ow does Garmin compute VO2Max? I mean the actual calculation, not an arm waving explanation about what it is, which is all that comes up when trying Google searches.
ow does Garmin compute VO2Max? I mean the actual calculation, not an arm waving explanation about what it is, which is all that comes up when trying Google searches.
TL;DR it basically compares pace to HR and assumes there's a linear relationship, afaik. The faster you can run for a given % of your max HR, the better your VO2 Max will be. Personally I've always found…
Hi everyone, thanks Messner for pinging me. I was on vacation, but I'm back this week and will take a look through this to see if there's anything I can add - although it gets tricky sometimes because…
I think that many of the discrepancies are due to the fact that the different concepts of VDOT and VO2max are completely confused here.
VDOT is simply a method to predict the possible results at other distances…
V02 Max is calculated relative to your body weight. The relationship is inversely proportional, which means the heavier you are, the lower your V02 max will be
To clarify, VO2max is a measure of the maximal amount of oxygen your body can use during exercise. The VO2max result is presented in two ways, Absolute in L/min and relative ml/kg/min. By presenting it in relative values (ml/kg/min) it becomes easier to compare individuals. Certainly, if your weight changes then the relative VO2max will change. In reality, your absolute VO2max (L/min) is likely to be unchanged.
I'll make one more attempt to illustrate that VO2Max (in ml/kg/min) can be estimated/predicted without knowing body weight (or age, gender, height, etc.), if anyone cares. Obviously if you are directly measuring oxygen uptake in a lab, then you also need to measure body weight, to get a value in ml/kg/min. But Garmin watches and online race predictors don't ask you to strap an oxygen mask on your face....
TL;DR One way to predict VO2Max (in ml/kg/min) is to do a 1.5-mile all-out running test and look up your VO2Max value in a table. You'll see that the table makes no reference to anything except the time to run 1.5 miles, and the corresponding VO2Max value. Yes, if you lose weight, your VO2Max will go up -- for the purposes of tests like this, it's because you can run faster when you lose weight.
One way you can look at it is that if your VO2Max is being estimated based on running speed, then your bodyweight is implicitly being accounted for -- it takes more oxygen to move a heavier person at the same speed as a lighter person. But the heavier person weighs more, which cancels out the absolute difference in oxygen uptake, when you express VO2Max in ml/kg/min. e.g. If we assume the relationship between speed and oxygen uptake is linear, then person A who weighs 200 lbs should take twice the oxygen to run at a certain speed than person B who weighs 100 lbs. Since person A weighs twice as much as person B, the VO2Max relative to bodyweight for A and B should be the same. That's my understanding, anyway.
Does FirstBeat use body weight in its calculations? Maybe -- only FirstBeat knows for sure. But runalyze, online race predictors and all these other VO2Max test based on running speed don't need body weight.
My understanding is that FirstBeat (and Runalyze) refine this kind of test by analyzing data from sub-maximal efforts -- this is done by assuming that the relationship between pace and HR is linear, so they can extrapolate your all-out effort from a lesser effort by comparing the HR of a given run (at a certain pace) to your max HR.
https://www.collegesportsscholarships.com/measure-oxygen-consumption.htm
(Note that this page reverses the common definitions of "absolute" and "relative" VO2Max, but in this discussion, we all know we're talking about ml/kg/min, and that's what this field test and table refer to.)
"One of the most popular field tests used to determine oxygen consumption is the 1.5-mile running test. (A 1-mile running test is more suitable for adolescents and the elderly.) The primary objective of this test is to run 1.5 miles in the least amount of time. For this field test to be as accurate as possible, distance must be precise and running must take place on a level surface. Running on an indoor or outdoor track is preferred. While the results of the 1.5-mile running test are an excellent predictor of oxygen consumption,3 it’s important to realize that this test favors runners.
Table 1 lists predicted values of oxygen consumption based on the time it takes to run 1.5 miles. Various running times are given in five-second intervals between eight minutes and 15 minutes 55 seconds. These values are an absolute measure of how much oxygen is consumed in milliliters per kilogram of bodyweight per minute (or ml/kg/min)"
| Table 1 | Predicted Values of Oxygen Consumption | ||||||||
| Based on the Time to Complete a 1.5-Mile Run | |||||||||
| Time | Value* | Time | Value | Time | Value | Time | Value | ||
| 8:00 | 63.84 | 10:00 | 51.77 | 12:00 | 43.73 | 14:00 | 37.98 | ||
...
| 9:55 | 52.18 | 11:55 | 44.01 | 13:55 | 38.19 | 15:55 | 33.83 | ||
| * ml/kg/min | |||||||||
I am not arguing with you that VO2max can be estimated without knowing weight. It can also be estimated without undertaking any exercise too for that matter. I was just seeking to clarify how VO2max is measured and then presented. For the record, estimations derived without bodyweight show a higher error than those derived with - Validity of VO2max equations.
