000 of the weight-bearing sport, going from 15 to 16 kilometers power, let's say 17, 18, 19, 20,

01:19:16.240 and so on, basically what you see, but this is also observable exactly from a metabolic heart as

01:19:20.080 well. And that is when you go at, let's say, a low intensity, you will see that your oxygen

01:19:23.820 consumption also is higher at the perceived lower intensity in running than it is in cycling.

01:19:28.660 Because cycling is not weight-bearing, so it's easy to go down to a very low intensity. Still

01:19:32.800 keep a normal cadence because you can keep a cadence of 80, 90, for example, and it feels

01:19:38.000 comfortable. Feels comfortable and easy in this kind of way. But going down to those cadences

01:19:41.520 in running is impractical because it means walking. You're not running anymore, so you're walking.

01:19:45.500 So the modality is changing more in running. What kind of muscles are involved are changing more

01:19:49.540 in running than it does in cycling. Coming back to the question, original question of

01:19:53.580 why isn't this more useless in running, it is gradually getting more and more attention there,

01:19:57.940 more and more people using it. I think what is still to be determined is to agree on, let's say,

01:20:04.780 a common standard for what numbers should be outputted. In cycling, one could ask the question,

01:20:10.540 so when the guys are on the bike, are they paying more attention? So let's just talk about them doing

01:20:15.800 a four-hour ride in an Ironman. Are they more concerned, because you're triangulating between

01:20:20.820 RPE, heart rate, and power. How are they prioritizing those things? How they feel,

01:20:26.220 what the heart rate is, and what the power meter says. How are they regulating effort based on those

01:20:31.720 things in a race? In training or racing?

01:20:33.900 Race.

01:20:34.400 In racing, I would say that training, we use also many more sensors, and then we limit it a little bit

01:20:39.160 more in racing simply because it's not practical to do it there. But I would say that this is also a

01:20:43.920 very interesting topic because then they touch a little bit more on psychology, but we'll come

01:20:47.000 back to that another time. But in racing, obviously, first and foremost, Ironman racing is so long.

01:20:52.900 It's so long. And with all this practice, you shouldn't be slave to the numbers, for sure,

01:20:57.360 because you can have suddenly your hero day where you just are able to go faster than what you

01:21:00.940 normally do. So you need to listen to your body. So I would maybe say that RPE is the most important

01:21:05.400 one in some senses. Or RPE is also a place where you can also talk about one dimension,

01:21:09.460 two dimensions, because RPE is normally, let's say if you take Kipchoge, when he did a two-hour

01:21:14.760 marathon, sub-two-hour marathon, if you put him on a treadmill and you did a normal step test on him

01:21:19.900 and you asked him when he was at 21 kilometers per hour and you asked him, how does this feel? He

01:21:23.340 would say, for the sake of the discussion, we say that seven is a round threshold or anaerobic

01:21:27.220 threshold. If you relate to FTP, something you would be able to hold around an hour for well-trained

01:21:32.680 athletes. Or what you would test for when you did a 20-millimeter protocol and then you

01:21:36.020 subtract the 5% to extrapolate it to that. If you asked him a 21-kilometer power, okay,

01:21:40.440 how do you feel here? And he gives you an RPE score. He would maybe say, ah, this feels like a

01:21:43.980 six when he has a 21-kilometer power or actually slightly above. If you ask him at the end of a

01:21:48.880 marathon, how do you feel? And he's still running the same pace. He would say probably 10. This is a

01:21:52.440 nine to a 10. Because we forget, again, there's a dimension and that is a duration. Let's say the

01:21:56.800 fatigue component that comes in there. And that's why it also helps sometimes to ask, I would say that

01:22:01.040 when I ask Christian Gustav or at least- But also his heart rate might be different there too.

01:22:04.700 When you throw him on for a minute at 21 kph and he says RPE 6, his heart rate's probably a lot lower

01:22:11.940 than it would be two hours later at that pace, right? But that brings us back a little bit also

01:22:16.080 to exactly utilization as well. Because one of the things that you would probably not be able to do,

01:22:21.540 you won't be able to get yourself up to a maximum heart rate at the end of a race even. So your max

01:22:25.920 heart rate, you could say in many ways are limited at this time. But this is also when you then say,

01:22:30.200 okay, but how on earth is it possible to ride at 4.5 liters when you have a five-liter view to max?

01:22:36.520 This is the key, obviously, in endurance sport. You want to increase this robustness.

01:22:40.880 So when you are finding this optimum of between of where should my view to max is,

01:22:44.700 this is what happens with specificity. If you're tuning this down, exactly where you start and

01:22:49.440 where you end becomes far less. And there's a whole range of implications to this because it comes down

01:22:56.280 to why is this important? Couldn't you just have a higher view to max coming into the race and just

01:23:00.220 have a higher view to max? Well, the problem with that is that the heart also is a muscle and it's

01:23:05.080 not very efficient muscle. So even the heart has energy. It's not insignificant. It's actually a

01:23:10.060 significant energy consumption as well. And if you have a heart that actually is trained for something

01:23:14.480 different than what you really need, it means basically also your heart is going to use energy

01:23:18.400 more inefficiently. How much lactate does the heart generate?

01:23:21.600 Probably you are better suited to answering this. I probably read it sometime, but I've never done any

01:23:25.940 research on that. I looked more at, let's say, when we broke down the body into components,

01:23:30.760 it's more like where I looked at it basically is energy consumption.

01:23:33.120 What's the mitochondrial density of cardiac muscle versus skeletal muscle? I mean,

01:23:37.320 they're both similar in some ways. They're striated muscles. I should know that. I mean,

01:23:41.280 your intuition would be it has to be very rich in mitochondria, right?

01:23:44.600 Yes. And then that's also why the heart obviously uses lactate also as a fuel source simply because

01:23:50.000 it is probably full of... I've never done a muscle biopsy of the heart itself. It would be

01:23:55.600 interesting just to see what's the ratio between type 1 and type 2 fibers there, but I would almost

01:24:00.100 imagine this is probably one of the muscles where you are as close as possible to...

01:24:04.260 Just to pure type 1.

01:24:05.140 Yeah, yeah, yeah.

01:24:05.920 Yeah, yeah. And I learned recently in the podcast with George Brooks that we can actually shuttle

01:24:10.940 lactate into the mitochondria for oxidative phosphorylation. I was completely unaware of that.

01:24:15.680 That would explain, of course, why the heart could richly use lactate.

01:24:19.380 Yes. And this also comes down to Mikael's constant, where you also look at the affinity for different

01:24:25.920 substrates, also for different muscles as well. And where lactate has the highest affinity for most

01:24:30.940 of these muscles as well, especially like the heart. But if you're going to bring this back up,

01:24:35.520 a lot of the research when you do apply research like we do, a lot of the times we can't actually

01:24:41.220 use a lot of the research that is there to base our decisions on. I don't know if we talked about

01:24:46.340 monocarboloxate transporters last time or not, but one of the things that is there is that this is

01:24:51.600 also a place where if you want to understand something, you can go in and you can do a

01:24:55.820 concentration measurement. The problem very often with research is that we get a partial view of

01:25:00.660 something, but it's only partial and it's not a complete view. And that means that, for example,

01:25:06.160 most of the time for us to understand what really works and doesn't work, we have to work with just

01:25:12.980 raw numbers. We talk about calorimetry, so we want to understand what kind of substrates are

01:25:17.060 being utilized here and how this is. One thing that we basically see when we're using isotope

01:25:21.140 traces and we start to dig into this, and we can also talk about actually maybe creating lactate as

01:25:25.660 an artificial fuel. So in the same way that you take glucose and you create supplements based on

01:25:30.780 fructose and glucose, and you use this as a fuel source or ketones or beta-hydroxybutyrate,

01:25:36.340 then basically lactate actually is very interesting in that context, simply because it is extremely

01:25:44.240 energy efficient. And that is in the end limitation for elite athletes.

