Any idea why Strava would display a different average power for a given ride than those shown in Connect and Training Peaks?

All the data is from the same power meter, uploaded from the same Garmin and synced across the 3 apps.

It’s not just a bit different either, 45W out.

Not a big deal since I use TP for the ‘important’ stuff, just curious.

Different criteria for autostop?

^ Probably this. How does average speed compare across the rides? Strava gives me much higher averages because it ignores everything below about 3 or 4 kph.

Average and max speeds identical.

It’s taken from a road race yesterday so no stopping, just 2 minutes difference between moving and elapsed times which would have been while I was sat waiting to get moving.

Could it be calculating it over the elapsed time rather than moving? That would be pretty daft seeing as it doesn’t calculate speed that way (as far as I’m aware)

Everyone else uses NP and strava uses their idea of average. Difference between can be huge, especially if the ride is full of big efforts.

NP was less than 5 watts higher than my average.

Can’t understand why they’d do it any differently tbh…

It’s something to do with zero values I reckon, strava AP and NP nearly always lower compared to Garmin & TP, less difference for me when AP and NP are similar (usually after a flat TT) It is a bit irritating though.