Related
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Different Strava Segment Results on device and on the web

beumont2
over 9 years ago
Hello,

I just switched from the edge 500 to the edge 520 and till now I very happy with the switch.

Today I rode a 30sek segment where I am the KOM and as I could see on the display Ive managed to ride a little bit faster going from 0:31 to 0:29. After coming home and uploading the activity I saw that the time was 0:31 again and not 0:29 as displayed on the device itself so I deleted the activity and uploaded it again but as I can see in the segment list I rode 0:31 and not 0:29 as displayed on the garmin. Any ideas why the time is different on the device and on strava?
  • 0 over 9 years ago
    This from Strava's FAQ, basically don't believe anything you see until you upload your ride...

    FAQs
    Why is my segment time different after I upload than it was on my Garmin?

    Garmin Edge devices use a different algorithm for determining when a user starts and finishes a segment than Strava uses, which can affect the elapsed time reported for the segment. To ensure consistency between users, the times shown on the Garmin during an activity are preliminary, and the times shown on Strava after uploading are final.

    ... So appart from "live segments" not actually being live, they aren't always accurate either
  • Former Member
    0 Former Member over 9 years ago
    This from Strava's FAQ, basically don't believe anything you see until you upload your ride...

    FAQs
    Why is my segment time different after I upload than it was on my Garmin?

    Garmin Edge devices use a different algorithm for determining when a user starts and finishes a segment than Strava uses, which can affect the elapsed time reported for the segment. To ensure consistency between users, the times shown on the Garmin during an activity are preliminary, and the times shown on Strava after uploading are final.

    ... So appart from "live segments" not actually being live, they aren't always accurate either


    Sounds legit.
    Last time I finished a half marathon, the results were preliminary. It would be a few hours before my results became official, and they still changed slightly in that time. (Not sure how, but it changed by a few seconds) As such, if you only beat your goal by a second or so, yeah, it needs to verify that crap. Now, if you beat the rival by, oh, 5 minutes, yeah, your good. You won't have the final actual numbers until verified, using the same system that counts these things. IE, strava uses their own formulas, garmin has a good idea how to match them, but it won't be perfect.
  • 0 over 9 years ago
    I would recommend making sure that the Edge is set to using 1 second recording when chasing segments rather than the default of "smart recording". The use of "smart recording" could lead to a few seconds difference between the times reported by Strava and the Edge depending on how Strava handles the gaps in the data and where these gaps fall in relationship to passing through the start and end gate.
  • 0 over 9 years ago
    Sounds legit.
    Last time I finished a half marathon, the results were preliminary. It would be a few hours before my results became official, and they still changed slightly in that time. (Not sure how, but it changed by a few seconds) As such, if you only beat your goal by a second or so, yeah, it needs to verify that crap. Now, if you beat the rival by, oh, 5 minutes, yeah, your good. You won't have the final actual numbers until verified, using the same system that counts these things. IE, strava uses their own formulas, garmin has a good idea how to match them, but it won't be perfect.


    I guess my point was that the average Joe sees "Strava" segments on their Garmin and the expectation would be that they are the "same" rather than Garmin interpret it one way and then the "real" Strava a different way.

    I do agree though, that the data on the website is the one that should count. It's just a shame that considering both companies went to the trouble to integrate this feature, that they couldn't use the same algorithm in both places to avoid questions and confusion such as this. I can see some people getting upset about this, demanding their Premium subscription cancelled etc etc as they seem to do.
  • 0 over 9 years ago
    Strava is the master in terms of being both the judge and jury. The Edge I think even stats that the results are provisional until confirmed by Strava if you set a new best time. I agree it would be better if both the Edge and Strava always agreed, but there are things that can be done when post processing data that are not really practical to do in a real-time environment on the Edge.
  • 0 over 9 years ago
    I think the ultimate problem is that the 520 is having to determine when you cross the start and finish in real time whereas strava has the luxury of post-processing the GPS track to figure out more accurately when it happened.

    For example, on Friday I did a segment that starts on an underpass and while it alerted me that I was nearing the segment, it was very apparent that the 520 couldn't tell I started the segment until I was quite a ways back out into the open. I am certain that after the fact it would be possible to estimate from the track into and out of the underpass the point at which I actually hit the start.

    So algorithms aside, there are reasons that the real-time determination may differ from the post-processed result. Now whether it has the processing horsepower to potentially do this while in the middle of a segment (i.e., correct the start time) is an interesting question, but I am doubtful it would be possible.
  • 0 over 9 years ago
    The problem is that both Garmin and Strava guesstimate where the start is based on the nearest GPS data point (and probably lots of other techo wizardry) so it is unlikely that they will ever match up. Strava itself never uses the exact same "line on the road" as a segment start or finish - often a KoM can be decided on data point locations rather than true "start to finish" lines, and the original segment is created from someone else's data points.

    There are many factors that can effect a devices recording and accuracy. As Alan pointed out, setting the Garmin to 1 second recording would be a good start. Using GLONASS would help too as likely to be more accurate than GPS alone.
  • Former Member
    0 Former Member over 4 years ago in reply to aweatherall

    thanks for that, just got a 4 second loss on a 23 second segment thanks to so called smart... settings changed