01:25:49.060 How difficult to deliver orally? I assume it's delivered in a salt?

01:25:53.240 This is actually something we started discussing. The interesting thing of it is in a salt is that

01:25:58.280 it could also then in that sense has a little bit the same effect as bicarbonate. So you could use

01:26:03.340 it as a buffer. Yeah, let's make sure folks understand the chemistry there. So I was going

01:26:07.640 to actually ask you about this and then we got off onto another topic. I want to come back to that

01:26:10.900 exact question, but we'll preface it with this question. People ask me all the time, hey, I

01:26:16.080 understand, Peter, that as my workload increases, my production of lactate increases. And as my production

01:26:22.580 of lactate increases, my capacity starts to fall off because as lactate goes up, it's buffered by a

01:26:28.920 hydrogen ion or it's married to a hydrogen ion. And that's what creates the acid part of lactic acid. 0.99

01:26:35.420 And it's that hydrogen ion that's causing all the trouble. It's not the lactate. We can tolerate

01:26:39.560 endless amounts of lactate. And lactate is crucial in metabolism. Yeah.

01:26:42.660 Yeah. We just can't tolerate the hydrogen that comes with it. And it's that hydrogen

01:26:45.660 that actually paralyzes the actin myosin filaments, prevents them from disengaging. And that's what

01:26:51.940 leads to that seizing up that you feel, the rigidity you feel when you exceed your lactate

01:26:56.380 threshold. So the question then becomes, well, can I buffer this? And everybody and their brother

01:27:01.040 talks about, hey, what if we took lots of Tums, anything that we could get sodium bicarbonate

01:27:06.520 into our systems? Looking at that literature, which I haven't done in a while, this strategy

01:27:10.700 didn't really pan out. There wasn't really a great way to orally ingest enough bicarb to make a

01:27:16.460 difference. Obviously intravenously you could, but that's probably illegal anyway. I mean, I don't

01:27:21.440 think water would permit that. But even if you weren't concerned with that and just asking the

01:27:25.580 theoretical question, unless you're on a stationary bike, it's not practical to have an IV drip

01:27:30.400 of bicarbonate to buffer your hydrogen. So what is the state of buffering agents to reduce and lessen

01:27:40.120 the impact of lactic acidosis? This is a little bit of a complicated domain.

01:27:45.740 Unlike all the other simple domains we've been discussing today. Yes.

01:27:48.940 The reason for that, I've been very fortunate over the last decade to be involved in a lot of edge

01:27:54.460 case researchers. Plenty that we are years away before this will be published. Not necessarily

01:28:00.140 because we don't want to publish in it, but there just remains work to be done, to understand it.

01:28:05.280 And for example, again, this is also a place where we have some indications now, but where some of my

01:28:11.740 colleagues, they have actually also tested on larger population. But Gustav's highest view to max

01:28:17.120 actually is done under bicarbonate utilization. Not Christian, but Gustav. 0.56

01:28:22.800 How? How is it administered?

01:28:24.260 So the way it was administered is that there's a company in Sweden called Morton. What they did,

01:28:29.680 this is actually something that is going to be studied now also in actually in health,

01:28:33.180 in medical settings as well, because it actually has some quite interesting applications there too.

01:28:38.600 How this was administered is basically that it is packed in a hydrogel and this allows you...

01:28:43.480 Oh, so you get rid of the gastric, you would bypass the gastric pH and you get it into the intestine.

01:28:49.160 Medicine, using the same mechanism. You basically pack the agents into a vehicle

01:28:52.600 to deliver it where you need it more efficiently. So this means that we can go to concentrations that

01:28:58.900 are significant.

01:29:01.560 What have you seen is the difference in his lactate tolerance with and without this buffering?

01:29:07.420 So this is a place again, where obviously the sample size is still limited.

01:29:11.940 Sure. But in him, in a world-class elite athlete.

01:29:14.960 So the interesting thing there is that here we have two different cases. And that's, for example,

01:29:19.020 I have some athletes that have almost twice as high lactate concentration in the blood when they're

01:29:25.300 using. This is why I said we are opening a can of...

01:29:27.700 The proverbial can of lactate buffered worms.

01:29:30.280 Yes, exactly. We also have them to remember because then we have to come back to the technology,

01:29:34.560 where do we measure...

01:29:35.720 And what is this called again?

01:29:37.260 So this brand is called Morton.

01:29:39.080 Morton, as in Morton salt.

01:29:40.560 It's written M-O-R-T-E and Morton. People go crazy over the gels and everything during

01:29:47.300 races. They have a gel that is actually based on the same principle. So we open up that box

01:29:52.220 as well. But basically Christian Gustav sits comfortably and not eating 160 grams of carbohydrates

01:29:56.340 per hour. And basically this is then quantified using isotop traces as well. So they don't only

01:30:02.400 eat it and it starts packing up in the stomach, they utilize it. 0.99

01:30:06.560 Yeah. I'm going to save time to talk about nutrition because I want to ask you about that.

01:30:09.200 So this is a commercially available product. Athletes are using it. So do you have a sense

01:30:13.440 of why some people find... Like Gustav seems to find huge benefit from it. Sounds like Christian 1.00

01:30:17.300 does not.

01:30:18.060 Well, I wouldn't say that Christian doesn't, but it depends a little bit on, let's say,

01:30:21.460 the setting we are using it in as well. And triathlon is inherently complex in the sense that

01:30:26.140 you have three sports that you put together and with a very varying intensity, all the things that

01:30:30.360 are happening there, which makes it a little bit more complicated. And then there also is some

01:30:34.180 benefits, but can also potentially be disadvantages with it. But the benefits that also comes with it

01:30:38.960 is that it actually also increases your plasma volume fairly instantaneously as well.

01:30:43.280 You're pulling fluid into the plasma.

01:30:44.920 Yes. Yeah. But early studies that have done on this, which will be published on this,

01:30:48.660 basically shows that there's a positive effect of using bicarbonate now or using,

01:30:53.020 for example, the molten product. We have studies now on larger population done by some of my colleagues,

01:30:57.200 which indicates this. Then on top of that, what I would say is that we don't fully understand why

01:31:04.100 it is like this. It's a one interesting hypothesis I presented to the research group is that because

01:31:11.060 on the one side, what we think is that, okay, this increases the buffering capacity. So we think,

01:31:15.380 okay, if it increases the buffering capacity, then you can basically do more, then you can go

01:31:19.000 more anaerobic or more into anaerobic or glycolytic resources. Because one thing that we are fairly

01:31:25.720 sure about is that you're never going to run out of glycogen, really. You come to a certain level

01:31:29.720 and the body will rather start to sense that it's really, really low. And that's why you shut down.

01:31:33.820 You don't shut down because you basically depleted every gram of glycogen in your body.

01:31:38.200 That's one. And then there are other preserving mechanisms.

01:31:40.700 By the way, that's easy to verify with a muscle biopsy at failure, right?

01:31:44.140 No. The problem with that is that even in muscle biopsies, you will see that if you basically do

01:31:47.920 sampling just across one muscle, it's so heterogenic.

01:31:50.920 If you wanted to do it, you would need to biopsy multiple times, multiple sites simultaneously,

01:31:55.540 or even make somebody radioactive and start to do nuclear measurement measurements on people.

01:32:01.220 Yeah. It's so impractical to really study that you have to go more by, again, coming back to,

01:32:05.960 for example, first order principles. For example, looking at gross efficiency and then calculating

01:32:11.240 us basically how much subsidy you're using basically is already involving inaccuracies to it.

01:32:17.700 That's why you just have to sometimes just say, okay, we are going to look at oxygen consumption

01:32:20.880 versus, let's say, mechanical power output. And then you just say, we don't care about whether

01:32:24.780 that's RQ correlates to this value. You just have to have the raw values.

