Strava changed uploaded data
Discussion
I've noticed this for a while, I use a Wahoo Roam and upload that data to Strava.
But Strava then changes that data?
I've had more dramatic examples of this, but I've just been out for a quick lunchtime spin.
Wahoo logged distance as 40.18km, Strava changed that to 40.05km
Elevation, same
Total time, same
Ride time, Wahoo logged as 1hr 5mins 44secs , Strava changed that to 1hr 6mins 3 secs
Speed Wahoo logged at 36.7kph, Strava changed that to 36.4kph
I'm not overly arsed, but it does seem strange as it's all using the same data?
But Strava then changes that data?
I've had more dramatic examples of this, but I've just been out for a quick lunchtime spin.
Wahoo logged distance as 40.18km, Strava changed that to 40.05km
Elevation, same
Total time, same
Ride time, Wahoo logged as 1hr 5mins 44secs , Strava changed that to 1hr 6mins 3 secs
Speed Wahoo logged at 36.7kph, Strava changed that to 36.4kph
I'm not overly arsed, but it does seem strange as it's all using the same data?
I don’t know the answer but some apps/systems convert mph to kph and may round up and down at the same time. Or may use more or less decimal places.
I know on mph cars the ecu will show in km but won’t be an exact conversion for that reason.
The timing , possibly have different definitions of when a ride is paused, or you are stopped.
I know on mph cars the ecu will show in km but won’t be an exact conversion for that reason.
The timing , possibly have different definitions of when a ride is paused, or you are stopped.
I regularly get "docked" a handful of yards by Strava, so I've learned to go a little over the distance I want to see on my feed, then crop it after I upload it.
It's about 0.01 miles missing after a typical run or walk, and anywhere between 0.02 and 0.05 miles after a ride. But then it goes from my old Garmin 910 watch into Garmin Express. That sends it to Garmin Connect, which syncs with Strava, so a couple of stages where any kind of data conversion could trim a little distance from an activity.
It only bothers me when I'm aiming for "aesthetically pleasing" numbers on an activity. I like things like 1.81, 38.83, 21.15, etc, mainly because I'm "a bit odd". Running an exact distance then getting it trimmed irritates, but so does cropping an activity after upload. It's just the way the tech works, though, so don't take it too seriously.
I thought It was a tech issue when my 2019 'RideLondon' came up as only 99.2 miles on my Strava upload. Turns out that one may have been down to a route change because Admiralty Arch was closed with scaffolding up in the roadway. Annoying, because I'm a bit of a distance cyclist, and if I ever get my Eddington Score near 100, it will mean having to ride an extra 100 miler to qualify, when I'd already ridden about 8 miles to get to the start but had reset my watch before crossing the startline at the event itself.
It's about 0.01 miles missing after a typical run or walk, and anywhere between 0.02 and 0.05 miles after a ride. But then it goes from my old Garmin 910 watch into Garmin Express. That sends it to Garmin Connect, which syncs with Strava, so a couple of stages where any kind of data conversion could trim a little distance from an activity.
It only bothers me when I'm aiming for "aesthetically pleasing" numbers on an activity. I like things like 1.81, 38.83, 21.15, etc, mainly because I'm "a bit odd". Running an exact distance then getting it trimmed irritates, but so does cropping an activity after upload. It's just the way the tech works, though, so don't take it too seriously.
I thought It was a tech issue when my 2019 'RideLondon' came up as only 99.2 miles on my Strava upload. Turns out that one may have been down to a route change because Admiralty Arch was closed with scaffolding up in the roadway. Annoying, because I'm a bit of a distance cyclist, and if I ever get my Eddington Score near 100, it will mean having to ride an extra 100 miler to qualify, when I'd already ridden about 8 miles to get to the start but had reset my watch before crossing the startline at the event itself.
Possibly how Strava handles zero speed sections. The wahoo will pause after a couple of seconds whereas Strava May have a different cutoff. This will affect the averages.
If I use my old polar v650 which doesn’t auto pause during the ride I get a very nice bump in average speed when I upload to Strava :-)
Do you use a wheel speed sensor? May give different results vs gps trace
If I use my old polar v650 which doesn’t auto pause during the ride I get a very nice bump in average speed when I upload to Strava :-)
Do you use a wheel speed sensor? May give different results vs gps trace
It’s all due to how Strava calculate distance based on the GPS track points in the data file, and how they prioritise information from different sensors.
Explained here: https://support.strava.com/hc/en-us/articles/21691...
Explained here: https://support.strava.com/hc/en-us/articles/21691...
Edited by CheesecakeRunner on Monday 15th May 18:11
Strava can be quite erratic at times, I'm just using it on a phone rather than uploading from a dedicated device so GPS accuracy (which isn't strava's fault) is part of the issue, but still...
I've frequently seen around 20% difference in distance & height gain between different people's strava recordings while riding together, even with no obvious outlier points when comparing the recorded route.
I've seen mine show a logged route fairly accurately on the preview map before uploading, then strava dumped most of the data and converted it to a straight line when it uploaded. Their helpdesk just said 'the gps data must have been flagged as inaccurate and rejected' but once it was uploaded there was no way to revert to the actual recorded data. Quite annoying as I wanted to record the position of an unmapped track to sync to Trailforks.
I've frequently seen around 20% difference in distance & height gain between different people's strava recordings while riding together, even with no obvious outlier points when comparing the recorded route.
