Help with TE of some activities

I was waiting for you to join the party 

Buuuut...holly molly...you're going too technical  i'm just a random runner that loves more the data than the running itself eehehehe

Etupes25 said:

- For a good threshold or VO2 max workout, look at your stamina graph. The workout will be optimized if your stamina hits the ground at the end of the last interval.

I'm not sure that i understood this part...not because your explanation is complicated but because my "Threshold" workouts that are labeled, normally, "VO2max" have a stamina of around 30-50% at the end.

Even my last Half with a TE "VO2max" and 5.0  as a value didn't go below 20% of stamina.

I'm not getting many "VO2max" workouts so i don't have good examples to use as a comparison...i found one from April/May where i was probably dying at the end but the min stamina was also not lower than 20%.

And it's not like when i do the "Threshold" workouts that are labeled "VO2max" i'm fresh at the end...be it because of legs or the breathing but i feel pretty exhausted during the last interval...cooldown or after the workout are not a problem but i doubt that i could go faster or longer to try to bring down more the stamina.

"Anaerobic" workouts are a different story and while i feel that the target is pretty slow and i run them faster(15-20 seconds on average) i bring also the stamina a lot lower...generally, around 10/20% but i have examples with stamina at 1% too and different TE(Anaerobic or Tempo).

Etupes25 said:

Focus on pace, not HR. If you don't max out during your threshold or VO2Max workouts, your stamina will not hit the floor at the end of the workout.  Either you have to run faster, or longer, or you have to do more repeats, and crash that stamina thing!

I'm definitely not focusing on HR, especially during "Anaerobic" or "Sprint" workouts but the difference in HR is one of the main differences that i can see.

https://connect.garmin.com/modern/activity/9117962583

https://connect.garmin.com/modern/activity/9700174222 

Looking better, could it be that the watch was expecting me to run, in the second example, 10(or more) seconds faster considering that the target was different?

While i always appreciate your technical inputs i would really appreciate even more if you could take a look at the activities and let me know your opinion on why the treadmill's activities are getting "Threshold" instead of "VO2max" or why the "Anaerobic" is getting "Base".

As you said already, it's up to me to experiment and try but it would be nice to get an input on what i should try to change based on what i did already.

Threshold/VO2max is not a big deal...it's more my personal curiosity especially now that harder(my opinion based on the HR more than anything else) workouts are getting "Threshold" instead of "VO2max". Or it could be put also in another way....considering that the heart effort is higher(due to temperature i guess) with treadmills activities what is the other factor(s) that is making them easier, for the watch, compared to the outdoor example? (It could be just the pace, that is all wrong, because it's an indoor activity.)

For "Anaerobic/Tempo/Base" it looks a bit more important as they are different categories of load. To try to get more "Anaerobic" primary effect do you suggest just to increase even more the speed or length of intervals? Or something different?

Sorry I went over board. I just love this stuff...

Anyway, your 2 activities are pretty close to each other. They have the same workout design (looking at intervals run/recovery).

Conclusion: you ran the 2583 intervals harder, but your performance condition was lower, so your had to work harder to get to a higher output, and you did indeed deplete your stamina more thoroughly.

It is interesting to see that altough the average power is the same, the average interval power difference (+3%) combined with a lower perfomance condition impacts the score. 2583 appeared to the watch as if you had a lower VO2Max (0 to -5 negative performance condition) therefore it looked more intense than 4522 (+5 to 0).

At the end of the day, these are not drastic difference at every level. 

Mladen T said:

why the "Anaerobic" is getting "Base"

I cannot rationalize the "base" training label. It looks wrong to me given that both factors are in the 3+ range. Given the labels from Firstbeat, I would expect a high aerobic label like VO2Max or Treshold, but maybe the hard intervals were simply too short)

Indeed, a 40s/3mn is going to be either a neuromuscular workout (powermax/anaerobic capacity at 150%+ of FTP) if the runner is really going flat out, or a "activation/base" workout, if the power is more in the threshold range (the kind you do at the beginning of a training program). In fact, you see "endurance" workouts that are like 10mn endurance, 10s FTP/FRC

Mladen T said:

what i should try to change based on what i did already

Good VO2 Max intervals are more like 1:2 in terms of hard vs. rest time. The idea is to accumulate as much time as possible around 130%+ of FTP through mutliple repeats until exhaustion (no stamina left). Some suggest that 1:1 is preferred. It really depends on how hard you go.

For a threshold workout, the target is more around 100% of FTP for 15mn+ intervals or 5mn at ~120% with 3:1 work/rest ratio, for example. You should be able to complete about 3 to 4 intervals like this and have a depleted stamina.

