The Problem with VO2max

There’s a good chance you’re using the term “VO2max” wrong…

If you’ve ever heard cyclists comparing stats (we don’t do that, do we?), or humble-bragging via Strava ride titles, then you’ve probably heard something like “My VO2max is 400 W”.

That statement is a bit nonsensical.

In the constant effort to be more precise in how we talk about physiology, fitness, and more importantly how we actually prescribe training, let me try to explain why.

This post was getting too long, so consider this first part more of an academic discussion on the misleading use of ‘VO2max’ in cycling conversation. The fun training advice will follow in a second part!

VO2max is a Process, not a Point

First thing to be clear on:

VO2max does not describe Power.

VO2 is the Volume of Oxygen consumption per minute, measured in absolute litres of O2 (L/min) or relative to weight in millilitres of O2 per kilogram body mass per minute (mL/kg/min).

VO2max is your maximum attainable rate of O2 consumption, usually determined with a ramp test protocol (we’ll get into that later…)

Simplified: VO2 is a measure of the input, Power is a measure of the output

VO2max is the maximal input your body can use, and therefore the limiting factor to producing Power Aerobically.

VO2max Power (PVO2max, or other terms like MAP, PPO, Pmax, Wmax, etc.) is a theoretical power target that corresponds to VO2max. However again I believe referring to a magnitude alone without providing a coinciding duration is incomplete.

Power @ Duration = Workload

The issue I keep hinting at is that VO2max does not refer to a single discrete point along the Power-Duration Curve.

AN Aer Curves_Inked2
Theoretical Power-Duration Curve with Aerobic and ANaerobic contributions to VO2max… if only real-life was this simple!

VO2max is typically thought of as your power over ~3-8 minutes. This is accurate enough, and actually a better way to think about VO2max as a range of workloads along the PD Curve. There is a minimum and a maximum workload – a range of Power @ Durations – that will elicit VO2max.

Let’s work through the general concept with a very rough summary of the physiology involved:

Minimum workload to elicit VO2max

At moderate workloads above FTP, fatigue begins to increase rapidly. Your body becomes less efficient at producing energy, and you have to consume greater amounts of Oxygen in order to generate the required power demand.

This results in VO2 rising over time.

You can see this in the slow rise of your heart rate during a moderate duration interval (think of that max ~8-minute effort).

Given a high enough workload and moderate length duration, your VO2 will reach VO2max.

However below a certain workload (generally at FTP) your body will be able to ‘keep up’ with the workload without losing efficiency and without the corresponding rise in VO2.

Clearly then, there is a minimum workload – a ‘moderate’ Power @ Duration – that will eventually elicit VO2max, as VO2 rises with fatigue.

VO2 Severe Moderate3_INKED2
Theoretical behaviour of VO2 under constant workload of various intensities. Arrows indicate exhaustion.

Maximum workload to elicit VO2max

We’re glazing over a lot of the physiology, but let’s move to the upper end of the Power-Duration Curve and consider ANaerobic Power.

Your ANaerobic system produces power in the absence of Oxygen, and can produce power much faster, and up to greater intensities than our Aerobic system. But it is also much faster to fatigue.

So as our power demand goes up to more severe intensities, we begin to use relatively more ANaerobic capacity and less Aerobic contribution – meaning we rely less on VO2.

VO2 is then no longer the limiting factor to fatigue, as we deplete our ANaerobic capacity to exhaustion before we can even reach VO2max.

Think of a very short 10-second maximal effort at the severe top-end of ANaerobic power. You might feel like you hardly have time to breathe during the interval! And you won’t be out of breath in the same way as during a sustained moderate-duration (eg. 8-minute) interval.

Therefore at some maximum workload – a ‘severe’ Power and low Duration – our Aerobic system will not reach maximum fatigue, and we no longer elicit VO2max.

