Magic VLaMax training

I recently came across Aerotune and the training philosophy of Björn Kafka. He has a german podcast which I listened to. I would call him a VLaMax-guy. The way he describes his principles are indeed very appealing.
He makes clear statements (and promises) about riding at Fatmax, lower or increase your VLaMax by riding in certain zones. And he also claims that he can influence the VLaMax value in a short time and can change a riders profile. And he can also change the fat/glycogen utilization by riding in certain zones (this is in a strong contrast to the explanation of e.g. Kolie Moore if I get it right).
He offers a test software which delivers values as VO2max, vlamax, fat and glycogen utilization only with heart rate and powermeter.
All this sounds like some magic training composition which makes me sceptical.

Björn is a successful coach of a world champion (e.g. Andreas Seewald - XCM) and several successful pros. He must have quite success with his training principles.

Nevertheless I know that there is a discussion if VLaMax is really a thing (I haven‘t understood the point to be honest - maybe somebody can jump in and explain it).

If his statements about the physiological effects don‘t hold true, I question if it just all comes down to the „all roads lead to Rome/Tokyo“ principle?

I also have to think about the Dr. Coggan statement that there is no switch between zones - training is training and you have to manage the trade-off of duration, intensity and recovery.

If I would do such a test, get a VLaMax value, do the prescribed training, would I really see a change in my values? How can this system hold up based on a doubtful vlamax-value work?

There is no magic training.

Training, in general, improves the thing you’re training. If you ride a lot, your endurance will get better. If you ride very hard (vo2max), your heart will get stronger. If you want to sprint fast, you need to bulk up. And so on. As everything in the body is connected, most training will also have side effects in improving other things, for example, if you ride a lot, your vo2max might get better too, even if you mostly ride easy.

The other thing is how you measure improvement. For example, you could compare two rides by how long you rode, how far, how many meters climbed, how many calories burned, how hungry you are, or how tired you are. You will get probably a good idea of how hard a ride was by using any of these measures, as long as you consistantely use the same measure. But some measures might be more appropriate than others to compare specific training approaches. For example, how much your legs hurt might be a useful measure after sprint training, but how many meters you climbed may not.

VLaMax is just another way to see what your training is doing. The problem with it is that a), its hard to measure (you need to take blood samples, following a specific protocol), and b), you measure something that is a just a side effect of something you are actually interested in. Its a bit like determining dehydration by the colour of your pee. Its sort of gives you the right answer, but you don’t actually measure dehydration.

The aerotune software does some sort of full circle - it estimates VLa from power, and then estimates other power zones (like fatmax) from VLa. (At least thats how I think it works, its been a while since I looked into it.

I think there a probably some things where the VLaMax approach is useful. My guess is, especially with pro cyclists, it helps estimating some things that are otherwise hard to measure, because they are all so very fit.

But yeah, to summarise again: there is no magic training. Most (reasonable) training philosophies will help you, but some might be a better match for some people than others.


I’m still not entirely convinced VLaMax is useful even after I’ve measured mine several times both running and cycling. There are numerous caveats in how you perform the sprint test alone and trying to calculate VLaMax from 1hz data is not ideal. It’s a fun topic to explore though. “The Science of Winning” by Jan Olbrecht is a good treatment of training around these measurements.

In my experience, I can use my change in W’ as a proxy for change in VLaMax.

These days if I’m curious about what my VLaMax is, I’ll use my measured VO2max and mlss to back calculate VLaMax. This person has been generous enough to share their work.

Here’s some background in VLaMax:


I commend you on your skepticism. It is well-placed.


I’ve enjoyed the podcasts I’ve heard with Bjorn Kafka. He sounds like a great coach.

After hearing Kolie Moore, @The_Cog, Mark Burnley, and Mikael Eriksson basically say that it’s BS, I would tend to believe them. (Note that some of the opinions are much stronger than others. It seems that a lot of people in the industry don’t want to call out Inscyd publically.)

I suspect that using one of these software systems (Aerotune, Inscyd), you are just getting another form of power profiling and modeling. Also, one could just look at a power duration curve and instantly tell if the rider is a diesel (low vlamax) or a sprinter (high vlamax). Thus, if you are diesel, do some training that will improve short power. If you are a natural sprinter and want to be a better endurance athlete, then do more endurance training.