If you want to use pace targets, you can map your pace targets to Joe Friel's pace zones (FTP workout at 100% of FTP pace, and VO2Max workout at ~90% of FTP pace). Alternatively, convert pace into speed and use the % of FTP power to establish your target speed, then pace.

As you do a threshold or VO2Max workout, repeat your intervals until exhaustion (you cannot complete the interval). You can use the real-time stamina to do so. You will get the most out of it, but it will be hard :-)

Note

I personally don't use the suggested workouts any longer, because the targets are a bit too low and they are too repetitive. I look at the type of recommended workout (rest, base, tempo, etc.) and I build my own workout based on FTP, by finding some examples online for variety.

Note:

Now about the stamina thing. You can deplete stamina running "endurance" or "tempo" for several hours. This is what happens when marathon runners "hit the wall". Depleting the stamina is just a good indication you put your best in the workout, mostly for intense workouts. Vice versa, it can be an indication that the watch is overestimating/underestimating your VO2Max. A few days ago, I tested my FTP test FTP of 240W, adjusting it to a 252W target for 20mn. During the last 10mn, the stamina hit the floor, and the watch was showing a constant 5mn stamina remaining, and I was able to finish the 20mn holding the target to near exhaustion. That gave me confidence that I had a good FTP number, but the watch was slightly underestimating my VO2max. 

Etupes25 said:

It is interesting to see that altough the average power is the same, the average interval power difference (+3%) combined with a lower perfomance condition impacts the score. 2583 appeared to the watch as if you had a lower VO2Max (0 to -5 negative performance condition) therefore it looked more intense than 4522 (+5 to 0).

About lower VO2max, that's absolutely true and that's why also the target from Garmin is different between the 2 activities.

Keep in mind that i started to run because of the watch and the daily suggestions, so my idea is to avoid starting to customize workouts too much...i know it's nice to understand exactly how everything works and play with the duration/intensity etc but, the second part is not something that i want to take in consideration...not in the near future at least.

I wanted to see the limits of the watch and the limit was definitely the "not ideal" detection of short intervals...got the chest strap and happy with it....the beauty of the HR in some graphs is undeniable.

Let's assume that we found why those similar activities have a different primary label...now the question is...is that correct? And not theoretically speaking but from Garmin side.

Etupes25 said:

I cannot rationalize the "base" training label. It looks wrong to me given that both factors are in the 3+ range. Given the labels from Firstbeat, I would expect a high aerobic label like VO2Max or Treshold, but maybe the hard intervals were simply too short)

Indeed, a 40s/3mn is going to be either a neuromuscular workout (powermax/anaerobic capacity at 150%+ of FTP) if the runner is really going flat out, or a "activation/base" workout, if the power is more in the threshold range (the kind you do at the beginning of a training program). In fact, you see "endurance" workouts that are like 10mn endurance, 10s FTP/FRC

As you wrote here, i would assume that the bases for the "Anaerobic" interval are definitely there...so, next step, why Garmin is giving something different?

In my opinion, it's because Garmin is considering that HR was too low during intervals for an "Anaerobic" label and the HR during warmup/recovery/cooldown is low enough to get "Base" label. I have other similar activities with label "Tempo" and considering that it was during hotter weather i'm assuming the reason is the same but the "Tempo" label was given instead of "Base" because the average HR during warmup/recovery/cooldown was higher.
I never had any other primary effect on "Anaerobic" workouts but my assumption is that if i run a faster warmup/recovery/cooldown i could get a "Threshold" or even "VO2max" label but the last one looks really difficult 

Based on the assumption of HR i compared few activities with max/average HR and this is the result.

Everything was nice until 2962 which has the same max HR of the "Anaerobic" activities.

So i thought to check more in details the intervals

The effort on 2962 looks definitely higher than the one on 0565...still 2962 got Tempo and 0565 got Anaerobic. 

0565 has more time in Zone 5 but then 2583 doesn't make sense...but 2583 is a different type of intervals so it could still make sense.

Should i just stop to try to find the reason behind some stuff? :D

By the way i have an "Anaerobic" session on treadmill later...as a first try indoor i would like to avoid with "pushing even more" option....i'll try to keep the usual 10/20 seconds between the target and the execution and let's see what will happen with higher HR due to the temperature.

