About HR zones, training load and balance assessment

Hello all,

I’ve been wondering for years (since I started running with a Fenix watch, and trying to get a consistent and productive training) how to set my training zones and use them.

I’ve probably spent hundreds of hours to try to understand how Garmin (and firstbeat) use the data we provide them to define if we train efficiently or not. I’ve tried every possible setting (HR max, lactate threshold, resting HR) and at some point, I guess I can share a few thoughts I’ve been able to pull from my experience. Of course I could still be wrong so if anyone disagrees, please comment ! It would be interesting to know if anyone has a different experience.

As I said, I’ve tried from the very beginning to set my HR zones right. So I went into a medical center and made an effort test to determine my VO2max, max HR and lactate threshold.

First thing I noticed was that my HR max didn’t fall too far off, and neither did VO2max measured by the watch ! Actually both were pretty close – which makes sense because they’re probably the easiest to calculate (as absolute data, since your HR theoretically can’t go above the measured value : the watch just has to grab the higher value it reads from the sensor, acknowledging it is accurate – which is another debate). Strange thing when the test report mentioned TWO values for threshold : VT1 & VT2 (Ventilatory Thresholds). Quite confusing !

Eventually I of course found out that there are hundreds of ways of describing HR zones, especially zones 2, 3 and 4. Pretty much everyone agrees on zone 1 & 5 (once again, makes sense as they are just minimum and maximum. Not too hard to figure out). So once again, I tried different settings, changing the values for each zone boundaries and checking if the training benefit of my sessions (tempo, VO2max, Anaerobic, sprints…) matched the intensity and duration. Results were that however the zones were set, training labels always seemed to consider percentage of HRmax as the only reference value for the intensity measurement. Which led me to think that zones settings are completely useless, or at best, they are just an easy way to see where you stand during your workout. I am pretty sure the zone settings aren’t baked into firstbeat’s algorithms. I think only HRmax is used for the assessment, and maybe (once again, not sure about it) lactate threshold – unless, as well, a percentage of HRmax is also used to assess it.

None of the ton of documentation I’ve read on Garmin or Firstbeat websites says otherwise, and their zones definitions are quite unspecific, so they can be interpreted very widely... ("easy to hold conversation", "hard to breathe", etc).

More specifically, Garmin seems to consider long workouts under 85% of maxHR as base, regardless of how thresholds or zones are set.

Also, one interesting thing I found out - despite someone wrote in a previous post that they were filtered and the data did not account for the training assessment - warm up and cool down do actually interfere with the type of workout label. So even a hard 2x 10x 30/30 session at VO2max will be labelled as tempo if your cool down mile is a bit too long or too fast. It is common knowledge in polarized training that WU & CD should account for low-aerobic training, but I’m pretty sure TE algorithms just ‘averages’ the workout intensity level over the complete duration. I was able to check that by splitting the workouts sessions in 3 recorded activities – warm up, exercise and cool down – and obviously got different values for aerobic and anaerobic TE. The split session would give base / VO2 max / recovery, while the single session would register as tempo (with 0 as low aerobic score, unlike the first method).
Even though they are the exact same workouts.

I still think the data we get from our watches is quite interesting and it can be very helpful to lean towards a better way of training ! But these two aspects about HR data interpretation are limitations that I believe are very much part of the Garmin ecosystem. You just need to be aware of them to get a better reading of all the data you get from it, and take it with a pinch of salt Blush

Top Replies

  • TE algorithms just ‘averages’ the workout intensity level over the complete duration

    I agree with your finding except this one. The TE uses peak EPOC and the dynamics of how it is building…

All Replies

  • Hi DRKGLB,

    Good info. I din’t have anything more to share than I agree with you, HR Zones are not took into account. 

    Regards. 

  • Great post - thanks for sharing your thoughts on this!