01:32:28.460 Back to the bicarbonate, what we can observe, for example, is that when we do blood gas analysis,

01:32:33.160 so when the athletes are exercising, basically a longer protocol, we are taking blood samples

01:32:37.600 to look at pH level in the body. And what is interesting to see there, even for Christian,

01:32:42.720 he does not have a doubling in his lactic concentration. So Christian have almost 0.99

01:32:46.360 unchanged lactic concentration in his blood.

01:32:48.660 At what level of exertion?

01:32:50.520 Whatever. Even all out.

01:32:52.080 Any given power?

01:32:52.920 Any given power.

01:32:53.780 Or any given VO2. Yeah. Okay.

01:32:55.380 Yeah. But the interesting thing here is that, the interesting thing here, this is easy to think now

01:33:00.260 that when an athlete has double the lactic concentration in the blood, then it has to be

01:33:04.800 double the contribution from the glycolysis. We have to remember, we measure in the blood,

01:33:08.600 this is a concentration metric, and the state can be completely different other places in the body.

01:33:12.700 Here's just a crazy idea. Have you ever done a muscle biopsy on the two of them to see

01:33:17.600 the relative differences in monocarboxylate transport density on their muscles?

01:33:23.000 No.

01:33:23.720 Think about this as, I mean, just for the listeners to understand what we're talking about,

01:33:26.660 the MCT transporter on the muscle cell must play a significant role in determining

01:33:32.600 the relationship between intracellular lactate and intraplasma lactate. It would be in an athlete's

01:33:39.460 best interest through training to increase the density of those, because the more you can get

01:33:44.640 lactate out of the cell, the more presumably you're going to get the hydrogen with it out

01:33:48.440 of the cell. We probably have a greater capacity to buffer acid in the plasma because we have the

01:33:54.360 respiratory drive to adjust bicarb than we do in the cell where that hydrogen is really poisonous.

01:33:59.580 So it just makes me wonder, as a hypothesis, maybe Christian has more MCT density, and that's why 0.96

01:34:08.720 he is less impacted by this buffering strategy. I don't know. It could be the other way around.

01:34:14.140 I wonder if all things equal, that would be... It's just so hard to believe that two world-class

01:34:18.780 athletes could be that different in their response.

01:34:22.260 One can be MCT transporters, for sure. One can be, but I doubt it to be significant.

01:34:27.960 Yeah. It's hard to imagine that's 2x difference.

01:34:29.900 And the reason why I doubt that to be significant is because if I look at purely the biochemical

01:34:34.460 efficiency, first order principle, looking at oxygen versus power output, cross mechanical

01:34:39.060 power output, this is so close that it can't explain it alone. One other difference that is

01:34:45.200 there, and which is maybe larger, but I don't know, maybe are closer related to this is, for example,

01:34:52.140 exactly plasma and blood volume. Because the blood volume and plasma volume in Christian and Gustav

01:34:57.880 is beyond significant in difference.

01:35:00.800 Why? They're not that different in weight, are they? 80 kilograms versus...

01:35:05.400 But still, it is borderline where the word significant doesn't do the difference. One of

01:35:11.020 the places where we don't have any definitive answers, it is just different. But then you can

01:35:15.240 say, well, it wouldn't have had implications on, for example, the VO2 max.

01:35:18.620 And stroke volume, cardiac output.

01:35:20.420 Yeah. Well, actually, that's the interesting thing, because if you talk about stroke volume,

01:35:23.600 I would agree, yes, there is a massive difference. Again, significant doesn't really do it justice.

01:35:28.600 We would talk about massive differences in stroke volume, but not in cardiac output.

01:35:32.980 What's the max heart rate of each?

01:35:34.520 Christian, ballpark-ish around 180, maybe 170, 180. I would say 178, something like this.

01:35:43.900 Gustav, around 200. So this is one, but also more interesting here as well, is that Christian 0.75

01:35:50.280 has a much larger also hemoglobin mass than what Gustav have. One thing we also have to remember is

01:35:56.520 that if we create a performance stream, more or less, and we put VO2 max on the top of this one,

01:36:00.580 where all these different factors are contributing, let's say, contributing factor to your VO2 max in

01:36:05.720 the end. That's why we say that VO2 max is the holy grail or such a good metric, because if there's

01:36:11.320 something broken somewhere in the system...

01:36:12.960 It trickles up.

01:36:13.580 It trickles up, yeah. But it doesn't mean that if something is broken here,

01:36:17.140 that that's a good metric, even though we could say that it has been used in research and other

01:36:21.440 things. So probably one of the most plausible explanations for why there are such big differences

01:36:27.520 in Christian and Gustav in terms of, let's say, where the biggest differences comes from more

01:36:33.380 is that Gustav has to circulate his blood much quicker than what Christian does, not relatively

01:36:40.500 speaking, but in absolute. So the absolute circulation of blood, cardiac output has to be

01:36:45.200 at least the same for Gustav, maybe actually a little bit higher than what it is for Christian.

01:36:50.120 But Christian compensates in other ways. And that's also an interesting thing then,

01:36:54.220 basically, when we talk about all these different things, whether it's MCT or other things that are

01:36:58.360 in the body, is that there are so many mechanisms in the body that we see that when you come

01:37:02.380 closer and closer to elite level, other systems have to start to compensate for you to take the

01:37:08.340 next step. In people that are not well-trained, then you can basically do whatever it is,

01:37:12.900 and the body will just prioritize to develop what is the easiest to develop. But at elite level,

01:37:16.760 it seems like more like now it starts to be like a really hard priority. And that's even before we

01:37:21.700 start discussing epigenetics. Let's go back and say one thing about temperature. There's some

01:37:26.260 reasonable data suggesting that a relatively high dose of acetaminophen can certainly improve heat

01:37:32.520 performance. So tolerance, race pace tolerance in warm temperatures, that it may even perform output.

01:37:38.700 And again, nobody knows why. Is it improving performance, i.e. absolute output, because it's

01:37:45.300 actually reducing and blunting body temperature, and that temperature itself becomes a bit of a

01:37:49.960 governing mechanism on output? Or if it's just blunting pain, and pain is part of the wall that

01:37:56.000 we face. But curious if you have any experience with high doses, 1 to 1.5 grams of acetaminophen

01:38:01.920 in any of these athletes?

01:38:04.240 No, we don't. We don't use it.

01:38:05.960 Have you experimented with it in training?

01:38:07.620 No. One of the reasons why we don't do it is because I believe at the moment you start

01:38:13.120 manipulating what is not manipulating. Already when you're using exogenous, for example-

01:38:18.060 Bicarb.

01:38:18.660 Bicarb hydrates for that sake, or anything like this, you can already say, okay, well,

01:38:22.760 this already also has an effect. But I would say that the problem when you try to go in and you try

01:38:27.940 to acutely, let's say, do something in the body, it means basically that you're trying to target one

01:38:33.880 part of one system of the body and you don't consider the other ones to be important. What

01:38:38.580 I more believe in is that, for example, if you talk about heat, for example, and heat tolerance,

01:38:42.620 if you have more of a natural approach to this and you build this into your protocols,

01:38:47.180 this means basically that, let's say, whatever is in the body that is there in order to help

01:38:51.920 increase your heat tolerance have to adapt to this. So for example, pain is something that is

01:38:57.160 trainable. And there are many mechanisms that are involved with this. Using, for example,

01:39:01.960 supplements to try to lower the pain, for example, or the perceived pain of something.

01:39:07.700 I think this is a place where, talking about, for example, performance-enhancing drugs,

01:39:12.640 it's interesting to see that despite post-big doping scandals that were in cycling, everybody

01:39:18.640 thought that, okay, now Tour de France will become slower. Did, short period, and suddenly it just

01:39:23.940 became faster, became faster, became faster. I don't think this is because people are doping more

01:39:30.080 now than what they did before. There are probably some cases where people still do doping or they

01:39:34.260 are really like venturing into the gray area. But obviously we have found techniques and other things

01:39:40.100 that are more powerful than using performance-enhancing drugs to get to this level.

