Strava’s IMBA Challenge

Our Community Forums Events Strava’s IMBA Challenge

Viewing 8 posts - 16 through 23 (of 23 total)
  • Author
    Posts
  • #984646
    hozn
    Participant

    @GregBain 67775 wrote:

    I think it’s fair for Strava to adjust the elevation (even if it’s less accurate) because you’re comparing yourself to other Strava users.

    It is annoying that it changes the elevation for the same route – I just checked my rides for the past 2 weeks and there is a +/- 50ft swing, which isn’t terrible, but isn’t great.

    Well, my issue is that they spend a lot of energy second-guessing Garmin and, if one is to believe the various support tickets, they make the data less accurate when they do so. I know that there are issues w/ barometric elevation readings, but their massaging algorithm seems arbitrary (I’m sure it’s not …).

    Note that the auto-correction is also different from what happens if you click the “Correct elevation data” button. That button applies the same algorithm, I assume, that is used for determining elevation when using phone GPS units. I have no idea why they auto adjust the incoming Garmin data when you don’t ask it to correct the data, though.

    Furthermore, for competitions like this I’m assuming that phone GPS users are already going to get screwed if they were to actually go off-road, though I don’t know how granular the elevation data is for the region and whether they are using that or using the known profiles of roads, etc. (I know that they have some explanation of this, but I’m not confident it hasn’t changed again and don’t care *quite* enough to go read it again :)) Mountain bike trails are extremely tight & twisty, though, so it seems unlikely that even fairly granular elevation data is going to reflect the actual elevation profile of the ride.

    In short, Strava probably shouldn’t have elevation competitions because the data is all over the place — even the same routes (I have variations much larger than 50-feet). That said, the imprecise and inaccurate nature of the data probably isn’t going to stop many people from giving the competition a go.

    #984647
    mstone
    Participant

    @GregBain 67775 wrote:

    I think it’s fair for Strava to adjust the elevation (even if it’s less accurate) because you’re comparing yourself to other Strava users.

    It is annoying that it changes the elevation for the same route – I just checked my rides for the past 2 weeks and there is a +/- 50ft swing, which isn’t terrible, but isn’t great.

    To be fair, the altimeter/barometer devices aren’t particularly accurate either. In general, this is an unsolved problem. I suspect that the consumer GPS units will improve over time more than the barometers will. (Eventually, I’d expect the retail GPS chips to implement WAAS and differential corrections as volume drives the cost down.) But it will take a while, and good vertical accuracy is currently pretty expensive.

    #984648
    jabberwocky
    Participant

    @hozn 67791 wrote:

    Up near Frederick — e.g. http://ridewithgps.com/routes/2439173 . Fantastic hills up there!

    …thats evil. I’ve ridden some of those hills (not all on one ride) and they are no joke.

    #984649
    consularrider
    Participant

    @mstone 67793 wrote:

    To be fair, the altimeter/barometer devices aren’t particularly accurate either. In general, this is an unsolved problem. I suspect that the consumer GPS units will improve over time more than the barometers will. (Eventually, I’d expect the retail GPS chips to implement WAAS and differential corrections as volume drives the cost down.) But it will take a while, and good vertical accuracy is currently pretty expensive.

    When I use the enable elevation correction on Garmin for my longer rides, I gain 25 to 50% more than was registered on my Garmin Edge 500. Shorter rides it’s more in the 5 to 10% range.

    #984655
    hozn
    Participant

    Interesting; it looks like they removed the ability to perform the manual correction. It sounds like it might happen automatically now? (https://strava.zendesk.com/entries/20965883-Elevation-for-Your-Activity)

    On a recent ride, here is what I got:

    Garmin elevantion: 1,962 ft
    Strava elevation: 1512ft

    I have typically found that Strava corrections are lower than my Garmin (Edge 510) records.

    #984685
    Vicegrip
    Participant

    @hozn 67791 wrote:

    Up near Frederick — e.g. http://ridewithgps.com/routes/2439173 . Fantastic hills up there!

    Nice!

    My legs just slid out of my pants, ran into the other room and hid under the coffee table. I am sensing a future weekend and it looks like it has a two scoop serving of suffer in it.

    #984702
    KLizotte
    Participant

    @DismalScientist 67758 wrote:

    Check out this thread: http://bikearlingtonforum.com/showthread.php?5588-So-why-does-Strava

    Have sent a service request to Strava to fix.

    #984706
    hozn
    Participant

    @Vicegrip 67832 wrote:

    Nice!

    My legs just slid out of my pants, ran into the other room and hid under the coffee table. I am sensing a future weekend and it looks like it has a two scoop serving of suffer in it.

    Yeah, those are some serious (category 3) climbs — and descents. I will take a day off at some point to go back up there. Really nice local-ish riding.

Viewing 8 posts - 16 through 23 (of 23 total)
  • You must be logged in to reply to this topic.