And a bit more reading that covers off a range of different methods of obtaining VO2max
And it is much too simplistic to assume that you need twice as much oxygen to move a 100kg person as a 50kg person - both could have the same absolute VO2max measured in L/min; obviously their individual relative VO2max will differ when measured in ml/kg/min. The relationship between velocity and oxygen uptake is not linear particularly at higher velocities. The relationship between pace and heart rate is not linear mostly due to heart rate drift. There is a demonstrated linear relationship between heart rate and VO2max.
I am not arguing with you that VO2max can be estimated without knowing weight.
I know you're not -- sorry. Maybe I shouldn't have responded directly to your comment.
But I've seen the argument in the forums many times that goes along the lines of: "because kg is in the denominator of (relative VO2Max), therefore Garmin/FirstBeat necessarily uses your weight to estimate VO2Max". And I just don't agree with that.
And it is much too simplistic to assume that you need twice as much oxygen to move a 100kg person as a 50kg person - both could have the same absolute VO2max measured in L/min; obviously their individual relative VO2max will differ when measured in ml/kg/min.
Fair enough. But isn't that the implicit assumption in the following situations:
1) 1.5 mile all-out running test which predicts your relative VO2Max from your time
2) Race predictor that uses relative VO2Max to determine race times and/or predicts your VO2Max from a known race time.
(Neither of those use body weight as an input, which seems to suggest that differences in body weight are cancelled out by differences in oxygen uptake. And there's formulas like oxygen cost and the ACSM formula for deriving VO2Max from velocity which don't take body weight into account, either.)
Obviously neither of those are perfect, but Garmin had been using 2) for quite some time until they switched to FirstBeat's race predictor. And the generic concept of associating VO2Max (or VDOT) with race times has been discussed for quite some time (without regard to the weight of the runner). It's the basis for all these online "equivalent race result" predictors, isn't it? You go from specific race result to predicted VO2Max (VDOT), which gives you all your other race results -- at least that's my understanding.
The relationship between velocity and oxygen uptake is not linear particularly at higher velocities.
I'm curious -- would those be anaerobic speeds tho? (In that case, VO2Max wouldn't apply, right?)
The relationship between pace and heart rate is not linear mostly due to heart rate drift.
Doesn't the FirstBeat method assume a linear relationship, according to the white paper? (Although it does refer to linear *and* non-linear relationships) Or is correcting for HR drift part of their secret sauce?
---
Having said all that, IMO lactate threshold is a lot more important for predicting race times anyway.... A lot of these sites which discuss VO2Max have mentioned that, too.
I'm curious -- would those be anaerobic speeds tho? (In that case, VO2Max wouldn't apply, right?)
What most people call anaerobic and what really is anaerobic are very different. As you say, anaerobic does not apply to VO2max as it literally means 'without oxygen'. However, some people consider anaerobic to be a hard effort when the reality is that it is only a high aerobic effort. So to answer your question, in general no it would not be running at an anaerobic pace, just a very high pace while still remaining aerobic.
But, (and there's always a but isn't there?) recent research using biomarkers is currently suggesting that our view of how our three energy systems interact (we have two anaerobic and one aerobic) is about to be turned on its head. The suggestion is that our anaerobic systems play a much larger part than we thought when we believe we are only exercising aerobically.
And as you note...
assume a linear relationship,
assumptions simplify calculations but do little for accuracy when relationships do not necessarily hold to the assumption.
Hi everyone, thanks Messner for pinging me. I was on vacation, but I'm back this week and will take a look through this to see if there's anything I can add - although it gets tricky sometimes because after a quick skim it looks like there are already some well-informed responses that cover things nicely. Glad to see folks taking such an interest.
I can quickly respond to the question you emailed me, Messner, which was about weight loss and VO2max.
The short answer is that, ceteris paribus, you can increase the VO2max on your Garmin watch by losing weight. This is because maximal oxygen consumption as displayed on your Garmin watch is properly referred to as "relative VO2max" - which simply means it is expressed in terms of oxygen utilization capacity per kilo of body weight. This allows for much more meaningful understanding of the feedback from a fitness perspective, compared to absolute maximum oxygen utilization capacity.
HermanB Sorry to beat a dead horse (since I know I've asked this in other threads).
AFAIK, just because losing weight will increase your relative VO2Max, doesn't necessarily mean that FirstBeat directly uses your weight (from your Garmin profile) to compute relative VO2Max.
For example, I linked to an all-out running test which estimates your relative VO2Max based solely on the length of time it takes to run 1.5 miles. No other factors (including weight) are used.
Also, the fact that relative VO2Max can be translated to race predictions (and race results can be translated to relative VO2Max), regardless of weight (or any other factors -- e.g. age, height, gender), also suggests that relative VO2Max can be estimated without knowing a runner's weight. Here I'm disregarding some of the subtleties like correcting for age, distance (especially marathons) and training (weekly volume, long runs, etc.) IOW, think of the old Garmin method of just plugging your VO2Max into a lookup table for race predictions (many online calculators still do this).
So my question is: does FirstBeat directly use the weight from Garmin user profile in its VO2Max calculations? Or is it just the case that losing weight makes you run faster (all else being equal), so your VO2Max will go up as the calculations (in this case) are basically all about speed vs HR?