01:39:45.620 I'll share two with you. I'm curious to your thoughts. We're going to talk about nutrition in a

01:39:49.640 minute. You already alluded to the fact that your athletes are routinely able to

01:39:53.700 consume 160 grams of carbohydrate per hour on the bike during competition. A hundred years ago,

01:40:01.640 when I was competing in anything, we were stuck at 60. It was really hard to get past 60 grams per

01:40:08.960 hour. For ultra-distance things like I did, it really became a bit of an energetics problem.

01:40:14.320 You had to start figuring out ways to get fat in the substance that you were consuming just to get

01:40:18.640 the additional calories, even though you didn't really need fat because you have enough of it.

01:40:21.680 It's glucose you're limited by. I very recently, literally the other day, asked Lance Armstrong,

01:40:27.520 we were talking about something unrelated. And I said, by the way, Lance, back in the tour,

01:40:31.740 how many grams of glucose were you consuming per hour? You know what he said? He said,

01:40:37.980 we didn't even pay attention to it. We just ate when we were hungry. So this is not saying that the only

01:40:46.000 reason those guys were exceptional is because they were using blood products. No, they were using

01:40:50.400 blood products and they were exceptional and they trained really hard, but their knowledge of

01:40:55.140 nutrition was very pedestrian to what it is today. As you know, if you wait till you're hungry to start

01:41:01.800 fueling, you're not fueling in an optimal strategy at all. So that to me is one enormous advantage in the

01:41:09.020 Peloton today. I think that nutrition science has evolved so much what these guys can do. I've

01:41:15.480 heard rumors that they can put down 200 grams per hour, that they've trained themselves up to that

01:41:19.900 level. I think a second interesting difference in the tour today, and I'd love your point of view on

01:41:24.860 this as well. I think people forget how big cyclists were 20 years ago relative to today. If you look at

01:41:31.760 the GC contenders in the era of Lance Armstrong, if you look at Jan Ulrich, Ivan Basso, Lance himself,

01:41:39.380 I mean, these were guys that weighed 70 kilos. Lance raced at, he'd start the tour at 74 and finish at 72

01:41:47.800 kilos. These are normal sized human beings. If you look at the GC contenders today, these guys are 58

01:41:56.000 kilos. I mean, they're very small. So in cycling, if we think about it as watts per kilo, it's not that

01:42:04.500 they've gone up that much on watts. Their absolute wattage is significantly lower than what it was 20

01:42:10.480 years ago. They've just gone down so much on weight. So I'm wondering what you think about those two

01:42:15.720 factors that are clearly stark differences between the world's best cyclist today and the world's best

01:42:22.660 cyclist 20 years ago as a way to bridge the gap between the use of drugs then versus not today.

01:42:29.680 There are also a couple of other interesting topics here as well, but I think, yeah, nutrition,

01:42:33.840 obviously, fuel in, speed out in the other end. It's crucial. And it doesn't help if you try to do

01:42:40.300 something with your VO2 max on the things if you're running out of fuel. It doesn't help to install a

01:42:44.000 larger engine in a car if your tank is. Yeah, exactly. I think one of the biggest differences is

01:42:50.340 exactly like you point out. Their attention to nutrition today in the World Tour, for example,

01:42:54.660 is in a completely different level than what it was a long time ago. I know also over the last

01:42:59.200 five years, these have also made a big, big change as well because it's large teams. So they're even

01:43:05.440 trying to devise tools and methods and other things that allows them to scale this a little

01:43:09.840 bit more. The benefit I have working with a couple of athletes is that we can go on a completely

01:43:14.260 different level and what they can do again. So we can even use metabolic measurements even out in

01:43:19.740 the field to do measurements basically just to see what is happening and become extremely detailed on

01:43:24.120 what we do. Even looking at basically post-exercise, even what's happening there, or pre-meals,

01:43:29.220 post-meals, what even happens with your resting metabolic rate because even this changes throughout

01:43:33.140 the day. And to keep up the consistency over time, you need to fuel accordingly.

01:43:37.440 Coming back to the question, and that is that I think one, nutrition, yes, this is probably one of the

01:43:41.740 main contributors. Of the two, this is the main contributor to why they are racing faster today

01:43:47.040 than what they did five, 10 years ago. Further, I would say that the watts per kilogram, this is a

01:43:51.860 place where I feel that we will see a change again. I think actually that the weight of athletes will

01:43:57.400 start to go up again. But the reason for why we start to see this going there is because if we look

01:44:02.400 at trying to understand why did we end up with the training programs we have done and so on, obviously

01:44:06.240 it's a very empirical approach to it, practical approach, which very often is extremely good

01:44:11.600 when it just gets enough time to evolve. But at the same time, also what haven't happened

01:44:16.140 is that we have adopted training strategies intact with the availability of information,

01:44:23.220 technologies, and all things that helps us do better fueling. Because one thing we also do know

01:44:28.140 is that in order for any growth to take place in the body, you need obviously oxygen. We already

01:44:32.600 talked about that, but we use oxygen as a proxy to understand metabolism. And metabolism, again,

01:44:36.620 is also a function of growth. When you do training over time, you're obviously trying to signal to

01:44:41.940 your body you need more muscles to be more efficient in one or the other way. And if you now start to

01:44:46.920 limit or you put a cap on fueling in order to drive down your weight, this will also start to impair

01:44:53.580 most likely ideal growth. Do we see this in triathletes? I mean, cycling is an interesting sport because

01:44:59.700 you don't really get a benefit from weight reduction if you're a time trialist, for example. In fact,

01:45:05.100 it's the opposite. It's watts more than watts per kilo that matters. But obviously, if you're in the

01:45:10.920 grand tours, all three of which are basically built around climbing, watts per kilo is ultimately the

01:45:17.600 metric. Now, we can debate whether you can get a more optimal watts per kilo at a slightly higher

01:45:21.940 weight than a slightly lower weight. But clearly, you're not going to see people winning a grand tour

01:45:27.140 at 80 kilos or 75 kilos. I think those days are probably gone. But in triathlon, how much of an effort

01:45:34.160 do these guys put into their weight? This is actually one of the interesting things because if you

01:45:39.360 talk about watts per kilogram, Christian actually is watts per kilogram have gone up with increasing

01:45:43.500 his weight. You said he's 80 kilos? 80 kilograms, yeah. And his FTP is? If I'm going to convert it to

01:45:49.320 FTP, it would be probably in the range of 400 and... If I did him on a 20-minute test, probably, let's say

01:45:56.940 for the sake of simplicity, 420 for something like this, probably if we tuned him. If that was how we did

01:46:02.680 testing and we started to standardize that, maybe even higher, I don't know. It's really hard to say

01:46:06.740 because we don't do FTP testing. Yep. What would be his average power in an Ironman for four hours?

01:46:13.760 So for four hours, typically there, they would be around... Let me see if I can guess. You said,

01:46:20.020 before we started the podcast, you mentioned they were holding 44 to 45 kph.

01:46:25.300 Or even slightly above, yeah, in the most extreme cases.

01:46:29.060 So he's got to be at 360 to 370 watts to do that?

01:46:35.440 Well, that's where aerodynamics come in and we are able to lower the numbers because obviously

01:46:39.400 staying at 360, 370 will have a quite big impact also on energy utilization and heat production as

01:46:44.320 well. And for example, coming also back to the question of basically using supplements to

01:46:48.420 try to mitigate pain or lower your temperature, we also know that this is not necessarily good.

01:46:52.840 But really also here's a question as I thought of that one, probably we're not going to continue

01:46:57.000 on that path now, but using, for example, supplements to try to lower your temperature.

01:47:01.900 I think this is where it's important to ask yourself first order principle,

01:47:04.900 where does that heat come from? The heat comes from basically mechanical power.

