You are reading a single comment by @ewanmac and its replies.
Click here to read the full conversation.
-
Ah, ok thank you. Training peaks gives me 242w NP for the second ride, so somewhere in the middle of Strava and Stravistix. It was a social ride but the last 10 miles was a fairly hard chaingang so that will have pushed it up.
I really ought to do an FTP test but I hate doing them inside and it's hard to find somewhere long enough to do one around here. You want a gradual, long, hill - right?
dglshrn
@ewanmac
As I understand it, "weighted average power" is strava's equivalent of Normalised Power (NP).
Essentially, time spent working hard on the ride (climbs, time on the front etc.) takes more out of you than equivalent amounts of time taking it easy (in the group, descending etc.). For example consider two rides: a 20 minute effort where you ride for 5min@220w, 10min@260w, 5min@220w, and a 20minute effort where you hold a steady 240w for the entire 20minutes. Both rides would have the same "average power" but the first effort would feel considerably harder. This means that average power can be a pretty unreliable metric for how hard you were working.
To try and get around this, NP (or strava's weighted average power) use an equation that gives more weight to hard efforts than easy efforts.
In a group ride you might expect to have a lot of 'down time' interspersed with periods of higher intensity an NP of 290 is essentially telling you what the equivalent power you would have to have held over the entire workout to achieve the same 'effort', even though your actual average power is likely to be far lower.
For these reasons NP is usually a better measure of effort (I use it for calculating my FTP from TTs), and in your case may suggest that your FTP is a bit low. However, it's not a perfect metric and if you look at how the equation works you can see that it is possible to game NP with a few short and hard efforts that artificially raise NP. For most people, the only effective way of knowing what your FTP actually is, is to test it.