I did two Incsyd tests about four years ago. My VLaMax was already quite low (phenotype? :man_shrugging: who knows, who cares). Despite that, I was still recommended a training program of endurance and tempo. They didn’t call it that. It was called “FatMax” and “Medio”, as you know. My initial impressions were extremely positive as I had been on a steady diet of too much intensity, misguided application of “flavor of the week” training ideas (even valid ones), one-size-fits-all online training programs, and no way to figure out who was full of shit. Remember, even the non Inscyd guys want to charge you money. Everybody wants your money. My other experience at the time, someone using WKO and regurgitating some lines from Cusick videos was also unhelpful. (of course, I could have just gone to Cusick, LOL…that would very likely have worked…but, long story). That was generally the state of affairs at the time.

Fortunately, I found a coach about the same time who debunked (privately) much of the hype around that concept, clarified why the training might work anyway, and helped me reconcile how that could be the case.

Looking back, it was the good coaching that fixed me, even within an imperfect system.

What did I learn: Magic VLaMax training = tempo + endurance. That will be $180, please. I accept venmo and you don’t even have to do the dumb sprint test. LOL…just kidding.



Great answers - very insightful. Thanks everybody!

My takes:

  • VLaMax is hard to determine outside the lab

  • It is rather an indirect marker (I like the pee color analogy :slight_smile: )

  • They try to build a system beside the „FTP system“ and proactively say that their system is better and does not rely on FTP

  • High VLaMax → sprinter, low VLaMax → Diesel - the same can be determined by the power duration curve

  • They use different terminology with doubtful descriptions of physiology

  • In the end they prescribe just „regular“ training

  • Anything else to debunk this topic?


Hey - just wanted to jump back onto this topic as I think there are a few wrong ideas around about VLamax on the one hand and INSCYD and what it does on the other. Important disclaimer upfront: I am a coach and use INSCYD as one of the tools to work with the athletes I coach.

Is VLamax the “secret sauce” or the “magic” that it is made out to be in some circles, certainly outside of high-performance sports? Probably not.

Are the insights gained from INSCYD (I don’t know the competitor software) a useful tool in your training arsenal, considering the cost ($200-500/test, depending on where you live and the protocol)? I think so and there are a number of reasons for that.

Through VLamax (whatever the usefulness of the knowledge of the actual value is beyond phenotyping) you can cross the proverbial bridge from the aerobic (VO2max) to the anaerobic and metabolic (VLamax), specifically via the glycolytic metabolism. This allows INSCYD to translate your power profile to metabolic requirements, specifically carb consumption (via lactate) vs fat consumption (via body composition). For example, the “FatMax zone” mentioned earlier in the thread is nothing else than the power levels where the energy contribution from fat (in calories) is the highest. It’s not a training zone per se (though some coaches use it).

INSCYD does not prescribe any training zones or training plans or a new way to measure FTP. In that sense, all it does is map power profile to metabolic requirements. That’s literally all it does. Of course, through VLamax and VO2max, MLSS (anaerobic threshold) follows deterministically. We can debate wether MLSS is better or worse of a measure than FTP and it may or may not be interesting for you, but to me personally it was good to know that after 5 years of very heavy sprint/anaerobic/vo2max training combined with very low intensity training (avoiding the term “polarized”, not to add even more fuel to the fire), my MLSS and FTP (via TR ramp test and/or AI detection) have deviated by ca 40W. This is important to me, as I have changed my own athletic goals for this year back towards more of a low VLamax/endurance type athlete (current VLamax 0.7). Without that knowledge, I would have continued on, thinking I’m a 330W guy for a 5-6 hour race, whereas in reality I’m more like a 290W kinda guy. No FTP test would have given me that info (well apart from a 1h FTP-like effort, urgh).

I believe the actual energy requirements that INSCYD provides, specific to your own physiology, for certain types of training and/or racing are very useful, e.g. how many grams of carbs per hour for which effort levels vs how much muscle glycogen you carry. I think that info is very actionable and relates directly to what you and I do on the bike (ie. fuelling, how much and when etc). So far I wasn’t aware of other tools that give you that info on such a personalised level.

Of course, the elephants in the room are the measurements and the validity of the calculations. As with any method, it’s a bit of a “garbage in/out” situation. The purely power-based protocol is very similar to other power duration tests (e.g. Wahoo 4DP) but are obviously subject to execution risk, the same as any lactate test that can be botched by a single drop of sweat or contamination. In that sense, I prefer the power profile as it’s very repeatable, relatively flexible and obviously lower cost than a full lab test (where I live, a remote test is $150-200/test vs $500/600 upwards for labs). You can also go to a lab and complete a lactate based protocol and spiro, too, if that’s your jam. Is the extra burden of the lab or and finger pricks worth it? I don’t know. What are the other inputs? Body weight and an estimate of body composition. Again, the better those values are, the closer the outputs are going to be to reality. If you really know what you are doing, you could amend most of the parameters yourself (or your coach can).