01:47:09.000 Is it lowering mechanical power? Because I think the argument of the acetaminophen literature

01:47:13.600 is that the body is bumping up against a couple of set points that are acting as governors

01:47:20.400 to output. One of them is pain. One of them is temperature. The body does not want to let you

01:47:25.680 get too hot and obviously doesn't want you to tolerate too much pain. Acetaminophen potentially

01:47:30.740 blunts both of those. You're arguing, yeah, but if part of the way it's blunting energy output

01:47:36.520 or temperature is by reducing mechanical work, gross mechanical work, then you're going to probably

01:47:43.280 pay a net mechanical work price. Yes. And also a capacity price, you could say in some regards.

01:47:50.020 I have no idea what the answer is. I would just want to test it in training. Literally alternate

01:47:55.160 weeks with, without as a placebo test with acetaminophen, without acetaminophen, with acetaminophen,

01:48:00.060 without. Because again, at your level, you get a 1% difference. It matters.

01:48:05.560 This is the wonder of the body. One thing that happens, obviously, if you start to use a lot of

01:48:09.020 carbide, so you go to 160 grams. We even cropped in at the highest numbers. We measured it's in

01:48:13.980 excess of 240 grams per hour of carbohydrate utilization. That is so... Can you tell people

01:48:20.360 what 240 grams of carbohydrates looks like if it were food? Do it in pasta or something like that.

01:48:26.640 How much pasta is 240 grams of carbohydrate? To make it simple, it would be probably as you took

01:48:33.440 pure gel and you fill this glass and you drank it. Take those pure little nasty, disgusting energy 0.91

01:48:40.460 gels and fill your 16 ounce glass there with it. And drink it. And what form are they consuming 0.92

01:48:47.680 the carbohydrate in? Normally they would do it in the form of drink mix. So how do they get that much

01:48:52.280 liquid in their body? That's a part of it. Normally they would consume around, I would say around minimum

01:48:58.600 of 1.4 liters, but say around two, a little bit more than two liters per hour. Good Lord.

01:49:03.440 What's the glucose concentration in that liquid? In that one, then I have to calculate.

01:49:08.280 So we could just do the math. If you said two liters, 200, that's 12%. That's a 12% mixture.

01:49:13.400 Yeah. Approximately. Probably, yeah. Yeah. 0.674.

01:49:16.280 Which again, that also completely flies in the face of traditional nutrition science, which says

01:49:21.760 five to 6% is the limit for gastric tolerability and is the sweet spot for absorption. At five to 6%,

01:49:31.700 your max. Now again, that's a different problem. That's optimizing for water absorption into the

01:49:36.920 cell, not your primary point. You're probably overdoing it on liquid and you're trying to

01:49:42.120 maximize glucose into the cell. But also absorption here, because this is also where it's interesting

01:49:47.320 or important and to go back to also a little bit like, what is this research? Like who is this

01:49:51.320 research done on as well? Yeah. And it's not done on these trained athletes.

01:49:53.820 Yeah. Okay. First question, how long did it take you to train the athletes to be able to tolerate

01:50:01.320 that? Because if someone's listening to us today and they're saying, oh, I heard these guys talking

01:50:05.400 about the 12% carbohydrate concentration, which again, just means for people listening, that means

01:50:11.100 120 grams of glucose per liter. That's twice the standard what you'd see in an energy drink.

01:50:18.120 And then I'm going to drink two liters of that every hour. I think within two hours,

01:50:22.760 a normal person is going to be puking their guts out. They simply can't get that volume of glucose

01:50:28.340 out of the upper gastrointestinal tract. So is this just like any other muscle where you can train

01:50:35.240 yourself to exceed capacity? Seems like it. Seems like most things in our body is actually extremely

01:50:40.400 trainable. That's also comes back to the nutrition part as well, where exactly also the training

01:50:45.140 programs we have today have not seen the same change as nutrition strategies have done over

01:50:50.320 the last years as well. So that means also that the power numbers that was output maybe

01:50:54.400 five, 10, 20 years ago, I'm not so sure if that's the limitation we're going to see.

01:50:59.780 Especially late in races. This is where it's going to make a bigger difference. It's not going to make

01:51:03.680 a big difference at the beginning of the race, but it could make a huge difference at the end.

01:51:07.580 Talk to me about formulations. Back when I was swimming great distances, one of the challenges I had

01:51:12.200 over a 12 hour event was literally fatigue of the same flavor. So part of the challenge was how do

01:51:18.580 you just mix things up? And after a while, sweet becomes horrible and you actually want something

01:51:23.680 salty, but then salty becomes horrible. Like how are you managing the actual practical implication of

01:51:29.240 this? So again, this is a place where more than they made actually their gels to be as neutral as

01:51:35.340 possible. And because it is formed as a hydrogel, it actually encapsulates the sugar inside the

01:51:40.920 hydrogel, meaning that the perception of sweetness is much lower than for any other gels. You will

01:51:46.380 still perceive it as slightly sweet, but it's much, much less.

01:51:49.700 Wait, sorry, this Morton gel is a significant driver of their calories as well as the bicarb?

01:51:54.600 No. So there are two different products because Morton has the bicarb product,

01:51:58.020 which actually also is encapsulated in hydrogel, but actually with carbohydrates.

01:52:01.780 But then they have the dedicated fueling products, which is basically consist of a drink mix,

01:52:06.540 which Christian and Gustav are mainly using, but they also have gels as well. So it's more

01:52:11.120 practical, obviously, to have a gel. Silly question. The bicarb capsule, 0.97

01:52:15.060 they're swallowing a capsule while they're on a bike drinking?

01:52:17.840 This is a two-component, or actually a three-component mixture. You have the hydrogel,

01:52:23.400 which is basically in a dry format in a sachet, and then you have the bicarb tablets, and then you

01:52:28.500 basically you take water. So you add water to this. First, you add water to the hydrogel mix,

01:52:32.500 and then basically you are forming the hydrogel. And then you take the bicarb and you mix it into

01:52:36.520 the hydrogel. And in this way, you wisp it around. And then basically this way, you're able to bring

01:52:41.680 it into the stomach and actually able to deliver it without it actually causes any problems.

01:52:46.840 I can't say actually too much in detail what kind of concentration we are on at the moment.

01:52:51.600 I will come back to this a little bit later because this is probably things that we will publish.

01:52:55.640 But to put it this way, 19 grams of bicarb, for example, that would be like a standard dosage.

01:53:01.480 You can go to 22 grams of bicarb as pure, and you can take this. And this doesn't require very much

01:53:06.940 training. What is the base of the carbohydrate in that brand that they seem to like? Is it straight

01:53:11.860 dextrose, multidextrin? No, mainly fructose and glucose.

01:53:16.860 What's the ratio? It's a little bit of a time since basically we went through

01:53:20.920 details of the law, let's say looking at the balances between this, but let's say it's a 40-60.

01:53:25.200 So it's basically high fructose corn syrup. Yeah. The genius part of it is actually not

01:53:29.220 the carbohydrates itself. It's the packaging mechanism.

01:53:32.460 Yes. Exactly. That's the difference. So if you were consuming 60 grams per hour,

01:53:36.740 spare the money and buy orange juice or even put honey inside and drink it. Where it differs itself

01:53:42.540 is when you are really starting to push the limit of the concentration. But you can even train yourself

01:53:46.620 today. This is something we know. You can even take honey today and you can mix it and you can train

01:53:50.440 yourself to go to higher values than what have previously been published, like pure normal products.

01:53:54.620 If you want to get to 160, you need the technology. Yes, exactly.

01:53:58.700 No fat and no protein, any amino acids throughout that whole race?

