tk_bike said:when using NP from a ~1 hour race effort such as a crit is there a caveat of the VI of the ride to bear in mind? my peak 1 hour NP occurred during a ~3.5 hour road race 3 weeks ago.
The NP was 292 w which seems a high estimate of ftp for me, i estimate more like 280 and at the time was using 270 as my ftp, the average for this hour was only 236 w so there was a big VI (1.24).
Does the VI of the ride, or that hour at least, make this useless for estimating ftp or am i just being too conservative? although it occurred in a long race it was at the beginning of the race so i was still pretty fresh.
I'm going to try some 20 min efforts at 290 to see how do-able they are, should give me an idea i guess!
A high VI, per se, doesn't necessarily invalidate the use of normalized power for this purpose. However, the algorithm seems to be only +/- ~5% accurate (i.e., that's my guesstimate of the confidence limits), and (catch-22) you only need to rely on the algorithm when power is highly variable in the first place (since with non-variable data, you can just use the average power). What I therefore tell people is to use a higher-than-expected normalized power (intensity factor, actually) as an indication that they should reassess their functional threshold power using, e.g., whatever approach they used to establish it in the first place (just as your planning to do w/ the 20 min efforts).
Specific to your data, I wouldn't necessary change anything: your normalized power of 292 W is only 4% greater than your estimated functional threshold power of 280 W, which is w/in the apparent error-of-measurement.