Regarding your other suggestions i'll definitely save the post for future reference but, as said earlier, the idea is to not customize too much some stuff and in any case i need to digest all the technical parts

When you look at Firstbeat labels, base training has always anaerobic benefit below 2

https://www.firstbeatanalytics.com/en/features/workout-labels/

I put your data in a table for easy comparison. Only the base workouts look obviously wrong. It is as if there was a bug in the categorization of base, where 3.0 is the new 2.0

You should consider reaching out to Garmin Support and share this data with them. While a label is cosmetic, if there is an issue with categorization, it might impact the training load focus calculations...

At any rate, you are not going to get into trouble if you follow the training suggestions. This is a great feature and simple to use.

I noticed that the latest version of the firmware list tweaks in base suggestions for cycling. Today, my base workout suggested 155W instead of 130W in prior versions. This makes much more sense in my case as I typically target 160W for endurance.

I will see how the future workouts evolve. On prior FW, tempo was targeting 195W, while 200W is my target, so it is very close and good. However, threshold intervals were at 230W for an FTP of 240W. This would be more like a sweet spot target (which also aims at enhancing FTP), but then the intervals are a bit short for sweet spot training.. I prefer targeting 100-105% of FTP for the proposed interval duration, but somebody following Garmin's suggestion would train just fine.

Etupes25 said:

You should consider reaching out to Garmin Support and share this data with them. While a label is cosmetic, if there is an issue with categorization, it might impact the training load focus calculations...

Will definitely try to contact them and see if they have any explanation.

The impact on the load focus is exactly one of my concerns too...when i get primary effect of "Sprint" i don't remember getting any other type of load(low or high aerobic) also with values of 2.6 or 3.0 on aerobic side. 

Getting a primary effect of "Base" or "Tempo" with an "anaerobic" training gives me both anaerobic and low/high aerobic loads.
In this case, i guess, that the amount of anaerobic load is lower than what it should be with a correct label....and high or low aerobic load is increasing when it shouldn't.

I'll keep an eye on the similar trainings in future...for today, i've got my screenshot of the load before the workout and will see what happens after...whichever label i get.

I'm back with something less serious but still curious to understand if there's a real reason or if it was just a miss from Garmin?

I know i will never get a certain answer but it's ok.

I was really surprised for the "Base" effect on the following activity https://connect.garmin.com/modern/activity/11040715443 (it was supposed to be a "Recovery" one).

A comparison that i used is https://connect.garmin.com/modern/activity/9339387940 

I understand that i run "a lot" faster than the target but the HR was ok for a "Recovery" run(if they didn't change anything it's supposed to be 130 in my case).

Few points that i would highlight:

- max hr was lower in the "Base" activity(140 vs 143)

- avg hr was lower in the "Base" activity(131 vs 135)

- i guess i get a bit more spikes in the "Base" run but i don't see any anaerobic load

- in the "Base" run i run for 28 minutes vs 32 of the "Recovery" one

The training load is 49(recovery) vs 50(Base) but why? Isn't it based mainly on the HR?

I had 99% Stamina instead of 100% but could that explain it? Altitude? But we are talking about a big difference here while the difference in load is only "1".

The avg power is definitely different but that's a consequence of the higher pace.

Etupes25 waiting for your comment

Etupes25 said:

Also keep in mind that the TE is a combination of 2 factors: the peak EPOC from the exercise and the training history (frequency and intensity). For a less intense training history, the same EPOC will yield a slightly higher TE and vice-versa

If acute load is another factor then this could be the reason as in this period it's around 500 while it was around 700/800 during the other run with "Recovery" effect.

But you're saying that same load will give different TE level and not that same effort will give more or less load right?

Or could that be the case too? Based on the activity history you generate more or less load with the same effort? 

That could explain why i had a tiny higher amount of load for a lower effort? I know that 49 vs 50 is nothing but it's 1 point more with less time run and lower avg/max HR.

With Garmin, the load is measured by the peak EPOC. The good thing is that is the closest measure to the metabolic effort and therefore highly individualized, reflecting fatigue and fitness etc. The bad thing is that it is estimated from HR and that comes with the problem of HR capture compounded by the error of the model. In their white papers, Garmin doesn’t express the average error as a %, but as an absolute value. From the data set they illustrate I estimate the error is around 10%.

Another parameter of the EPOC estimate is the resting HR.

So you get a pretty OK approximation of the best way to quantify the metabolic cost for a specific individual in specifc conditions (compared with just looking at mechanical output like pace or power in TSS).

It is hard to compare the TE across activities because of (a) the average error (b) the different conditions and (c) variations in recent training history.

As an additional note, the watch is using the HR kinetics, HRV and pace/power to evaluate the part of the EPOC stemming from using the anaerobic system. 

I keep forgetting that pretty much everything is still an estimation with more than few % of possible error