Garmin Race Predictor

Vladimirnapkin

I've just noticed my Garmin Fenix 3 HR (Great value on Amazon, BTW) includes a VO2max calculator and a Race Predictor, based on this VO2max. My suggested race times are not crazy, but pretty optimistic. Anyone else have this feature? I'm curious about your experience.

Duck_Puddle

My Garmin race predictions seem to think I’m Meb and not a middle aged, chubby woman who thinks marathons are like mullets and the party is in the back.

That’s a tad bit of an exaggeration, but I suspect the vo2max estimate it’s using to generate the predictions is quite generous.

dewd2

Mine tells me I should be 1 minute faster for a 5k, 2 minutes faster for a 10k, 2 minutes faster for a HM, and 30 frickin' minutes faster for a marathon. My mind tends to agree with this but my body never seems to go along...

DX2JX2

Mine is about ~15% optimistic.

lporter229

My Garmin does not have this feature. I have always used the Daniel's Training Tables to predict my race times. I think most people find them to be fairly close assuming a) you are properly trained for your race and b) you are experienced at running that distance. Otherwise, they are known to be a bit generous (the consensus seems to be that Daniels tables are about 20-30 minutes fast for a first marathon). Check it out and see how Garmin compares? I would be interested in hearing the comparison, as the subject of the Garmin predictor came up recently on another thread.

BruinsGal_91

I've not seen the Daniel's Training Tables before. Looking at the HM prediction, I think I must have accidentally ticked the "you are Paula Radcliffe" box.

Vladimirnapkin

BruinsGal_91 wrote: »

I've not seen the Daniel's Training Tables before. Looking at the HM prediction, I think I must have accidentally ticked the "you are Paula Radcliffe" box.

LOL

scorpio516

Mines about 1 minute slow for a 5k, pretty close for a 10k and unknown over that
FR230

Vladimirnapkin

I need to look up my measured VO2max (treadmill test) from some time ago and compare to what the watch is telling me. As with all these things, there's the "VO2max" and there's proper training for the distance. Historically, as I'm getting back into shape, I can bang out some fast intervals but still perform poorly in races until I get my volume up. (consistently for a few months)

Duck_Puddle

In reality, mine is about 15% optimistic for races that I’m adequately and appropriately trained to run. Which also means that it thinks I can complete a marathon in a mere 32 minutes longer than it took me to do a 17 mile race (where I blew my A goal out of the water). I’ve not done a marathon yet, but I suspect that’s extremely optimistic as I don’t generally run 4 minute miles.

For longer races where I’m undertrained and/or under-experienced, it’s off by as much as 45%. Such as my first half where I’d only run 4 or 5 runs over 10 miles and my weekly mileage peaked at 25 or something.

It’s more consistently about 15% optimistic now, but it’s a straight lookup from what it is calculating as your vo2max. And from my actual race performances, I’m guessing that’s also about 15% optimistic.

Tacklewasher

I think mine is ~ 5 mins optimistic for 5K (and 10 mins for 10K). But it's winter so maybe it's right and I can run that fast if there wasn't white *kitten* on the ground...

spiriteagle99

A lot of predictors assume you are training 70+ miles a week. If you aren't doing that kind of volume, you definitely need to be more conservative.

DX2JX2

lporter229 wrote: »

My Garmin does not have this feature. I have always used the Daniel's Training Tables to predict my race times. I think most people find them to be fairly close assuming a) you are properly trained for your race and b) you are experienced at running that distance. Otherwise, they are known to be a bit generous (the consensus seems to be that Daniels tables are about 20-30 minutes fast for a first marathon). Check it out and see how Garmin compares? I would be interested in hearing the comparison, as the subject of the Garmin predictor came up recently on another thread.

The Daniels tables and Garmin predictor are pretty close assuming the same VO2Max/VDOT.

The problem is how they work...the Daniel's tables use known race results to assume your cardio condition and then use that to extrapolate race times from lookup tables (likely the same tables used by Garmin). The base assumption about cardio condition is fairly solid because it's based on actual, hard data.

The Garmin predictor assumes VO2Max based on your workout performance and then uses that assumption to predict race times for all distances based on lookup tables. The main flaw of the system is in the VO2Max assumption...the Garmin software can be reasonably accurate if using a chest strap HR, but it can be a crapshoot if only using the optical sensors. The general flaw is an overestimation of VO2Max and therefore, overly optimistic race predictions.

The Garmin predictor and Daniel's tables would likely sync up pretty well if Garmin provided a way to override the device's VO2Max assumption with real data/race results. I'm not sure that it does.

Vladimirnapkin

I've used the Daniels tables in the past and found them to be useful. I was just wondering what folks experience was with the Garmin race predictor. Runners World took a look at some studies of watch-based VO2max estimators. https://www.runnersworld.com/sweat-science/can-your-watch-estimate-your-vo2-max

I was surprised by how close the Garmin VO2max estimate came to the measured data.

As I noted previously, your VO2max is one thing, and training is another. Most of us have the 'aerobic capacity' to run way faster than we ever will, because of limitations in our training, diet, injury, etc.

This may be why most people find the race predictor to be wildly optimistic. I've run faster than all the predicted times Garmin has for me, but I'd rattle to pieces before I could hit any of them today.