another strava live segments question

The pacing is taken from the ride that the athlete did that you are racing against.

The pacing is taken from the ride that the athlete did that you are racing against.

That's what I'd assumed but it really doesn't feel like that's what's happening.

Obviously on a constant flat/gradient segment it wouldn't be noticeable but one specific one I did I was nearly a minute ahead but finished nearly a minute behind.

you can do a compare on strava and I was never 30 seconds ahead on the flat..I was about 4 seconds in front - so I sat back :(

https://www.strava.com/segments/1129787

(my ride is set to private so you can't do a compare)

I can see it being a second or two out out but not 30 seconds. Not the only time I've seen this

also doesn't it say "virtual partner" finished at the end NOT "virtual racer"

partner = average speed
racer = actual gpx

I fairly convinced that I'm following an average not a previous ride

if so makes it a bit useless on varying grade segments

I can assure you that it is using the pacing from the activity for that segment. If you unpack the segment file you can see this.

Virtual partner is just a generic term.

If you are importing a gpx file then that is not the same as using a segment file delivered by Strava unless the gpx file has been built using the pacing from the activity.

Sorry for the confusion. I mean Gpx as apposed to average pace. I'm obviously importing the segments via strava > garmin connect.

I can understand id see this issue if either the KOM or my garmin has a bad fix - but then that would show up on the compare function in strava.

I'm really not convinced that this works as you suggest. The garmin website has exactly 6 useless articles about it and strava has nothing useful at all.

I'm not disputing that the segment file on there 520 contains that position/time data but that doesn't mean that the time ahead/behind actually uses it.

Fwiw I'm on 6.20 - is this a preemptive change so they can allow the 520 to follow a course as the same time as use strava live segments?

I'm pretty sure I see this behaviour on other segments too - racing my own times. I'm normally fairly consistent and so there should be little variance from the previous times. Try it on a segment that gets much harder or much easier. Racing against your own time.

I see no other reason why the strava compare segment function would be so radially different from the data that was showing on the garmin during the ride ?

Let me collect some data against some segments here, and see what I observe.

thanks Alan

I'm in central london and most segments are pretty flat - so it's hard to tell. You really need to do this test on a longer segment with a single, severe gradient change halfway. Certainly what I saw matches more or less exactly what I'd expect if the garmin was taking an average speed of the KoM ride

If I get chance I'll try to find something similar this weekend.


here's the segment is was pretty obvious on.

KoM average speed = 24.2km/h
my average speed = 19.6km/h

I was doing about 33 km/h for the first half before it kicked up to 15%
the 520 had me 30 seconds ahead at half way
strava compare has me neck and neck for the first half
no gps glitch as this would be shown in strava

All these figures would be correct if I was racing against a constant 24km/h rider not the actual KoM pace/postion data

MUCH harder to see on a flat segment