Let me put it this way… There really is no substantial difference between a focus duration of 1 hour and 3 hours (they’re both 100% pure LOW XSS workouts). The next best options are to filter either by duration and/or difficulty.
But there is a substantial difference between a focus duration of 20 minutes and 1 hour, but there is no way to filter out the sub 1 hour duration workouts.
In terms of the math, there’s actually not a large difference in the contribution of High/Peak systems between 20 min power and 1+ hour power (the difference between those two power outputs are generally rather small when looking at the power-duration relationship). This is why we decided to move all activities with a longer focus duration into the Endurance categorization.
A focus duration of 20 min and focus duration of 60 min are virtually indifferent in terms of the % breakdown of Low/High/Peak strain. Both require a overwhelmingly large contribution of low XSS, so if you’re looking to do a workout with a focus duration between those, you can simply do any workout that is predominantly low XSS (that is, any workout with endurance focus 
) . Hope that makes sense why we did it this way!
So you are saying that there is little difference mathematically between the % breakdown of Low/High/Peak strain between for focus durations greater than 20 minutes?
This sounds right, but there is a huge physiological difference between them.
Given the idea that VT1 may be considered the binary threshold for HRV/ANS recovery it strikes me that the XERT model has no way of accounting for this. Perhaps splitting Low Strain into one category below LTP and another between LTP and TP could work?
You’re exactly right! The model doesn’t yet account for this difference. I think there still remains some work to be done here, and I could see if going a few different ways…
- there is contribution from High (and even tiny, tiny amounts of Peak) systems at sub-Threshold intensities. This might explain why Sweet Spot workouts can be effective for some athletes
or
- Low strain can be further subdivided into Low-A (above LT1/VT1) and a Low-B (below LT1/VT1).
What do you think? Perhaps I missed another option?
I can see benefits to both approaches.
As I see it #1. may be a way of tracking things based along the lines of how you calculate CHO and fat usage. Muscle fibre type recruitment kind of stuff. It may provide a lot of new information especially about the different recovery and improvement patterns that people with high or low HIE values experience. The data would all be derived though, so there would be a lot of room for error.
Approach #2. would give data that would be more reliable and it would allow you to see the different results achieved from time spent in Low-A and Low-B. Throw in daily HRV measures and I reckon this approach may prove to be very powerful.
I’d also lean towards option 2 (but would switch A and B, to have A at the bottom…?)
I do think once you also (presumably) start tracking loads for A and B you’ll get even more insight into recovery - it’s much easier to recover from a ride below VT1 than tempo / sweet spot.
Not sure how that would work in practice e.g. would you have another signature parameter? or just be defining say 90% of LTP as VT1 for training load and workout intensity purposes… or maybe instead of LTP being the artefact resulting from a signature defined by (TP, HIE, PP), it’s that TP is the artefact resulting from a signature defined by (LTP, HIE, PP)…? or maybe it makes no difference as I obviously don’t know the math behind it!
1 Like