Indoor trainer - Garmin EDGE 810 + Strava = BS
 

Running into a bit of an issue and there isn't a reasonable explanation for it.

Equipment used:

Minoura FG540 Hybrid Roller
Garmin EDGE 810 with GPS turned OFF
Garmin Speed Sensor on rear hub
Apple iPhone 6S 11.0.1
Garmin Connect App
Strava App

Fairly new to using an indoor trainer but have figured out enough to get everything working. Garmin 810 has the GPS OFF for an indoor ride. It was manually turned OFF by me before the ride started. Garmin Speed Sensor is registering and distance and speed are both registering.

Proceed to do a 40 minute interval training ride. The wheel is turning on the roller, and the sensor is registering the speed and distance for the entire ride. It was a good ride, and my average speed was 18.46 mph on the Garmin. The distance was 12.4 miles. Using the Garmin Connect app, the ride was uploaded from my Garmin to Strava. Everything is fine, until I look at Strava. Upon inspection, Strava has me at 12.3 miles, and an average of 17.3 mph. This is disappointing to say the least.

Why is there such a difference in the average speed, and sometimes the max speed? I had posted before regarding a friend using the app on a phone vs me using the Garmin and uploading to Strava. It was decided that differences would be there due to how the GPS data is recorded and processed by the phone, the Garmin app possibly, and Strava. OK, that part is fully understandable. What is not understandable is why a ride with the GPS OFF is showing differently on Strava than what was recorded on the Garmin. If the GPS is OFF, why isn't it transferred over as is, and if it is being manipulated, why?

This year I am keeping a running total of all of my rides, trainer, fun, commutes, every single mile is being recorded. At this point, Strava has not proven itself accurate enough for me to keep my records. Because the GPS is what I am using on the road/trainer, that is what I am going with. After searching a bit online, I see that people have created documents to keep in the cloud that will track everything from miles to HR and cadence, etc. Strava is fine for the social part, maybe those who are not nit picky about how their miles transfer over, etc. The stats part of it has been fun so far, and seeing such a big discrepancy is a huge turn off.

Strava is accurate. It sounds like it may just be a question of whether the garmin was auto pausing or including zeros. Did you stop at any time during the ride on the trainer?

The distances match so the moving time is likely different on Strava vs Garmin. Likely just need to change configuration on the Garmin.

Strava was not running for this ride, only the Garmin. Strava was only used after the ride was over and the Garmin had stopped recording and the ride was saved. The ride did stop once for about 25 seconds, because my fan setup was not working well. I paused the video I was watching and the Garmin is set up to auto stop and it did stop. Once I started back up, it was right back into 20+ mph where I left off.

Strava does the exact same thing to me.

I sync rides from Polar GPS to Polar Flow and then out to both Strava and Training Peaks. The Polar GPS, Polar Flow and Training Peaks all agree but Strava consistently under reports both distance and average speed.

Saturday I rode 66.2 miles @ 14.4 MPH but Strava says it is two miles less and 0.6 MPH slower. Yesterday I rode 1 hr 58 sec on rollers at 22.2 MPH vs 21.9 MPH according to Strava. Group rides are the same - the Polar GPS head unit, Polar Flow and Training Peaks all agree with the other riders but Strava always says I rode less miles and slower than the group.

Three different Polar GPS' on three different bikes, each with their own sensors. Indoor or outdoors, GPS on or off, auto pause enabled, it doesn't matter. Strava always gets it wrong.

I've even tried stopping the ride during long store stops, starting a new ride after the store stop and then combining the two rides using the GoToes Utility but it doesn't help.

@TenSpeedV2, you could get a free Training Peaks account just to test. I'll bet that TP agrees with your GPS and Garmin Connect.


-Tim-

@TimothyH - thank you for the valuable information. It is also good to see that it is not just me losing my mind over here in frustration trying to figure out what the heck is wrong with my rides. It doesn't happen all the time, but it is enough to make me notice, and this last one just set me off. Not that I go on Strava and brag about what my ride was, but I follow a lot of faster people. This was definitely one of my better rides, average speed wise at least. So when I uploaded it, the feelings were pretty good. Then I saw what Strava said. So I uploaded a picture of the screen of my Garmin and called BS on Strava.

Training Peaks looks interesting. May give that a run and see how it is on the free end. Gathering that it is quite a bit like Strava?

That's odd. Mine are always within .1kph and the difference is just due to how Strava and Garmin calculate moving time.

For the numbers in your OP the moving time on Strava would need to be around 43min. You should be able to look at the graph of the ride and see whether there's 43 min worth of data there.

I use a Garmin 520 but don't see why that should be different.

Well, to be clear, I am not loosing my mind. Neither should you. It is a little frustrating and at first I thought my gear was bad. Now I just laugh at it.

Training Peaks is not a social platform. It is a serious training platform, very different. It will quantify the stress you put on your body, how well recovered you are from that stress, resulting improvements in fitness and help you understand the balance between the need to be fresh vs not loosing any fitness (tapering) going into a race. It is designed for a coach to be able to access your workouts. I was sending data from Polar to TP for over a year and hardly ever looked at it before I started using it seriously this past fall.

