Hiya. Anybody have an idea why I’m getting such a big discrepancy between Strava and Google Fit?

This is my morning commute.

Strava: 11.4 miles, 602 calories.
Google Fit: 9.5 miles, 350 calories.

gps resolution is spotty. I have found that my strava will actually adjust a bit after it uploads, sometimes a few tenths of a mile. look at the actual data and zoom in, chances are one of them has a section that is messed up.

you can tell that strava knows when you start and stop, whereas google fit is showing the same for total and active time, average speed is simply calculated, and calories are always bull**** so ignore them anyway. the real issue is why is the distance off by almost 2 miles. is this data from the same ride on the same day with both apps running? or is it different dates/times along the same route?

Yep, this is the same ride, same day. I actually hit the Strava record button before beginning my ride but don't do anything for Google Fit. However, they both give the correct start time of 6:53 am.

I have Strava feed into Google Fit, so at a certain point the Strava numbers will overwrite the numbers in Google Fit.

Good to know. Since I have Strava feed into Google Fit, the GF numbers are now gone, replaced by the Strava numbers. However, I'll keep this in mind for next time.

I don't really use GF that much, just thought the ~2 mile discrepancy was curious.

Your beginning/ending points (dunno what direction you were heading, the one on the lower left) aren't the same, either. Don't think that makes up to a 2 mile difference, based on the distance of the rest of the track, but should account for some of it.

I'm more curious as to the difference in Strava and a good bike computer. I'll have to test that out

My cheap wired Bell brand computer measures within a few hundreths of a mile on the same route every time, well within any minor distance variations I would expect from trip to trip. On a day I'm getting good GPS reception, it is usually within 1.5% of the GPS (I have a voice that announces distance every 2 miles, so I can look down at the trip meter).

The biggest difference I notice is that while the computer is always spot on, my GPS reception can vary from day to day. On my normal after work ride, I've had days where my outbound leg reads 8.0 and the exact same inbound leg reads 7.6, where other days they read spot on with one another.

When I'm running my gps vibrates every time I cross a mile. It happens generally within a few 10s of feet to the same positions every time. My ending mileage is always within a few 100ths of a mile. But that's only ~5 miles.

Strava is an optimist, Google is a pessimist.

I'll go with Strava -- I need all the help I can get!

HA

but seriously I wonder if Google Fit is "snap" fitting the actual data to known roads and essentially "smoothing" the GPS readings, which would definitely account for the decreased mileage. in other words, it isn't tracking when you change sides of the road or ride in a circle at a red light for example. you could check this by drawing the route in google maps or mapmyride and see which one is closer. you could also check this with an actual bike computer known to be calibrated correctly.

Hmmm, when I map the bike ride out on Google Maps, I get 12.2 miles, which makes me think going w/ Strava's 11.4 is OK.

Just some data I grabbed on my ride tonight:

At 2 miles on my GPS, my bike computer said 2.00

4 miles on GPS, bike computer was 4.04

6 miles on GPS, bike comp was 6.05

End of outbound leg at brewery, GPS was 7.79, computer 7.81.

Return leg I didn't pay attention to intervals because I didn't reset my computer, but the exact same path in reverse my GPS computed 8.06 miles, and my bike computer 7.77. I can easily chalk up 0.04 miles to variances in passing, cutting corners, exact start/stop, general inaccuracy, etc, not so much a quarter mile.

For the record, my GPS app is Cyclometer running on Windows Phone 10 and linked/uploaded to Strava

Thanks for the info. I have a Cateye Strada on my bike. I've got to remember to zero out its trip setting sometime and compare it to my Strava/Fit results.

I've got two other bikes with Cateye Velos I found on clearance at a LBS, they work well. The best advice I can give, though, measure your tire's actual size, don't use the built in settings. My touring bike is still setup with a default size, it is nowhere near as accurate as the other three bikes I have computers on.

I took a slightly different route today and GF/Strava were within a tenth of a mile of each other. Go figure.

Resurrecting a slightly old thread, but I started using a Cateye Stealth evo+ for indoor riding a few weeks ago and noticed that after uploading to Strava there would be big discrepancies between what was on the computer and what would show up on Strava. Just to use one ride as an example:

Computer (and Cateye Sync):
Distance: 18.53km
Time: 0:33:17

Strava:
Distance: 10.8km
Time: 0:18:22

Seems to happen quite frequently that there's a discrepancy between both time and distance, and oddly, when that happens, Strava shows a max speed that is lower than my average speed (e.g. average speed 33.6kph, max speed 15.8kph).

I've contacted Cateye and they say it has something to do with my computer repeatedly searching for a GPS signal. When Strava detects a GPS signal associated with the data, but you flag the ride as a "Stationary Trainer" ride, it ignore the GPS data entirely, only keeping those data points that don't have a GPS signal associated with them.

I've contacted Strava to ask them about this; will let you guys know what I find out.

Anyone else seen this before, or is it some oddity between my computer and Strava?

What I've noticed is that if Strava is high and your computer is measuring via rotations, often it's because the battery at the sensor is going out and is sending an intermittent signal even if it says it's still sending a signal when you do diagnostics. 2 different measurements when both use GPS is new to me.

I use Locus Map pro for my gps with my Galaxy smartphone. The data given and recorded is more diverse than strava. Plus it can be used for turn by turn if I upload a .gpx course into the app. Either way, when I am done and I save my recorded ride, it auto uploads to Strava. I have Locus Pro set to use my phones baro sensor for the most accurate elevation readings and yet every time it uploads to Strava, Strava recalculates and my elevation is ALWAYS lower than it is on Locus. I am not sure which to believe. For a ride the other day, I set both apps to record so I could compare the results side by side before any uploads. Locus said I had an average speed of 13.2 and Strava said my average speed was 13.7. Locus said my elevation gain was 843 feet and Strava said it was 591 feet. Go figure.

And yet some riders insist on logging their miles to the second decimal point.

You've got a problem with precision? I don't "insist" on it, that's just what was on my computer.

What good is precision if your measurements are not accurate? I know that most cycling computers read out in tenths or hundredths of a mile, but the accuracy of the number is anywhere between 1% and 5% of reality. Say you ride 50 miles. At 1% accuracy you are +/- 0.5 miles. Said another way, your real mileage is somewhere between 49.5 and 50.5 miles. At 5% its +/- 2.5 miles. All those numbers after the decimal point don't mean much unless you improve the accuracy of measurement.

Check out : About the accuracy of cyclecomputers, GPS, distance and speed measurement on a bicycle