Garmin watch 'Race Predictor'?!?

cool story, bro' wrote:

On May 13 I bought a Garmin Forerunner 235, which I’ve been wearing practically 24/7.

Sorry, my mistake, the model of watch is the Garmin 245. (I didn't buy the music version of this model because of various potential reliability issues with that feature, and at the time, I wanted to get a new GPS watch ASAP, and the store didn't have the music version in-stock.)

I don't see what the problem is. It seems pretty acurrate?

????? wrote:

I don't see what the problem is. It seems pretty acurrate?

Eggzactly! Crazy-weird that it predicts my 5k to be the exact same time as the split summation of that workout...(and I didn't program a workout into the watch, but just hit the lap button for the reps.). My respect!

And even though it's giving me a VDOT of 48, it knows not to line up across the race distances (without the training behind me), as follows:

RACE PREDICTOR

5k...21:54

10k...46:18

Half...1:44:52

Marathon...3:52:50.

Before, I was taking those predictions with a grain-of-salt. But now?

And some further detail (for those not very familiar with the VDOT), someone capable of only a 7min mile (my time trial last Saturday) has a VDOT of 40.7, with a 5k equivalent of only 23:46, per Daniels calculator:

https://runsmartproject.com/calculator/

Garmin just makes a guess based on your HR at certain pace. On garmin 235 predictions are a bit optimistic for most of the runners, so they slowed them down a bit with 245.

Mostly it is just luck that watch is predicting same result as your predictor workout. But it will be within +-15s most of the time.

1) It only works if you use a chest strap to measure HR. Wrist HR is too unreliable, except in a few lucky individuals there will be a deviation of 10-20 bpm on average.

2) The "Peaking" is a result from suddenly running faster at a given HR - this is usually an effect after reducing training load/volume, which makes you very fresh and efficient. So if you are able to hammer a workout like you did, you probably went well rested into it - hence the "Peaking" status.

3) It is just an ESTIMATE. If you train in hot weather, or on trails all the time, it will UNDERPREDICT your race performance if you race in good weather on the road or track. Similarly, if you always train at 6 am in the morning in 60F, and then race at 1 pm in 80F heat, it will OVERPREDICT of what you are capable of.

4) Lactate tolerance and mental strength is highly individual. The estimates are based on an average, balanced runner. Someone might still feel quite relaxed at 5 mmol lactate, whereas someone with a max lactate of 6 mmol will be dying at the same 5 mmol. Same is true for HR - some have their threshold at only 85% of max HR, someone else might have it at 93%.

"Peaking" for Garmin means exactly a combination of 2 things:

- your estimated Vo2Max is growing lately

- your training load is decreasing and is inside "recovery" zone or very close to it

Nothing more, nothing less

Regarding the original question, well, it's a nice coincidence that a really rough estimation done by watch matched slightly less rough estimation done by workout. And you don't know a real race result, which is very likely totally different because you're not highly trained and are using bogus wrist HR.

I think it's worth mentioning here that you should only train in hot weather if:

1) Your race will be in the heat

2) You're 2 weeks away from the race

It takes 2 weeks for endurance athletes to acclimate to the heat appropriately (only 1 week if they're very well-trained). Any training in the heat done prior to that is wasting potential adaptations.

I agree but, please, stop talking about "lactate tolerance", go educate yourself. Lactic acid is very easy to understand. I learnt that as a junior in highschool.

Kipchogefan wrote:

Mostly it is just luck that watch is predicting same result as your predictor workout. But it will be within +-15s most of the time.

I agree for the watch to predict the same 5k as the predictor workout, down to the second, was uncanny (and was the main reason for me to originally post). Even if the watch really does typically come within +-15s I think that is somewhat of an amazing tool.

Halvard wrote:

I agree but, please, stop talking about "lactate tolerance", go educate yourself. Lactic acid is very easy to understand. I learnt that as a junior in highschool.

+ 1 ! Absolutely correct by you Halvard. There is no "lactate tolerance" , only lactate threshold.

I know what you say is typically accepted. I checked it a few times against a manual pulse measurement (after running and stopping), and it was spot-on in those instances, though probably a strap provides a greater frequency response in times of very rapid HR changes. And maybe around Max HR, the wrist measurement is questionable? During my intervals, the watch says I topped out at 165 bpm.