Determining VO2max in the Lab

Now we should understand why VO2max can be elicited by a range of workloads along the PD Curve (roughly ~3-8 minutes). So how do we test to determine VO2max?

The generally accepted method to directly measure VO2 is with a cycle ergometry ramp test to exhaustion while wearing a mask that measures respiratory gases. This requires visiting an equipped lab with a qualified technician to interpret the resulting raw data (… for now).

ramp test simple_inked1
Sample Ramp Test to exhaustion, used to determine VO2max (power data from a Wahoo Kickr, which artificially smoothes the power output)

The general test protocol starts easy and ramps up the intensity in stages until the subject reaches exhaustion. The peak rate of Oxygen consumption attained during the test before failure is then interpreted VO2max. And the power at which VO2max occurs is your VO2max Power.

Easy enough, right?…

However a problem arises when looking through the literature on VO2max testing protocol.

VO2max Ramp Test Protocol Variability

Ramp Test protocol vary in 3 significant ways:

  • Starting workload
  • Rate of workload progression per stage
  • Interpretation of VO2max Power

Starting Workload

Starting workload usually varies between 1 W/kg to 3 W/kg for the first stage. This mostly depends on the population being studied. eg. untrained subjects will typically start at a lower intensity than well-trained or Elite study participants. This makes sense and probably has the least effect on the test results.

Rate of Stage Progression

The stages can progress by anywhere from +20 W to +50 W, and each stage can be 1min, 3min, or even 5min in duration.

We know that intensity is inversely related to duration, so the stage duration & intensity will affect your rate of fatigue and rate of VO2 increase toward VO2max.

You can imagine that it will take less time to reach exhaustion and VO2max if each stage is progressing by +50 W, compared to +20 W. And less accumulated fatigue means you’ll probably be able to hit a higher peak power number!

ramp tests2_INKED1
Two theoretical Ramp Tests of differing stage duration and progression, showing possible difference in resulting VO2max Power. Arrows indicate exhaustion.

So a test with longer stages or lower progressive stages will usually produce a lower maximum power, and a test with shorter stages or greater progressive stages will produce a greater maximum power.

VO2max Power

Finally, Power at VO2max – PVO2max, MAP, PPO, Pmax, Wmax, etc. – may be defined in different ways.

VO2max Power may be interpreted as the peak power of the final stage attained (even if you only complete one second?) or the peak power of the  final stage completed (even if you go on to complete all but one second?).

Other definitions include the average power over the last 1min or 3min of the test, or the peak 1min, 2min, or 3min power attained during the test.

Ramp Test demonstrating the potential difference between ‘peak’ and ‘last’ 1min power.

You can see how trying to compare your “VO2max power” with another athlete – or even with yourself over time! – will be utterly dependent on the particular methodologies and interpretations of the tests each of you used.

(references below for a sampling of test methodologies)

So.. to summarize:

  • VO2max does not describe power: it is the greatest rate of oxygen consumption, and represents the greatest attainable Aerobic workload.
  • VO2max can be elicited by a range of workloads along the PD Curve, with a minimum and maximum intersection of Power @ Duration along your individual PD Curve.
  • VO2max Power is not a discrete point, and depends on testing methodology and data interpretation.

Redefining MAP and time >90% VO2max

Now that we’ve broken down the concept of ‘VO2max’ into a million shattered pieces, how should we actually talk about VO2max? How do we compare numbers and humble-brag on Strava?

Next time someone asks what your VO2max power is, be sure to provide both a Power and a Duration. Compare your 5min Power, or talk about the 6x2min @ 350 W workout you were just barely able to complete!

If that’s too boring, I do think we can find useful, functional terms that helps inform our individual training prescription.

Let me propose what I use to be precise in my own training prescription.

Max Aerobic Power (MAP)

I use MAP to refer to our upper limit to VO2max; that is, the maximum intensity (Power) that elicits VO2max.

MAP corresponds to a high power and short duration, in which we deplete our ANaerobic capacity and just barely reach VO2max before hitting exhaustion.