I've seen mine show a logged route fairly accurately on the preview map before uploading, then strava dumped most of the data and converted it to a straight line when it uploaded. Their helpdesk just said 'the gps data must have been flagged as inaccurate and rejected' but once it was uploaded there was no way to revert to the actual recorded data. Quite annoying as I wanted to record the position of an unmapped track to sync to Trailforks.
joropug said:
I don’t know the answer but some apps/systems convert mph to kph and may round up and down at the same time. Or may use more or less decimal places.
I know on mph cars the ecu will show in km but won’t be an exact conversion for that reason.
The timing , possibly have different definitions of when a ride is paused, or you are stopped.
All KPH here!I know on mph cars the ecu will show in km but won’t be an exact conversion for that reason.
The timing , possibly have different definitions of when a ride is paused, or you are stopped.
lufbramatt said:
Possibly how Strava handles zero speed sections. The wahoo will pause after a couple of seconds whereas Strava May have a different cutoff. This will affect the averages.
If I use my old polar v650 which doesn’t auto pause during the ride I get a very nice bump in average speed when I upload to Strava :-)
Do you use a wheel speed sensor? May give different results vs gps trace
Just GPS trace.If I use my old polar v650 which doesn’t auto pause during the ride I get a very nice bump in average speed when I upload to Strava :-)
Do you use a wheel speed sensor? May give different results vs gps trace
Strava filters the data so that it looks nice on a map - as with most measurements, the GPS data from your device will only be accurate to up to a certain percentage and would look crap if used in its raw form. They explain it here:
https://support.strava.com/hc/en-us/articles/11500...
Interestingly they say they try to use the overall distance from the device file.
https://support.strava.com/hc/en-us/articles/11500...
Interestingly they say they try to use the overall distance from the device file.
ecs said:
Strava filters the data so that it looks nice on a map - as with most measurements, the GPS data from your device will only be accurate to up to a certain percentage and would look crap if used in its raw form. They explain it here:
https://support.strava.com/hc/en-us/articles/11500...
Interestingly they say they try to use the overall distance from the device file.
If you ever do a cyclocross race you really notice this (especially my local East kent league- "why use two corners when 8 will do") https://support.strava.com/hc/en-us/articles/11500...
Interestingly they say they try to use the overall distance from the device file.
ecs said:
Strava filters the data so that it looks nice on a map - as with most measurements, the GPS data from your device will only be accurate to up to a certain percentage and would look crap if used in its raw form. They explain it here:
https://support.strava.com/hc/en-us/articles/11500...
Interestingly they say they try to use the overall distance from the device file.
That still doesn't really explain it, but 'Strava'. Westcoast mafia ;-)https://support.strava.com/hc/en-us/articles/11500...
Interestingly they say they try to use the overall distance from the device file.
Julian Scott said:
Yes, for some reason it apparently calculates power rather than just taking the actual power data?
Anything like this will be just a discrepancy of settings between devices. I set any Garmin to records data every second vs smart (perhaps wahoo have this setting too?) and get rid of autopause and include zeros in everything. I’ve never had any mismatch in metrics that matter but obviously my average speed in the unit is always lower than when it gets to strava as it scrubs out the stationary s
t. But when doing time trials for example there was always an exact matchokgo said:
Julian Scott said:
Yes, for some reason it apparently calculates power rather than just taking the actual power data?
Anything like this will be just a discrepancy of settings between devices. I set any Garmin to records data every second vs smart (perhaps wahoo have this setting too?) and get rid of autopause and include zeros in everything. I’ve never had any mismatch in metrics that matter but obviously my average speed in the unit is always lower than when it gets to strava as it scrubs out the stationary st. But when doing time trials for example there was always an exact matchHow can strava calculate power if it’s taking data from another device? Assuming we are talking average here not weighted average.
It does make estimates if there’s no power data in the file, but not sure what calculation it’s making in terms of ‘average power’ with a wahoo/Garmin file with a power meter? (Their weighted average is something else and obviously their version of NP but that’s another story). Average should be the same as your unit all settings being equal.
In answer to your OP - while we see the elapsed time is the same, do we know the difference between when the devices stop registering movement? Do you have autopause set?
It does make estimates if there’s no power data in the file, but not sure what calculation it’s making in terms of ‘average power’ with a wahoo/Garmin file with a power meter? (Their weighted average is something else and obviously their version of NP but that’s another story). Average should be the same as your unit all settings being equal.
In answer to your OP - while we see the elapsed time is the same, do we know the difference between when the devices stop registering movement? Do you have autopause set?
Edited by okgo on Wednesday 17th May 16:09
okgo said:
How can strava calculate power if it’s taking data from another device? Assuming we are talking average here not weighted average.
It does make estimates if there’s no power data in the file, but not sure what calculation it’s making in terms of ‘average power’ with a wahoo/Garmin file with a power meter? (Their weighted average is something else and obviously their version of NP but that’s another story). Average should be the same as your unit all settings being equal.
In answer to your OP - while we see the elapsed time is the same, do we know the difference between when the devices stop registering movement? Do you have autopause set?
No idea. I'm not a power obsessive.It does make estimates if there’s no power data in the file, but not sure what calculation it’s making in terms of ‘average power’ with a wahoo/Garmin file with a power meter? (Their weighted average is something else and obviously their version of NP but that’s another story). Average should be the same as your unit all settings being equal.
In answer to your OP - while we see the elapsed time is the same, do we know the difference between when the devices stop registering movement? Do you have autopause set?
Edited by okgo on Wednesday 17th May 16:09
Gassing Station | Pedal Powered | Top of Page | What's New | My Stuff