For example, say I ran two 10k races a week apart, at the exact same pace and HR, but I falsely set my weight for the second race in Garmin Connect to be 10 lbs lighter than the first race. Would FirstBeat calculate a higher VO2Max for the second race? If the answer is yes (2nd race would have higher vo2max) , then how does that reconcile with the idea that a given race result can translate directly into estimated relative VO2Max without regard to weight, or the idea that relative VO2Max can be meaningfully compared for two random people, regardless of their weight? If the answer is no (2nd race would have same vo2max as first), then does that imply that the weight in your GC profile doesn't matter?
To put it a different way, even if I had really lost 10 lbs for the second race, wouldn't the fact that I had somehow moved less weight for the same time and distance at the same HR cancel out the supposed increase in VO2Max? In other words, maybe losing 10 lbs but running at the same speed, distance and HR means that your absolute VO2Max goes down, but your relative VO2Max stays the same.
And if FirstBeat does use your (GC profile) weight, is it used as a "corrective factor" or is it a major part of the calculations?
I will also say that the FirstBeat whitepaper I linked above makes little reference to weight, except as part of the definition of relative VO2Max.
I know none of this really that important for training, but I'm just pretty curious about it....
You've formulated that in a particularly interesting way. In reading it, I found myself going yes, yes, ok, wait what? Which means that I got lost along the way, but don't worry it's not you, it's me. ;-) Have you actually performed a test similar to the one you describe? I'd be interested to hear how it works out.
There are certainly tools and methods out there for estimating VO2max using only distance and time. Here's one: https://www.brianmac.co.uk/gentest.htm (scroll down, it's in the middle of the page).
Firstbeat VO2max Calculation steps; The following calculation steps are used for VO2max estimation:
Source: https://assets.firstbeat.com/firstbeat/uploads/2017/06/white_paper_VO2max_30.6.2017.pdf
If you are asking why AGE is listed as a necessary condition, it's because it's used to estimate HRmax sans a personalized entry. It is my understanding that, when available, weight, age, gender, etc are all utilized to interpret the pace and intensity data and transform it into your VO2max.
HermanB thanks for the response. I’ve read that part of the white paper (and quoted it a few times). My point is that:
- the white paper itself makes no reference to using weight, age or gender to help estimate vo2max
- runalyze uses pace vs. hr to estimate vo2max, and it doesn’t know my weight, age or gender
- a 10km (or other standard distance) race result can be used to estimate vo2max (and predict other race times), with no reference to weight, age or gender. By this method, any two ppl who run the same race result will have the same VO2max
- the old Garmin race predictor just plugged in your vo2max into a lookup table to predict race times (the reverse of the above point). Any two ppl who had the same vo2max would have the same race predictions.
My point with all of that is it’s certainly possible to estimate relative vo2max from intensity and speed without knowing weight. To take that further, if you estimated relative vo2max from intensity, speed and weight and got significantly different results for different weights (but the same intensity and speed), wouldn’t that imply that all of those other methods are nonsense?
I’ll try the example again with some made up numbers: Someone runs a 10k race at 41:00, 180 bpm avg (their max is 200) and their estimated vo2max is 53. We all agree that if they run the same race with the exact same results and HR next month, their estimated vo2max will be 53 again. We also see that if you plug their race results into a vo2max/equivalent race results calculator, they will get the same answer both times, because it’s the same race results. It’s also clear that none of those calculators cares about weight or any other physiological parameters.
Now let’s say the 2nd race was somehow run 10lbs lighter while having the exact same result and HR. Maybe the runner lost ten pounds but also lost some fitness. Or maybe the runner just changed the weight in their Garmin profile to see what would happen. I think that all the online calculators which don’t care about weight would give the same estimated vo2max as the first race. I also think runalyze would, too. But what would firstbeat do -- report that the 2nd race was run with a higher vo2max because the runner weighed less? Wouldn’t that also imply that relative vo2max would not be comparable between two runners of different weights?
---
e.g. the ACSM equation for indirectly estimating relative VO2Max (in ml/kg/min) from running speed makes no reference to weight:
VO2max= (0.2 · Speed) + (0.9 · Speed · Grade) + 3.5
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3743617/
I certainly understand that if you directly measure relative VO2Max by strapping on an oxygen mask, then weight is important because relative VO2Max is expressed in terms of ml of oxygen per kg of body weight per minute.
But if you indirectly estimate relative VO2Max from speed, aren't differences in body weight cancelled out by the difference in oxygen cost to transport different weights? IOW, if two runners, weighing 100 lbs and 200 lbs respectively, could run a 10k race at the same speed and %max HR, then I would (naively) expect the second runner's absolute VO2Max (ml/min) to be twice that of the first runner (because they're moving twice the weight), but I would also expect their relative VO2Max (kg/ml/min) to be the same. And that's what all of those VO2Max/race equivalency calculators would predict anyway (since they don't ask for weight.) (I realize that the dependency between oxygen cost/VO2Max and speed is not necessarily linear.)
---
Or to put it as simply as possible, don't these methods for indirectly estimating relative VO2Max from speed and intensity already indirectly take weight into account, by looking at how fast you can run at a given intensity? Because it takes more oxygen to run at a given speed/intensity if you weigh more.