01:54:01.900 No. This is something that was also done research on, but we ended up in the end,

01:54:06.700 pure glucose and fructose. And it actually has to do with actually how oxygen is being prioritized in

01:54:11.160 the body. Because we have to consider two things here. It's easy to think that while you're only

01:54:15.520 racing at 80% of your VO2 max or 70 or 90, whatever percentage it is of your VO2 max. But that's a

01:54:21.980 bit of a short circuit. Because what we have to think of is that this oxygen, more oxygen is going

01:54:26.700 to contribute to more heat as well. So it's not like you can just go, whatever, I'm rising at a

01:54:30.840 solo percentage of a VO2 max, so I can go whatever and do in the beginning. Because it always will

01:54:35.480 accumulate towards the end of your race. That means already from the beginning, it's critical how you

01:54:40.280 pace and your efficiency as well. Because every inefficiency you have, either it's biochemical or anything

01:54:45.840 like this, it will basically end up towards the end of the race. So this means also that at the

01:54:51.000 moment you start putting any other substrates or nutrients into your mixture that the body have

01:54:57.660 to prioritize somehow in the system, that means also that there are going to be less oxygen available

01:55:03.400 for pure propulsion. And in the end, it's a pure propulsion that really sets the winner apart

01:55:08.100 from the rest of the people there. So you basically want to peel away absolutely everything that doesn't

01:55:13.360 contribute to forward propulsion. Use every milliliter, every mole of oxygen purely for that purpose and

01:55:19.400 nothing else. Because oxygen is one of the limitations. We talk about different kinds of

01:55:24.360 substrates and these kind of things, which is still also a place where we don't know. We are actually

01:55:28.780 doing quite some interesting research on different kinds of substrates from glucose, fructose.

01:55:33.300 Have you looked at BHP, beta-hydroxybutyrate?

01:55:35.900 Yes. So we looked at BHP as well. And one of the things, if you purely look at the stoichiometry,

01:55:41.660 there are interesting things here. But the problem is that we cannot only do stoichiometry.

01:55:45.320 We actually need to know the enthalpy, and we also need to know the Gibbs-free energy that is

01:55:49.980 available, and ideally even the entropy. Okay. So the Gibbs-free energy, if you're making the BHP

01:55:55.600 yourself, much more complicated. But if you take it purely exogenously and consider it an additional

01:56:01.780 substrate above and beyond glucose, do we have reason to believe that we're going to get more ATP

01:56:07.280 per mole of oxygen? No, because the practical limitation of this is that when you do racing,

01:56:12.660 if you think of this more first order principle again. So if you do racing, and this was actually

01:56:16.520 one of my first questions, because we have been working together with one of the leading brands,

01:56:21.060 or arguably the leading one, they are situated here in the US. But one of the limitations that

01:56:26.720 you have to look at is basically when you are racing, this has a certain fuel demand. So there's

01:56:30.240 a certain amount of fuel that needs to go through the system there. And that means that in the context

01:56:33.940 when we talk about glucose and fructose, we are looking at basically trying to replace as much

01:56:37.780 as possible. We said, okay, glucose, fructose, or basically a carbohydrate seems to be a very,

01:56:41.640 very oxygen-efficient fuel source when you do movement. But that means also that in order not

01:56:47.980 to run out of this, we are trying to replace as much as possible or this to be able to race faster.

01:56:52.420 Because if you said, this is the fuel source I have available in my body, I'm going to race this

01:56:56.260 fast. Obviously, I cannot go faster than this. So certain power number, certain duration will

01:57:00.920 deplete this energy source or basically bring me close to depletion of that source. So that means,

01:57:05.560 okay, instead of then switching to other sources, or you're basically always using other sources as well,

01:57:09.260 but trying to get more from that source means basically you have to replace more of the power.

01:57:13.540 So more of the power now, we have to create an environment where more power can come from the

01:57:18.020 same source. Let's say you were originally around at 80 grams or 90 grams, let's say 80 grams for

01:57:24.700 simplicity, say 80 grams of carbohydrates, and you have a biochemical efficiency of 20%, for example,

01:57:29.680 that means basically you're getting out, or if you say 4.2, let's say you can get basically,

01:57:33.900 so one gram would be then the equivalent of, let's say, one joule per second or one watt,

01:57:37.720 more or less like this. So you could get from that one. So if you double this from 80 to 160 now,

01:57:42.200 that means basically now you can get double the amount of what's coming from carbohydrates here.

01:57:46.680 The problem with ketones is that the strategy is not to basically fuel the amount you are using in

01:57:52.780 a race. You don't fuel with ketones in the same way. Like if you go to 70, 80, or even 100 milliliters

01:57:58.900 of ketone consumption in an hour, you're going to have so high levels of ketones in your body

01:58:04.680 that you will start to feel almost like you are getting diabetic. There's a feeling of bonking

01:58:08.900 almost. So the thing that is there is more like it's inducing a state in the body rather than you

01:58:13.880 actually fueling, using it. Because in order to fuel, so they said your idea was that you're going

01:58:18.880 to replace what you're losing from your body, because that's driving your body also out of

01:58:22.860 homeostasis as well. You are going to try to keep homeostasis as much as possible, or try to keep the

01:58:28.280 state of the body as unchanged as possible. That would mean that you would need to ingest so large

01:58:33.240 amount of ketones during a race. And then you have also the issue with the salt, if you're bringing

01:58:37.500 it in with a salt versus an ester. Okay, a couple of things I want to chat about. So one, we talked

01:58:41.520 a little bit about this before, but let's kind of go back to it. So remind me, did Christian and

01:58:45.960 Gustav both compete in Tokyo Olympics? Yeah, both did compete in Tokyo Olympics, yes.

01:58:50.460 And how did they do in Tokyo? Christian won. Gustav came seventh, eighth, seventh place, yeah.

01:58:57.240 After that Olympics, they went back to Ironman distance for the next three years.

01:59:01.360 And then Gustav did not compete in Paris. He had an injury. Christian did, and I think you said he

01:59:06.780 was 12th. Yes. How much of that was the mismatch in distance? Again, he's much more optimized for

01:59:13.820 Ironman, Olympic distances. That's like asking a marathon runner to go and run a 5K. Very unusual.

01:59:21.560 What was the process like to get ready for that? And what was the difference in his personal

01:59:26.660 performance between 2024 and 2021? We went there for two reasons. One, we thought it was possible

01:59:33.180 to get back on the podium. And I would say even after disappointing wrestles, I would say that we

01:59:38.220 are more convinced today that it's possible to go back than we were. But there are other reasons for

01:59:42.640 that, coming back to that. Sorry, meaning in Los Angeles, you're saying? In Paris, yeah, even there.

01:59:47.080 So if we then look also back to there, Christian already had in training put out far better

01:59:53.540 performances than what he was capable of getting out on race day. I would say that there are two

01:59:58.460 things that looking back at, let's say, the process leading into Paris that for me explains most of

02:00:04.980 this. And one part of it is one of the differences leading from Rio, leading into Tokyo, was that we took

02:00:11.540 a massive shift in how we did intensity control. And again, now we're talking about intensity control,

02:00:17.860 that doesn't mean necessarily about talking about just threshold or anything like that. This means

02:00:21.780 exactly like pinpointing different intensities and working on them in order to optimize the human

02:00:26.360 body. And then how we find this. One thing that we did, actually, we got maybe a little bit too

02:00:31.420 complacent the last year. And that is basically, we've done it so much that it's more like, okay,

02:00:35.440 we had more projects. So it's also one place you just think that, okay, you are getting really good at

02:00:39.720 controlling the intensity and these kind of things. So we stressed it a little bit. The key

02:00:44.340 that really made a difference between Rio to Tokyo, we became sloppy at.

02:00:50.600 Rio to Tokyo or Tokyo to Paris?

02:00:53.060 Yeah, from Rio to Tokyo, that was the massive shift going from unknown to basically Olympic champion

02:00:57.640 to basically where we, the last year, I would say in particular, let's say the two last year,

02:01:03.680 were much more sloppy on simply because you just think that, okay, after you've done intensity control

02:01:07.420 for so long time, you would think that you have such a good calibrated body for this.