Think about a max effort 2-3min interval: For many of us this will feel almost easy for the first ~60-90sec, as our Anaerobic reserves are able to fulfill the power requirement.

Then our legs suddenly turn to lead, and we find ourselves gasping and grinding out the second half of the interval. This is where we’ve depleted our ANaerobic reserves, and our Aerobic system has to work at maximum to keep our legs churning over.

Time at 90% VO2max

Although it doesn’t exactly roll off the tongue, I use T90VO2max to refer to the duration spent at >90% VO2max. [UPDATE: I just try to talk about time accumulated >90% VO2max now, without the awkward acronym).

This is the value we’re trying to maximize with interval training to optimize our VO2max adaptations. (look for a forthcoming post for the details!)

By carefully manipulating power during training intervals, I’ve seen studies where participants spent over 26 minutes straight at >90% VO2max! That’s wild! When we consider that VO2max is typically thought of as your ~3-8 minute power, it’s clear how impressive that is.

DISCLAIMER: These definitions are not universally accepted within the literature. You’ll find these two and other terms used interchangeably, synonymously, or to refer to completely different measurements. I’ve tried to be consistent in how I define MAP & T90VO2max based on reoccurring terminology and simple preference. I use them precisely, and I hope accurately.

With these two points, I can precisely talk about a maximum intensity at the high-end, and a maximum duration at the low-end of the VO2max range along the PD Curve.

As long as I know these points for any of my athletes I can optimize interval prescription for them within this range to maximize adaptations to VO2max.

Optimizing VO2max Interval Prescription

I have to leave this as a tease for now! In a forthcoming post I’ll dig into the literature and what I’ve learned in practice, on how to optimize interval prescription for VO2max adaptations.

We’re just getting to the fun part!

Just a tease of workouts to come 😀

The most interesting references

The rest of the references

9 thoughts on “The Problem with VO2max

  1. Quick question on pVO2 max. Wouldn’t a 5-minute all out interval not be a true measure of pVO2 max since there is quite a bit of anaerobic energy contribution?


    1. Hi Todd, let me give you a long answer to your quick question 😋

      What do you mean by ‘PVO2max’?
      What do you mean by ‘all-out’?

      PVO2max doesn’t have a standardized definition in the literature. As this article talks about, ‘power at VO2max’ depends on the protocol used to determine VO2max, and the characteristics of the ‘VO2max’ training interval being performed. It might simply mean ‘the power that will elicit VO2max during a work interval’. In which case the duration of that interval will determine the intensity that will elicit VO2max, and that intensity will change for subsequent intervals and subsequent workouts.

      And you’re correct, no matter how you determine VO2max or PVO2max, any workload that elicits VO2max will necessarily also include significant anaerobic energy contribution (I mean.. anaerobic (glycolytic & alactic PCr) metabolic pathways are constantly contributing to energy production at all intensities, but that’s a longer discussion).

      ‘All-out’ does have a typical definition in the literature. It refers to a maximal, non-paced effort, where you are essentially trying to go as hard as you can for every second of the effort. That means you start at a full sprint and keep sprinting as hard as you can for as long as the effort lasts. You’re not trying to maximize average power, you are trying to maximize instantaneous power at every single second.

      The most common all-out intervals are the 30sec Wingate and 3min Critical Power tests. You can go look up these tests. I’ve done both many times, but actually haven’t talked about them much yet.

      Therefore by design, a properly executed all-out interval will deplete virtually all of your anaerobic resources as fast as they can be depleted, leaving you with only oxidative metabolism to produce power at the highest rate it possibly can, ie. at VO2max. In the case of the 3min all-out Critical Power (CP) test, your power will rapidly decline and stabilize at CP at the end of the test (by definition) while you are essentially at VO2max through the entire interval. To clarify, this means at the end of the interval you will be at VO2max, but only producing power equal to CP (analogous to FTP).