  • TE algorithms just ‘averages’ the workout intensity level over the complete duration

    I agree with your finding except this one. The TE uses peak EPOC and the dynamics of how it is building up (rate of change and intervals, pace/power derived intensity). Because it is peak, it is not averaged.

    So even a hard 2x 10x 30/30 session at VO2max will be labelled as tempo

    Yes, tempo will be used if the patterns/criteria don't fit the other labels. This could be because the rest periods were not long enough, or the warm up/ cool down periods were too high and gumed up the recognition required EPOC dynamics. I know there is a Garmin article with a note describing how high intensity could end up being labeled as tempo, but I couldn't find it again.

    In practice, there is no need to separate the warm up, recovery periods, cool down from the workout. Although low aerobic training is essential, it is not achieved through these periods but through long duration easy efforts at VT1. These intermediary periods will not affect the peak arobic EPOC significantly unless too long and/or too hard. Just use the TE data fields to observe real time how they evolve during a workout. Worst case, the TE components will be valid, but the label will be off.

    EPOC will decay during rest periods if the intensity is low enough. The watch is looking for these "OFF" periods as part of the EPOC dynamics analysis. So I recommend to walk for at least 30s during recovery/rest periods and then job very easy if you are concerned about the execution score. This will ensure a more accurate EPOC estimates.

    Still to help the watch analyze, having longer recovery periods is helpful. If you do 30s/30s anaerobic, you might get a tempo, but if you do 30s/1mn+ you will get better anaerobic recognition.

    Of course, this is all about getting the watch to give you due credit. Any of the workout above is a good anaerobic workout.

    Finally, EPOC is cumulative, so if you do a workout activity just after a warm up activity, the resulting EPOC and load should be the same or similar to a single activity, except for the decay during the switch of activity.

  • People seem to be very concerned with individual activities's workout labels, and whether they feel correct. But I have the impression that the label really is just what it says, a label for the "primary benefit". Regardless of the label, an individual activity can still affect other components of the training load focus as well. So if a longer/more intense cool down period shifts the label/primary benefit from Threshold to Base, it doesn't mean that you wouldn't get high aerobic points to load focus from that exercise. It just means that you got enough low aerobic to change the label.

    Or have I understood this wrong?

  • Garmin doesn't publish the rules they use to apply a workout label. This table might provide a hint of the type of dynamics/rules the algorithm is looking for

    https://assets.firstbeat.com/firstbeat/uploads/2015/10/white_paper_epoc.pdf

    Aside from the (fundamental) fact that an anaerobic workout that is labeled tempo by the watch is still an anaerobic workout, provided that the effort during the work interval correspond to a duration * intensity anaerobic effort, the calculation of each component "aerobic" "anaerobic" is the most important. One has to look at each value and assess whether they achieve the objective, regardless of the workout label.

    Remember, "tempo" is a label that will be used as a catch-all for "not-quite-right" threshold and above workouts.

    When you look at the graph below (posted above), you can see an illustration of impact of a warm up on the accumulation of EPOC. Yes, it is possible that the contribution of WU or CD will be noticeable. When I ride indoor, I monitor my TE data fields. At best, I have seen 20mn WU (Z2/Z3 power zones) contribute about half a point to the aerobic and Coold Downs (10mn Z1/Z2) contribute a tenth of a point. Your mileage will vary.

    Microsoft Word - EPOC_based_training_effect_assesment_REVISED 2.doc (firstbeat.com)

    After a workout, regardless of the label, the load will be distributed among "low aerobic", "high aerobic" and "anaerobic". I checked and it is not obviously linked to the value of the 2 TE components. So, the training focus will always reflect the nature of the workout. The exercise load will use the workout label, but the colors in that graph are less useful anyway. The training focus is the one that helps plan the training, and the TE numbers give you the benefit that you got out of it:

    Microsoft Word - EPOC_based_training_effect_assesment_REVISED 2.doc (firstbeat.com)

  • Very interesting ! I'm going to go deeper in that document. Thanks for the insights !!