How does Suunto App decide, what kind of training (aerobic, anaerobic, etc.) is it?
-
@FunkyLarma That’s what I thought - TSS is like an overall value “how fatiguing the workout was”. Zones determine what energy system of the body is used and what capabilities are hit the most - and therefore gonna probably improve the most. And that brings us back to my first question, that probably noone knows - what conditions need to be me met to get the workout tagged “VO2max”. If it’s “majority of the workout is zone 5”, then it makes sense I don’t have any workout tagged like that, because I get into zone 5 rarely and even when the workout is hard, I spend there ~ 25 % of the time at most.
My former question was probably set too vaguely, I didn’t want to dig into zones and TSS too much. I was just curious, what are the conditions which determine, how is the app tagging the workouts and whether someone have a precise knowledge. Like real numbers, not estimations like “for aerobic, you run on low HR”.
@Maryn IMHO yes - there is the explantion in the app. TSS ( hr ) for heart rate, ( r ) for running pace, etc. Just “open” the workout and you can see
icon there.
@wmichi are these intervals hard enough? I mean I believe so, but the only thing how SA could say it’s aerobic/long aerobic, is due to really short time in higher zones. I can imagine you go really hard but for short bursts so your HR don’t need to climb so high. Like when you are doing sprints - you’d achieve much higher HR with continuous hard 5 minutes than 60 m dashes with rests between. My last run was pretty easy, but just few of these hills there were enough to rise the HR and SA tagged it aerobic/anaerobic. -
@Mazel yes, as you say the reason they are tagged a certain way is probably down to a complicated algorithm. To be fair other platforms suffer from similar issues and it is very difficult to try and apply a broad brush to all users. I recently carried out a bike workout that contained 1 minute efforts at 110% of threshold. This is something that would hit and target a VO2Max adaptation and another device makers platform informed me I was completely an endurance ride
From your example it would be incredibly difficult to spend long durations in those higher zones, which gets you into the topic of VO2Max training which is always a hotly debated topic with many nuances.@wmichi idea of training the model is a good start but again this would be so dependent on the athlete. Not everyone will hit peak VO2Max at the same outputs or percentages of threshold, You only need to enter into the ‘your ftp is the power you can hold for 1 hour’ debate to see the fun and games.
Personally I would take all of these with a pinch of salt, if not a very large bag full of salt
-
@FunkyLarma I must say, when 1 minute hard efforts are presented as an endurance ride, the algorithm is really… unconvincing
Yes, I feel the same, it should be taken as an overview. And like I say to my dad, who is using older Garmin watch with wrist HR monitoring - you should not believe exact values (unless you got chest strap in this case) but you can use it to have an overview about the progress. Whether the pace during intervals is going up, HR during easy runs going down, etc. That’s why I take all these calculated values like VO2max with grain of salt, but was curious, how to get VO2max tag.
Btw I was browsing through some workouts of other people in my vicinity and found one man with workout tagget VO2max - and it was like 30 minutes in zone 5. So it seems this tag is really given under the condition of long time in zone 5. -
@Mazel I think it depends on how long you have been in zone 5 in relation to the total duration of the workout. I am pretty sure, that if I was 30min above LT and 1 hour in zone 1/2 during a workout, SA tags this as just aerobic. Yes, the majority of the time this workout was easy, but the benefit of this workout is definitely not low aerobic base training (like it’s tagged). I have had this a lot of times since this feature has been added.
If Suunto Coach is AI driven (I read that a couple of times), then it fails to recognise some sorts of interval trainings correctly. If there’s an ongoing training of this AI, then it should get better eventually. I don’t know, just speculating…
-
@wmichi Yup, I got the same impression. If it’s using machine learning or any type of self-improving algorithm, it should definitely got better learning dataset. I can imagine it’s difficult to distinguish between easy run with some hills which appear like “intervals” but typical interval training is pretty easy to define - speed, HR, time, everything there follows particular trends. It might be interesting to check, whether manual laps help - whether it takes into consideration when you “lap” your warm-up and cool-down and gives it lower weight in the overall classification.
Today was the victorious day! I’ve managed to get “VO2max” tag with the last possible workout structure I haven’t tried before - go almost full out without any intervals, etc. Just run long and hard. So it seems this is the main component. I’ll try shorter distances in the future, but half-marathon in hills did the job. I have checked all my previous workouts since 2020 (2019?) - don’t know when the tags started but it’s my first VO2max workout including all races, etc.
