Maximal Oxygen Uptake: Does it matter?

Along time ago in a galaxy far, far away, I had time gates to jump through that involved training two or three times a day, attending tutorials or lectures, grabbing a meal or even catching up with friends (usually while training). Oh, and sleeping! Through this halcyon period (“the 90s”), Danger (Damien Angus) and I discussed two major topics: physiology and time optimisation. I know you were all expecting me to say girls, but the topic of ‘fairer sex’ is covered under the umbrella of physiology and time.

One prominent difference between training now and in the 90s is the availability of data. Heart rate, speed and distance was really all there was. I used to volunteer for scientific studies so that I could objectify my performance. This included maximal oxygen uptake (VO2max) and lactate threshold (LT) – using Douglas bags to measure expired oxygen and carbon dioxide with Lode bikes setting wattage. But once I had participated in a few trials, it became obvious that my VO2max wasn’t changing much. It was just a number that gave me an idea of the limits of my heart rate. It didn’t add much to my training, aside from giving me the confidence that I had a decent motor, which I still needed to learn how to use and tune.

With the glut of information we now try to understand on a daily basis, are we over complicating things just a little?

Fast-forward 20 years and on top of the basics, we have power, rpm, left-right balance, pedalling efficiency, temperature, elevation gain, TSS (training stress score) or equivalent, and more. But I feel as though data collection may have jumped the shark. With the glut of information we now try to understand on a daily basis, are we over complicating things just a little? Do you really care what your normalised power was on the weekend ride? Does it really add to your (or your hypothetical coach’s) understanding of the session? Do you even know what normalised power is and how it changes your power number?

Professor John Hawley and his mates essentially covered off all that I am about to blather below in a paper examining the five best predictors of triathlon race time in 20001. In summary these were: LT at steady state wattage of four watts/kilogram on the bike; lactate at 15 kilometres/hour running; cycling peak power output; peak running velocity, and cycling VO2 maximum – you can’t measure all of these without a lab.  But he doesn’t mention normalised power anywhere!

So, in homage to Danger and my chats, usually about Hawley/Jeukendrup or Noakes papers, let’s address the more important measures versus the time wasters, with the goal of spending more time training effectively, and less time in front of a screen trying to understand what the numbers mean. Or, in other words, how others are simply ‘Instagram training’!

VO2max: This is a useful number to have measured once, in the context of finding out a total number in litres per minute. Knowing your maximum heart rate is valuable, and the deflection point of ventilatory threshold gives a solid guide for LT. Note that bike VO2 is usually 10% lower than a running test, because you utilise less muscle mass (thus less oxygen). Would I spend money on running out and getting this testing done? Negative. But if you’re interested in doing this test, I’d recommend volunteering in a study at a university or you could perform the Hawley test described under ‘power’.

Lactate threshold: When partnered with power, this is a valuable tool. By way of illustration, Derek Clayton ran 2:08:33 for the marathon in Antwerp in 1969. This stood as a world best for 12 years. It still stacks up today as a fast time. He had a VO2max of – elite, but not that elite in comparison to many who get a VO2max into the 80’s (Cadel Evans) or even 90’s (Bjorn Daehlie). But he could hold ~99% of this maximum for over two hours (i.e. Antwerp marathon). Taking blood is a drag and measuring lactate is also costly, which is why FTP (Functional Threshold Power) is so often used as a 20-minute test of training performance. It’s a functional and meaningful test of your ability to hold power, therefore a steady state lactate.

Performance anxiety: Having access to performance data can give you neccessary feedback on your training progress, but overdoing it can be unproductive.

Heart rate: This is a great measure to get to know. It is simple and easy to measure. It is a little prone to drift and is variable according to weather etc, but you can still look at it with confidence once you build experience. Danger has 20 years of Excel spreadsheets of heart rates and uses them to benchmark his form, year on year. But it’s a powerful tool that he understands, and it allows him to guide effort and nutrition by knowing his limits. Add power, and you have a useful data set.

Power: This is by far the most objective number you can attain – if you have calibrated the crank! It precisely tells your ability to push the pedals in watts, which is then expressed per kilogram. A peak power test2 can be used to predict VO2 max with a simple step test if you have a Wahoo KICKR or similar. Look at the Hawley paper and try it at home.  An average power can be a much better predictor of a training session than any other measure.

Even if you do take a scientific approach to training, there’s just no sense in comparing minutiae data on a day-to-day basis. Given how inconsistent temperature, pressure, wind, time of day, sleep, nutrition, fluid status and recovery can be, there is too much noise in all your data to draw exact comparisons. But if you need to run the numbers, do it inside on a trainer that measures power at the same time each week. It’s reproducible and thereby comparable data.

Spending time on Strava, Garmin Connect, MapMyRide or the equivalent is just time spent on social media – it’s cyberstalking or general amusement. But it’s not helping your training much. Using these applications in a productive manner may include finding new and interesting routes to train, especially in a new location (i.e. if you have moved). For example, I use heat maps when I’m travelling and can’t be bothered wasting time exploring when I can be doing productive training instead.

Lastly, to complete my rant – DO NOT spend any of your God-given time hitting ‘like’ buttons on Strava. It’s a really strange modern phenomenon. You ‘like’ my training session? Whatttt? It creeps me out. If you are going to stalk me, at least don’t alert me to the fact that you are doing it.

See you on the road! Mitch likes this.


1. Prediction of triathlon race time from laboratory testing in national triathletes. Schabort EJ1, Killian SC, St Clair Gibson A, Hawley JA, Noakes TD. Med Sci Sports Exerc. 2000 Apr;32(4):844-9.

2. Peak power output predicts maximal oxygen uptake and performance time in trained cyclists. Hawley JA, Noakes TD. Eur J Appl Physiol Occup Physiol. 1992;65(1):79-83.



Mitch Anderson

Dr. Mitch Anderson is one of the premier sports doctors in Melbourne working out his practice Shinbone Medical in North Melbourne. The former professional triathlete is your go-to triathlon doctor.

Leave a Reply

Your email address will not be published. Required fields are marked *

Sign up for the mailing list

Enter your details below to stay up to date with whats going on.