In the Progression view with XSS as part of the graphic I cannot understand how the training load would remain relatively constant when the Xert stress attributed to different rides varies quite a bit. Perhaps understanding how the training load is determined (the underlying variables and formula) would help.
It’s a rolling average… more details here.
http://baronbiosys.com/glossary/training-load/. There are formulas out there on the web somewhere… I used to calculate it in an Excel spreadsheet.
This doesn’t explain everything in detail (particularly the rolling average part) but it’s good place to start.
What is the interval of activities over time that is included to make up the average? I would suppose that this interval is physiologically significant. That said, in what way?
Typically 42 days, but you can configure this in your profile. It’s an exponential decay such that workouts farther in the past have less influence than more recent ones.
There’s also recovery load, which uses the same formula but with different constants. It’s a typically 7 day rolling exponentially weighted average.
You can search for “coggan chronic training load” and find lots of material. There are whole books on training with a power meter.
Yes. I have read many of these books.
If I understand correctly, you are saying that all of the XSS stress is averaged over the previous 42 days worth of activiites to form the training load. Furthermore, there is some formula attached to these that lessen the impact upon the average the further they are in the past. I surmise that the accumulated fatigue is recognized and factored into projections of future performance and the recovery process is modeled on a 7 day period with similar variable changes based on time from exposure to the initial load (as you said).
It’s not quite as simple as that: In the ‘classic’ PMC, as used in WKO, the exponential weighting is taken as 42 for CTL and 7 for ATL. Here are the equations:
Today’s CTL = Yesterday’s CTL + (Today’s TSS - Yesterday’s CTL) / 42
Today’s ATL = Yesterday’s ATL + (Today’s TSS - Yesterday’s ATL) / 7
Xert does something similar but splits XSS into Low, High and Peak and uses different exponential weighting for each:
It then adds them back together to get the overall training and recovery loads.
Thank you Mike.