02:01:11.640 We made the same error actually already in 2017 to 2018, where 2016 to 17, we saw massive

02:01:18.180 improvements in performance. From 17 to 18, we have done so much measurement. We lacked it and

02:01:22.480 other things like this. Just thinking, okay, they must have a good feeling for this, really good

02:01:25.760 calibrated. To a couple of months, we saw that, okay, this doesn't work. We have to go back to the

02:01:29.660 practice. The same error we did the last year going into Paris. Purely in terms of performance,

02:01:34.920 Christian basically put out performances that was far exceeding what he did on the run in Paris. In

02:01:41.000 Paris, he lacked around four minutes per kilometer over the 10K. That was the difference between the

02:01:46.440 first place and the 12th place. Sorry, say that again?

02:01:48.860 Sorry, four seconds. Okay.

02:01:50.240 Four seconds per kilometer. Four minutes would be between them and me.

02:01:53.720 Yeah. Yeah. So yeah, so four seconds per kilometer.

02:01:57.240 So 40 seconds he lost on the run.

02:01:58.940 Yeah, around there. But one of the differences that also were between Tokyo and Paris was that

02:02:04.500 Christian actually sat there. When he came out of the water, Christian's swimming have never been

02:02:08.720 his forte. That's been the worst discipline for him. And the same actually made, this was also,

02:02:14.380 we are an arena where we would see even bigger differences. So the poor swimmers in Paris,

02:02:19.620 some of them even didn't finish.

02:02:21.140 A lot of people heard about how disgusting the river was. How did that create more separation

02:02:26.180 between the excellent and the good swimmers? It's the current in the water. There was a

02:02:30.460 massive current in the water, basically to the extent that if you said it was still water,

02:02:34.880 most of the triathletes basically set 400 meter record, world records swimming the first 400

02:02:39.160 meters. Usually that closes the gap between people when the current is strong.

02:02:43.440 The only problem is that you have to turn back up. And then the problem also with this,

02:02:47.080 because this is in a river, you can normally say that if this were basically a place where you just

02:02:50.980 knew you had laminar flow. So you had laminar flow and it was homogenous all over the place.

02:02:55.260 Then you could say that really it would make that much of a difference or should not. A little bit

02:02:59.780 it would do, simply because the swim in distance is the same, but the duration becomes slightly

02:03:04.520 longer. And if it's a slightly longer duration, you open up a little bit more of a gap.

02:03:07.980 The challenging part with Paris and the Seine was also that flow is not, especially in the Seine,

02:03:13.680 not homogenous. It is highly heterogeneous. Meaning basically, if you look at the swimmers there and

02:03:19.680 you see that they basically come around the first buoy and they are aiming for the buoy next to them,

02:03:24.140 suddenly people go in a large parabola because they underestimate actually how much more

02:03:29.080 up current they have to basically point in order to get around. And then after that as well,

02:03:33.760 when they get closer, because there are two bridges that went across the river as well.

02:03:37.800 And around this one, you would normally think that the current is actually slower next to the

02:03:45.060 pillars that are into the water. The fastest swimmer that was suddenly in the front there,

02:03:48.780 suddenly became second or even down into the group simply because the people that went a couple

02:03:54.160 of meters, two meters further to the right there, swam suddenly past them. The fastest swimmer

02:03:59.420 suddenly, it looked almost like they were, came, had good speed and suddenly standing still almost,

02:04:04.080 where they had to deviate out from the pillar and basically, so it becomes a far more tactical race.

02:04:09.620 How far did he come out in the water after the leader in swimming?

02:04:13.920 I don't remember. That was quite significant. That was, to be honest, the only thing I remember

02:04:19.000 from that is that he came out, I think, let's say of that long, because also what happens here,

02:04:23.900 instead of that you get a pack that is swimming, it gets a long, long line instead. And that's also

02:04:28.740 challenging in this context as well, because you get a completely stretched out field instead of that

02:04:33.640 people come more like in patches out of the water. I think Christian came out from the last guy in 0.98

02:04:39.560 the first group. He was, I think, 17, 20 seconds behind him, but out on the bike, he became,

02:04:46.300 I think, 40 seconds down from the lead group. One of the differences that we already anticipated

02:04:53.580 from the first year to the second year also was that one thing that I think people got a little

02:04:58.220 bit shocked over the first year, so that was basically during the test event, was how quickly

02:05:02.820 in that course the groups behind basically catched up with the lead groups on the bike.

02:05:08.840 Like, this one thing we knew that basically the people that obviously were in the first group,

02:05:12.840 they would benefit everything from basically going a little bit harder,

02:05:15.720 and maybe even starting devising strategies where they even got domestics. Like, this is an

02:05:19.960 individual race, but people actually starting getting domestics and sacrifice people in the

02:05:24.280 races from a nation perspective to be able to keep up, let's say, a different racing strategy.

02:05:29.100 Because they're allowed to draft in triathlon now?

02:05:31.160 Yeah. So in short course triathlon, you can draft. There you can draft. This is the difference

02:05:34.660 between that and long course like Ironman racing. What did this mean for Christian? It meant for

02:05:38.960 Christian that basically in Paris versus Tokyo, he had to pedal 30% harder for three-fifths of the race

02:05:48.540 in order to just come back in the group, and you don't get 30% for free for basically 25 kilometers.

02:05:54.500 30% extra power for 25 kilometers, that has a massive cost. So even though we could say that his fitness

02:06:01.820 have increased beyond what it was before, which made it confident that if this had been a pure time trial,

02:06:07.580 in let's say where you had like a normal swimming conditions, this kind of thing,

02:06:10.540 he would be faster than what he was in Tokyo. But because this is a place where obviously the dynamics

02:06:15.660 is completely different, that's extra fitness. 30%. I mean, I'm surprised he could even do that.

02:06:20.920 Yeah. That has an implication when you get out on the run. Because obviously here,

02:06:24.340 this is not a time like you where you go all out on a bike.

02:06:26.560 What was his 10K time?

02:06:28.840 30 high something, I think. Let's say around 30 minutes. I have to go back and check.

02:06:33.980 Sometime I'll head a little bit different place.

02:06:35.820 Still so fast. Good Lord.

02:06:37.700 Yeah.

02:06:38.460 Last thing I want to just chat with you about is where you are using AI today,

02:06:43.600 or where you see AI going to make your insights better and your efficacy better.

02:06:50.600 So with all the data we collect, one of the things that I would be able to discover quicker

02:06:57.340 with AI is that his utilization went up too high in the month leading into the Olympics.

02:07:04.940 Utilization.

02:07:05.700 So meaning basically where you are trading away a little bit on your anaerobic capacity and power,

02:07:11.160 or let's say at least anaerobic power, maybe not capacity, you don't want to trade that away,

02:07:14.560 but let's say anaerobic power, in order to increase your aerobic power slightly more.

02:07:19.160 So let's say off your view to max, you're able to raise sustainably an even higher percentage

02:07:23.140 off your view to max. So that means also substitute utilization would normally then be improved as

02:07:28.100 well. Whether that would make a difference, to be honest, I don't know because of how the race

02:07:31.920 played out. That's one. Where AI really makes a difference is that for me, we have used AI for

02:07:38.440 some while. We started developing our own systems. I think this must have been back in 2020. And that

02:07:45.700 is that with all the data, because of all the in-depth research we have done, we obviously have a lot of

02:07:51.180 data and interesting findings where we can build AI systems that allows us to take new data that comes

02:07:58.820 into the systems in real time and help us highlight what is important and what is not important. Like

02:08:04.600 if there are things that we need to prioritize more, or I can even put attention to this.