In that sense, the only thing “secret” about INSCYD is that the calculations underlying the outputs are proprietary and closed source. The actual science around it is most definitely not.

So taking @Rad-ler 's points:

  • VLaMax is hard to determine outside the lab [it’s very hard to determine in the lab too]
  • It is rather an indirect marker [correct in that lactate is a byproduct of your glycolytic/anaerobic metabolism]
  • They try to build a system beside the „FTP system“ and proactively say that their system is better and does not rely on FTP [One of their outputs is MLSS/anaerobic threshold, hardly novel, debatable if it’s better or worse than other methods/systems]
  • High VLaMax → sprinter, low VLaMax → Diesel - the same can be determined by the power duration curve [agreed to some extent]
  • They use different terminology with doubtful descriptions of physiology [no]
  • In the end they prescribe just „regular“ training [no, they don’t prescribe training. Well, INSCYD doesn’t]
  • Anything else to debunk this topic? [as with anything, you need to decide how useful the insights I describe above are to you and your training/goals. You most definitely don’t need it, the same way as you don’t need a power meter or a heart rate monitor]

My uneducated answer: if you would try to do a classic 2x20 threshold workout, you could save 500 bucks and know that you aren‘t a 330W guy and your FTP is set too high. It‘s really not hard to find out. Unless ones ego is in the way.

A recent, and much more educated information by Kolie Moore why the Vlamax model has incorrect assumptions:

So why to rely on a model which has proven wrong, is expensive and doesn‘t provide more information than the power duration curve?


So, long form FTP test (KM test): free
icu.intervals: free or donation
WKO5: $170 and lasts for years with free updates

I just don’t see what Insycd provides that you don’t get from WKO5/ + seeding efforts and a real FTP test.

I do see that Insycd provides a pretty report and a point of sale for coaches. I don’t believe that it can really map metabolism (fatmax, VT1) to a power profile other than in the most general way based on averages.

But if one really wanted VLAmax and a nice report, one could buy Aerotune’s test for 25 euros (month’s use) or 200 euros per year.

Personally, I think you can’t beat WKO5. The learning curve feels steep but it’s cheap and you get updates for years for one low price.

Got a podcast for that?

I also question what is the value of knowing your “fatmax”? Why would I need to know it and how would knowing Fatmax influence my training for the better?

1 Like

I don’t recall. Maybe twitter comments. He only diplomatically said that he didn’t find it useful for his athletes and stopped using it. I do recall that they use to advertise remote Inscyd testing packages in every podcast.

Edit: check this:

Mark Burnley, IIRC, much more strongly suggests that it’s BS.

Search twitter for “vlamax” and look at the debates and comments. You have to scroll down a lot.

I know which coaches and physiologists that I trust.

Another edit: on the Empirical Cycling podcast there are two Andy Coggan episodes - they talk about vlamax (short version: BS).


Thanks for the further reading/listening links! Obviously have some more reading to do.

As for the usefulness of model, I believe I should have made clearer that I see INSCYD as an additional tool, rather than the sole “source of truth”. It’s a model, therefore implicitly has its shortcomings. The challenge for the community is obviously to come up with a better one.

As for FTP detection, yes there are simple, easy, low to no cost alternatives to using INSCYD for standalone FTP/MLSS determination and I generally rely on those, too. I don’t see that as the primary use case for the software either (but it should obviously line up with the real world power data). Re the personal anecdote, it obviously highlighted to me the shortcomings of leaning on “quick fix tests” (such as TR) myself recently. Will obviously dive deeper into that.

Re the metabolism/consumption mapping, I believe those outputs are useful for what they are: model outputs. Therefore implicitly subject to some variability. But that doesn’t mean they are not useful to contextualise energy consumption and fuelling for (key) workouts and goal events, specifically carb consumption.

Single, standalone PPDs are obviously a different discussion in terms of cost and usefulness. As a part of a wider, more comprehensive and longer-term coaching package I believe it adds value.

Not a coach. I’ve done INSYCD testing and it wasn’t worth the cost.

1 Like