EPOC/Garmin Load analysis

Training
1

Has anyone experience of analysing and gaining insight from EPOC data, ie Garmin Load?

I’ve manually copied some data out of Garmin into a spreadsheet and I’m realising that it’s not really the same “thing” as TSS.

Some of the values seem really high, some values seem to be highly affected if it’s the second workout of the day, others seem to be tiny.

2
Type Cycling Indoor Cycling Multisport Open Water Swimming Pool Swim Running Strength Training Virtual Running Walking Grand Total
Date (YW) Load (Sum)
2021-W25 93 292 399 219 1003
2021-W26 386 52 254 212 904
2021-W27 69 186 487 404 7 3 1156
2021-W28 171 136 401 362 1070
2021-W29 115 57 146 239 557
2021-W30 32 85 251 454 822
2021-W31 80 457 537
2021-W32 30 136 98 562 10 836
2021-W33 199 385 389 973
2021-W34 114 155 207 271 747
2021-W35 315 162 395 872
2021-W36 19 145 490 8 662
2021-W37 21 231 291 94 637
2021-W38 38 114 809 147 1108
2021-W39 246 567 114 22 37 986
Grand Total 976 2104 1316 114 4057 4056 17 219 11 12870

Indoor cycling values look like the range you might expect for TSS but per workout they might vary 50% for very similar sessions. I’ve emboldened one such where a 100tss ride is only 57 Load.

So in the multi sport category the first two are PB effort olympic triathlons, both sub 2h30, the third entry is a 12hr PB Ironman.

The pool swim values seem high generally, but the week 38 entry of 809 doesn’t seem credible. There was a swim longer than usual but still only an hour and a half.

For context, week 36 was a weak marathon performance, 490 seems reasonable.

3

Check out the white paper from Firstbeat, it’s pretty informative.

My own experience is that it is more exponential in terms of measuring load than TSS. Endurance rides are generally < 1 Epoc / TSS, but can be > 3 for races. I personally find TSS between endurance rides and races can be the same but feel completely different and require different amounts of recovery. EPOC generally tracks with my own perceived effort a lot better.

The one downside is that since it uses HR and HRV, the data can be worthless if the source is contaminated. IE, older / cheaper optical sensors, or cheap straps in general.

2 Likes
4

Have you used it in a prescriptive way, ie I want a session that will generate X load, or for my week to total Y load?

The Garmin interface gives you enough for the here and now but this historical analysis I’m looking at is not served.

At my peak running my run load in 2021 was around 1000/month, but I can’t say whether this a good target or not.

So I looked at the individual outliers, and as you say there were some maximal heart rate points that might be erroneous or valid, as they were all really strenuous sessions. I feel like I’m trending back towards the “descriptive, not prescriptive phrase” of Coggan fame…

1 Like
5

I don’t know what’s good / bad either. I typically stay between 800-1500/week. :man_shrugging:

I wouldn’t know where to start for programming a ride to hit a target epoc. I can do rides of 3 hours and relatively similar kj and tss that have the same work, but have epocs that vary quite a bit. And even then, am I trying to hit a specific number? And if so, why?

I’m just doing good to hit 15-18 hours a week and be able to recover from it, much less getting nuanced with anything else :laughing: After all, there is no magic metric that replaces volume and frequency for endurance adaptations.

2 Likes