Free means no risk and I thought it might be helpful for you to determine whether it was Strava actually causing the problem or if your gear was somehow the cause of the problem.


-Tim-

Ride for today completed. Used everything normally as listed above. Incorporated the Training Peaks app/account as well.

Garmin Edge 810 - standard @ 45:00:97 & 13 miles - 17.35 mph average
Training Peaks - super accurate @ 45:00:97 & 13 miles - 17.3 mph average
Garmin Connect - very accurate @ 45:01 & 13.01 miles - 17.3 mph average
Strava - cheated me again @ 44:54 & 12.9 miles - 17.3 mph average

This was an indoor ride, GPS was turned off on the Garmin. There should be absolutely no discrepancy when uploading a non GPS based file. This is raw data, time, distance, speed, etc. It should transfer over exactly as it was recorded. Very bummed because I like using Strava and comparing to other athletes in the area and friends. However, I don't like inaccuracies within my rides and me losing out over the period of a year could be significant.

6 seconds and .1 of a mile.

That may not seem like much, but my goal for the year is 5000 miles. If I ride them, they should count. It seems like the longer the ride, the more discrepancy there is. Several long rides are planned for the year so I want the mileage from them.

that is really weird that they would mess with the mileage like that. And the speed discrepancy is strange too. I would thing strava users would be complaining about this all over the internet.

Here is an example. This was a gravel ride done Nov 24. I'm pretty sure it has something to do with time as well as distance. Strava always has the "Moving Time" a little longer than Polar or Training Peaks "Duration". Heart rate data is typically accurate though.

Polar Flow

https://www.dropbox.com/s/u12g2x52p8...polar.JPG?dl=1



The same ride synchronized from Polar Flow to Training Peaks...

https://www.dropbox.com/s/xb7vv3g1p1...24.tp.JPG?dl=1


The same ride Synchronized from Polar Flow to Strava...


https://www.dropbox.com/s/zotjq6oidf...trava.JPG?dl=1


-Tim-

So it seems like Strava truncated your ride by 7 seconds, which likely resulted in a rounding down of mileage (I think Strava just rounds down to tenths, btw).

I don't know why Strava wouldn't give you the full time - you're syncing via Garmin, right? Because Strava won't record trainer rides by itself (no GPS).

I've listed my equipment in the original post. There is no reason that the data doesn't transfer over as is. As of now, I have come to the conclusion that Strava is more of a social app that does some tracking, while the others are far more serious and accurate.

I understand wanting to hit a number for the year, I felt similar the first time I went for 5k, but there's going to be a difference in how Strava calculates moving time and distance based on how they interpret the data sent to them from Garmin. Nothing you can do about it.

I didn't sweat anything smaller than a mile difference, but if the 1/10ths of a mile are that big of a deal, set the resistance lower on your trainer and spin faster. It's not like trainer miles are real miles anyway. Part of the reason I switched to using Zwift for indoor rides. If I wanted to inflate my numbers I could spin on my rollers at 30mph and inflate my numbers that way. At least Zwift gives me a somewhat more accurate representation of real world milage based on my power.

Sure there is. I can stop using Strava and stop with their games and their data transfer idealogoy. The Garmin Connect app is a far better and accurate representation of each of my rides vs Strava. Training Peaks is unbelievable and puts Strava to shame. Strava is just short of Facebook now in the social aspect.

I meant there's nothing you can do about it on the Strava side. But whatever, it seems you've figured it out. 100 miles a week is easy. I wouldn't worry about it too much.

TP had better put Strava to shame...it is 2x the price. Already at $60 for premium I'm done paying for Strava. Their Facebook-i-ness finally pushed me passed my enough line.

Try out the 1 week Premium Trial of TP. It is much different than Strava. Paying that much for what I use it for is not worth it, so my trial ran out. You still have access to your rides and what not, but almost everything after that is for Premium users. It was good to try it out at least and see what else is available.

You have to pick a package and stick with it. I'm using RidewithGPS, zwift, the Sufferfest, Wahoo and training peaks. I get different numbers from each. I've decided that I'm going to use the RidewithGPS numbers as they are downloaded from my Wahoo Bolt for both indoors and outdoors and just rely on that. If they are off by .01% so be it. Any indoor miles are just an approximation anyway and don't equate to true effort. Change your measurement to watt-hours or joules to get a better equivalence.

I suggest picking a ride with a big discrepancy, opening the raw data in Golden Cheetah, exporting it to Excel, and coming up with your own totals.

I'm going with Stava.

What's better than photos of what someone ate for breakfast and where else can you get a guy who calls his son, dog and Quintana Roo "My children?"

Strava isn't perfect but the ability to change the world one ride title at a time makes the inaccuracies worth it. I'm heading over there right now to express my thoughts about Donald Trump through my riding. Send a PM if y'all want to follow me. This is important stuff people.

https://www.dropbox.com/s/7db5t14g2x...kudos.JPG?dl=1


-Tim-

A man with 2 watches never really knows the time.

I've noticed that they're off as well, I think Strava computes things a bit different than TP or other applications.


Now worries though on the total mileage for the year goals, because riding on the training doesn't effing count.

A Strava art image of Trump's hair or it didn't happen. :lol: