Exercising at VO2Max

If you did a treadmill test, you can exercise near VO2max by running intervals at the speed where you reached VO2max on the treadmill (or faster/slower depending on wind, temperature, elevation, etc.). 

If you measure your VO2max running on a treadmill, the numbers are useless for sports like biking or rowing. Every activity has its own VO2max and you can't calculate your way out of lab tests unless you accept very inaccurate numbers.

Heart rate takes minutes to reach VO2max bpm, so aim for speed unless your VO2max intervals are long enough to reach VO2max bpm (which would make it a very hard workout!).

If your Garmin VO2max equals your lab VO2max consider yourself lucky. Last time my Garmin said 55, the oxygen mask said 63. All my Garmin watches have consistently underestimated my VO2max by 6 to 8 points compared to lab tests.

Note that your anaerobic threshold heart rate is not necessarily 95% of your maximal heart rate. The percentage depends on training and (high) individual variation.The percentages in zone charts are about as useful as the average shoe size of the general population when you're out shopping for a new pair.

The accuracy of smartwatches estimates of VO2Max has been tested for a while now and although it would be obviously wrong to say that it's as good as a lab test, it would be equally wrong to say that it's one shoe size fits all.

The below and fairly new clinical study shows why my estimate of VO2Max at 45 shows a much higher accuracy than someone with a Vo2Max at 63:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9286863/

The problem is smartwatches manufacturers won't tell us this and it is for us, the users, to figure it out.

I think most Garmin users who care about VO2max figures will get a chest strap. Too bad they didn't use those in their test.

They explain why they have made this choice in the study. Also, they used 2 watches, one on each wrist (why if not to increase accuracy). I'm sure there exist other studies where they used a chest strap if in your view it is a must. Every study will have to make choices, it doesn't necessarily takes away the validity of the results.

Yes, I read why, but I disagree with their view on this. It is not a must to use chest strap, but it is interesting to see what accuracy you can get with the recommended setup. I assume it would be somewhat more accurate than with wrist heart rate.

I agree that using chest straps could have only increased the accuracy of the results, making the argument that one shoe fits all even less valid.

If you do actual research about the importance of wearing a chest stap vs the watch (or watcheS in this case !), feel free to share the results. I am myself interested but not enough to research this point specifically.

Since they had two watches, it would seem very easy to use a chest strap with one of them, to get an idea of how much it affects the results.

These are additional results I would have been interested to have too.

Not sure if a chest strap would make much of a difference. In all my pre-marathon lab tests, heart rates measured by ECG electrodes, chest straps, and my Garmin 255 (and my old Vivoactive 4) were all within a beat or two at speeds below my 5k pace. My Garmins only got erratic near the end of the tests at paces that produce data so erratic (step frequency misread as heart rate) that Garmin ignores that part when calculating VO2max.

The problem with the Düking et al. paper cited above is not the chest strap but the sample size. Splitting 23 lab rats in three arbitrary fitness groups regardless of sex makes the problem worse, since low/medium/high VO2max values are not the same for men and women.

Each group in the Düking paper had an average of 4 male and 4 female runners. That can only yield very shaky statistics.

Of course this bites most exercise physiology papers. The literature is full of published hobby projects and pilot studies that score low enough P-values by accident and never ever get a proper follow-up.

Btw, the authors themselves write: "within the lower ˙VO _2peak category, the smartwatch tends to overestimate the runners’ actual ˙VO _2peak , whereas within the higher ˙VO _2peak category values tend to be underestimated."  Low numbers pushed up, high numbers pushed down... the Garmin-calculated values indeed converge towards average shoe size!