02:08:09.040 So AI for me is also where I think research, just purely in terms of longevity, in terms of human

02:08:14.860 health, everything will really supercharge us to a completely different level. Researchers and other

02:08:21.100 people that I work with today, and one of the main limitations when they come in and they basically see

02:08:24.660 the work we do, is to say, okay, the research we do is insane. It has a depth to it that basically

02:08:29.840 is unmatched. But it's not because they don't have necessarily the competence in a group or anything

02:08:35.800 like this to do it, but it's just the sheer amount of work that is required in order to utilize it,

02:08:41.240 or even use it in a study in itself. And when you have AI, like a lot of work that actually is done

02:08:47.240 in research today is still manual work. There are plotting that is still done manual, there are

02:08:51.620 measurements that are still done manual, but where instrumentation AI really can change this to the

02:08:57.180 point where I'm not talking about one study can be done with maybe one more instrument or something

02:09:02.940 like this, but where you can use a multitude of extra of instruments and be able to dig into the data

02:09:08.700 on completely different levels than what you could do before. AI will not replace the decision-making,

02:09:14.480 but it will help us in terms of information.

02:09:16.900 Basically extract patterns you might not otherwise be able to direct your own statistical analysis

02:09:23.680 towards? Yes. You can also imagine here, AI, how we train today is that for a normal person,

02:09:29.440 I don't have a PhD in mathematics, I don't have a PhD in machine learning, but the good thing is that

02:09:34.960 we can basically take... You're using large language models to do this?

02:09:38.620 No, we have built an agentic system. So large language models is basically the interface for us.

02:09:43.280 So I would use that, for example, I would say that I want to look at data in a certain way,

02:09:48.140 so I'm using more the LLM or NLP to inform which agents that should be utilized in order to understand

02:09:56.260 or to dig into the data. So I still have to produce the question. I still have to produce the question.

02:10:01.940 AI for me is basically just an autonomous system. I know this is a big debate on how intelligent these

02:10:06.700 systems are, but one of the best cases is that because I work so much in edge cases, so there was

02:10:12.620 a study that just came out now and people say, oh, it seems like AI has the possibility. It came out of

02:10:17.160 Stanford, I think it was, and it was quite a lot of research involved in it. And I say that AI has the

02:10:21.740 possibility to seemingly to produce quite a lot of novel ideas on par with humanity. But where I would

02:10:27.700 say that, well, it depends on what kind of perspective you look at it. And also where I would say that

02:10:32.820 if it doesn't have data that it's been trained on for this, the best thing it can try to do is

02:10:37.600 interpolate between data it already have access to. That means edge cases or basically cases where

02:10:42.760 you are in the, let's say in the domain where you need to extrapolate, they greatly suffer. This you

02:10:46.740 can see again and again and again. But the good thing is that you empower a group of people or even

02:10:52.680 a single human with PhD level capabilities in mathematics, in biology, physics, all these kinds of

02:10:59.560 things. So you use the NLP, you ask a question, and basically it can target this kind of data.

02:11:04.760 Like even programming is a limitation for many people today. They are able maybe to formulate

02:11:09.580 great questions, but they don't have the capability or the resources available to really put or execute

02:11:15.560 and look into it. But when you have AI, suddenly now you can, without necessarily skills and programming

02:11:20.900 other things, you're able to formulate the question and you can ask the questions into the data.

02:11:25.080 And these different agents can now suddenly employ extreme level competence from multi-domains by the

02:11:33.060 hands of a single user into the data and give you insight back that otherwise would be impossible,

02:11:38.900 even for large research teams. As you think about 2028, if you think about the next four years,

02:11:44.580 if you had to just sort of speculate wildly, what percentage improvement do you think this will

02:11:50.100 bring, not to the insights, but to the actual performance?

02:11:55.080 I think that, okay, first to answer it a little bit more high level, I think that today Norway have

02:12:01.620 been a superpower in winter sports. A lot of the reasons has been because of the technology research

02:12:06.940 and other things that have gone into, for example, purely optimization of skis. You could come to the

02:12:11.520 Olympics, you can be a better trained athlete, but you want to win simply because the way the

02:12:15.980 technology makes a difference on the skis, for example. I think AI will be, or I'm convinced AI

02:12:21.600 and the systems we are building now will have the same implication on purely human performance and

02:12:26.820 training. Meaning that if you basically don't employ AI at some point, you will be at such a

02:12:31.660 deficiency. So you have great people. What is important for me to say here, coaches are very

02:12:36.420 often extremely advanced, almost super intelligent in the space of exactly making people fitter.

02:12:43.000 They are maybe not, or not even closely remote to the understanding of everything we have talked

02:12:49.200 about today in terms of even knowing exactly the definition of such a widely common expression

02:12:55.140 like, or terminology like FTP, but still they produce some excellent athletes because they have

02:13:00.320 observational skills and developed an intuition of what works and not works through a lot of

02:13:05.320 experience, empathy and so on, that they are brilliant. But AI is not conflict to them, rather again,

02:13:12.600 a superpower where they can employ their level of understanding into much better questions

02:13:19.080 into what is happening here with my athlete. So it shouldn't change the way you do coaching,

02:13:24.340 but it increases the precision of what you do in the coaching. And the biggest change thus in

02:13:30.920 performance will come through even better consistency in the training. Not that single workout that is so

02:13:36.580 much more brilliant, but purely in the consistency of the training and adaptations you do to the training

02:13:41.740 programs. Is there an area in particular that you see the biggest gap between where your insights are,

02:13:49.020 your coaching insights are today versus where you hope AI is going to close a gap?

02:13:54.820 Yeah, being able to be far more proactive than what I am today. I still, even though I would consider

02:14:00.160 myself as one of the leading persons on using technology and also in applied research when it comes to

02:14:06.740 elite performance, I can still say that I am only able to utilize in the daily life a fraction of

02:14:12.700 the data that we are collecting. Then you can say, well, why do you collect all of the data? Well,

02:14:16.600 very often it becomes insight. You come back, you learn something from it that you otherwise wouldn't

02:14:20.500 be able to do. Well, and you're going to have training data.

02:14:22.240 Yes.

02:14:22.660 Great training data.

02:14:23.280 Yeah. Yeah. But being able to be proactive and make adjustments, individual adjustments much more

02:14:28.660 proactively will be one of the biggest differences. And this doesn't go only apply to elites. This

02:14:33.900 applies even to amateurs or normal people. Well, it's been a fascinating discussion. Thank you again

02:14:40.780 for making the time to swing by on the way to Flagstaff. And as always, just enjoy talking about this

02:14:46.120 stuff. And hopefully folks enjoyed this as well, even though admittedly it was a little technical at

02:14:50.280 times. Thank you the same. Thank you for listening to this week's episode of The Drive. Head over to

02:14:57.520 peteratiamd.com forward slash show notes. If you want to dig deeper into this episode, you can also

02:15:05.200 find me on YouTube, Instagram, and Twitter, all with the handle peteratiamd. You can also leave us

02:15:11.220 review on Apple podcasts or whatever podcast player you use. This podcast is for general informational

02:15:17.780 purposes only and does not constitute the practice of medicine, nursing, or other professional healthcare

02:15:22.760 services, including the giving of medical advice. No doctor patient relationship is formed. The use of

02:15:29.220 this information and the materials linked to this podcast is at the user's own risk. The content on this

02:15:35.460 podcast is not intended to be a substitute for professional medical advice, diagnosis, or treatment.

02:15:40.740 Users should not disregard or delay in obtaining medical advice from any medical condition they have,

02:15:45.860 and they should seek the assistance of their healthcare professionals for any such conditions.

02:15:50.960 Finally, I take all conflicts of interest very seriously. For all of my disclosures and the companies

02:15:56.400 I invest in or advise, please visit peteratiamd.com forward slash about where I keep an up-to-date

02:16:04.040 and active list of all disclosures.

The Peter Attia Drive - January 13, 2025

#331 ‒ Optimizing endurance performance： metrics, nutrition, lactate, and more insights from elite performers ｜ Olav Aleksander Bu (Pt. 2)

Episode Stats

Summary

Transcript