      So it kind of doesn’t make sense to isolate or talk about VO2max or PVO2max in the absence of anaerobic contribution. It’s the process of drawing on anaerobic metabolism that helps stimulate a rise in oxidative metabolism toward VO2max. I know you’ve already read some of my other articles that talk about more practical interval prescription to optimize ‘VO2max training’. It sounds like you’re on the right track to putting together a more actionable, less pedantic & conceptual answer to your questions 😁 Keep them coming!


  2. Jem,

    If I would target maximum elicitation of VO2 and opt for a shorter interval length (in this case 2 minutes) e.g. 6-8 x 2:00 would it be better to:

    i) opt for shorter rest in between interval e.g. 1:00 in order not to get complete recovery and hence elicit a stronger VO2 response with a corresponding heart rate drift and total higher average heart rate? or…

    ii) opt for longer rest in between e.g. 3:00 – 4:00 in order to get more recovery so that the power output during the intervals could be maximized?

    Clearly in case i) heart rate is going to be higher but at the expense of power which will be something less in percent of maximum VO2 power (measured as maximum average power over e.g. 5:00), and likewise in case ii) power as % to pVO2max @5min will be maximized/optimized but at the expense of lower total average heart rate.

    I guess the answer – if wanting to get optimal VO2-training – would be i) with short rest in between intervals and sufficiently high power to pVO2max. Case ii) may have its benefits but probably not for VO2?

    Best regards / Björn


    1. Hi Björn. Great question. Short answer is: I don’t know!

      It’s a question of what is a better proxy for adaptive strain: metabolic intensity (dose = duration x VO2) or external workload (dose = duration x power)?
      And I don’t think we know the answer to that. This could probably be a 3000+ word post itself…

      My opinion based on my understanding of the literature is that to optimize high intensity interval training for performance outcomes:
      a) Power should be above FTP/CP/”Threshold” at minimum
      b) Higher dose of metabolic intensity (↑ duration x ↑ %VO2max) is correlated with physiological adaptations and performance outcomes
      c) But we don’t know mechanistically how that total dose should balance intensity and duration. eg. is less time at slightly higher %VO2max better than more time at slightly lower %VO2max?
      d) We also don’t know mechanistically if the adaptive benefit of elevated VO2 during rest intervals (ie. metabolic activity toward restoring homeostasis) is different from elevated VO2 during work intervals (metabolic activity toward contractile activity & locomotion)
      e) there are more possible levers to pull, eg. interval pacing, consistency over time, periodization, but the above is enough to lay the groundwork for my point here 🙂

      To address your specific options, you could take them to extremes and ask whether to perform sets of Ronnestad- or Billat-like 30/15s or 30/30s, in which case metabolic intensity remains elevated for essentially the entire work-set duration, despite only producing external work for 1/2 or 2/3 of that duration. Or whether to perform MAP TMAX intervals, where you hold MAP/PVO2max to task failure as long as possible, then rest completely (10+ min), in which case workload-stimulated VO2 is maximized.

      The nice answer is both will give benefits and can/should be used at different times, for different purposes! And yes, definitely balancing somewhere in the middle is also a completely valid strategy.

      In general however, and I’ll be talking about this more hopefully soon when my colleague’s meta-analysis is published, I think that it makes more sense to optimize for sustained work bout duration at elevated, but not maximal intensity. eg. longer work intervals (>4-min) at ~8/10 RPE. Remember, VO2 onset kinetics might be anywhere from 60-sec to 2+ min just to reach near VO2max, depending on workload and your individual physiology. So 2-min intervals might not give you a lot of sustained time near VO2max, meaning a lot of that workload might be produced anaerobically.. which might not be a bad thing! (to circle back around to the top of this circular answer 🙂

      There isn’t a straight answer, but hopefully you can understand a bit more about what, why, and when you can prioritize further toward one option or the other!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.