Confidence in AI FTP Detection up?

The varying opinions above is why we use power, and zones.

1 Like

How about the TR over-under workouts that are classed as threshold? They can easily push you into 9-10 RPE.

1 Like

RPE of 9-10 should not be overall feeling but per peak. For example, description from McAdie +2 (95% U / 105% O)

I.e. RPE is moment-to-moment feeling, just like power.

Only the stupidly stupid ones doing dumb stuff like 98%/115% :smile:

I tend to do o/u at 90%/110% and even pushing those out to, say, 3 × 18 I’ve never really got near a 9.

And if I did I’d pull the plug because the stimulus to fatigue ratio is too low at that point.

1 Like

In reality it’s gotta be both, right?

Otherwise we have no way to qualitatively assess how hard anything is.

1 Like

Right, that’s what makes RPE unreliable – it brings in unrelated axis “how long has interval been going on”. I mean, 2h vs 8h Z2 ride power should be in certain range, but not sure many people can say same for RPE for each hour.

2 Likes

Hard agree. And, of course, everyone’s RPE scale is different - either subtly or wildly.

I’m glad I have actual metrics, because I think I’m pretty lousy at assigning a number to how I’m feeling. Sometimes I can get in touch with the feeling OK, but to then label that as a “6” or a “7”? Not so much.

3 Likes

Yeah, I have pretty tuned feeling for Z2 and SS but Z4 gets somewhat hazy, O/U is simply undescribable due my personal weakness (lactate shuttling) and VO2max is suddenly clear feeling again.

1 Like

Totally with you here. Then throw in that on one day a workout might feel easy and on another day the same workout might feel hard. The post-ride RPE system (or Body Battery or readiness or whatever AI you prefer) is really hard to rely on when it comes to AI assigning a probability to your ability to complete a workout on any given day.

1 Like