I’d love to see a more dynamic load calculation than TSS. Maybe even one unique to this platform.
TSS is hugely flawed when an activity includes work above FTP, and ignores the context for that work. For example, you could do a steady state hour at 50% FTP and generate 50 TSS, or you could do high intensity intervals for 30 minutes, then spin really easily for 30 minutes, and also generate 50 TSS. Hiit intervals right after a VO2 Max effort are also more taxing that at the beginning of the ride, and rest period infleunces it too. However, TSS simply looks at overall NP against time.
Xert have improved this significantly with XSS - for example it is possible to generate 150 training load in an hour if you are doing significant amounts of work above threshold.
Why is this a problem? With TSS it is difficult to accumulate load when the focus is on shorter high intensity sessions - when the actual physical load could be significant. This can lead to TSS chasing which generally means working for as long as possible in the grey areas of z3/4 - which it is generally suggested is best avoided to generate effective / efficient adaptations.
I am not sure exactly how to solve it without trying to directly copy Xert, but perhaps multiple calculations of NP power could be taken (say NP in 5 minute chunks) to get a better view of load across an activity.
I would need to be working from a published paper to do something different. BikeScore (Skiba) has already been suggested on this forum as an alternative. TSS certainly has limitations but it also has a huge ecosystem behind it. Lots of people joining Intervals.icu look very closely at the training loads and want a very close match with other tools, which is understandable.
I think the whole polarised training thing helps with TSS chasing.
I think TSS also overestimates longer/light workouts and underestimates shorter/hard workouts. It also might be impossible to find a model that works with all kind of workouts flawlessly. E.g. I can run downhill with relatively low HR and decent pace while the euro-muscular low is very high (and thus I need two days recovery). Stryd tries to fix this with power even though in mountain running it has its own problems…