I 'rested' for two days after a quite taxing 20x0.15mi workout on Wednesday evening; Thursday evening was a very slow jog on grass.

I'm giving the most prediction credence to the 5X1000m workout results, based on McMillan's comments (per his considerable experience) in the article I linked to earlier. He does throw in some caveats regarding similar weather and course.

Interesting. I could tell during the workout I wasn't just aerobically limited, but structurally (strength) as well.

To you, Kipchogefan and all the others that took the time to thoughtfully respond to this thread, I say "Thank you."

Darby wrote:

"Peaking" for Garmin means exactly a combination of 2 things:

- your estimated Vo2Max is growing lately

- your training load is decreasing and is inside "recovery" zone or very close to it

Nothing more, nothing less

Maybe you got this info somewhere from Garmin? I did essentially 'recover' from a very taxing workout 2 days earlier, but possibly the watch is taking into account several days of activity into its assessment.

Regarding the original question, well, it's a nice coincidence that a really rough estimation done by watch matched slightly less rough estimation done by workout. And you don't know a real race result, which is very likely totally different because you're not highly trained and are using bogus wrist HR.

I don't think an actual 5k race result would be too out-of-line from the Race Predictor workout results, based on McMillan's viewpoint. (See my earlier comments above, particularly the caveats he mentions.)

cbenson4 wrote:

I think it's worth mentioning here that you should only train in hot weather if:

1) Your race will be in the heat

2) You're 2 weeks away from the race

It takes 2 weeks for endurance athletes to acclimate to the heat appropriately (only 1 week if they're very well-trained). Any training in the heat done prior to that is wasting potential adaptations.

An interesting point is that 7min/mile pace workout last Saturday (that I stopped after four laps) was done in what have been typical evening temperatures around here, high 80's deg F with significant humidity; whereas, we got a significant break in temps yesterday from a storm that had moved in. During the end of my workout, there was actually a cool sprinkle going dropping. That break in temperatures may likely have improved my recovery the day of, and significantly helped during the workout.

Head coach wrote:

Halvard wrote:

I agree but, please, stop talking about "lactate tolerance", go educate yourself. Lactic acid is very easy to understand. I learnt that as a junior in highschool.

+ 1 ! Absolutely correct by you Halvard. There is no "lactate tolerance" , only lactate threshold.

I surmise this is in reference to adaptations of the 'lactate shuttle'?

I actually think it is pretty awful in the Garmin 235. I often do a 7.5km solo tempo run, and it normally predicts a race effort about 30s slower than I go through the 5k point of this run.

It is great if it gives an accurate prediction for you, but it seems garbage for me

DMulvee wrote:

I actually think it is pretty awful in the Garmin 235. I often do a 7.5km solo tempo run, and it normally predicts a race effort about 30s slower than I go through the 5k point of this run.

It is great if it gives an accurate prediction for you, but it seems garbage for me

Interesting, then, that Kipchogefan says Garmin slowed their race predictions "down a bit with 245" (the model I have).

I would like to get the HRM-Run strap and evaluate how its more accurate HR measurement compares with the wrist measurement. Also, that strap paired with the watch can supposedly assess 'Run Metrics' (like Ground Contact Time, Ground Contact Time Balance, Stride Length, Vertical Oscillation, Vertical Ratio, and 'Running Power').

cool story, bro' wrote:

[quote]LateRunnerPhil wrote:

1) It only works if you use a chest strap to measure HR. Wrist HR is too unreliable, except in a few lucky individuals there will be a deviation of 10-20 bpm on average.

I know what you say is typically accepted. I checked it a few times against a manual pulse measurement (after running and stopping), and it was spot-on in those instances, though probably a strap provides a greater frequency response in times of very rapid HR changes. And maybe around Max HR, the wrist measurement is questionable? During my intervals, the watch says I topped out at 165 bpm.

[quote]

Also, in attempts to improve its accuracy, I wish to mention that I shaved the hair on my arm where the watch sits, and I latch it down fairly tight, particularly for workouts.

My Felix 6 is predicting I can run a 5k in 17:02. My running has been going really well, but my PR is 17:40ish from almost a decade ago (I’m 40 now). If we get some good weather I’ll give it a try and report back, but I’m doubtful I can go much, if any, below 18min.