Results 1 to 25 of 34
Thread: Tracking the amount you climb.
-
08-01-2018, 10:56 AM #1
Tracking the amount you climb.
Fuck off stravasshole, etc. Yeah, got it.
That said, I am curious about elevation gain in rides. I have searched and there are numerous threads about why Trailforks thinks you climb more than Strava, even when they are sharing data. But none of them attempt to answer which is more accurate.
Maybe it doesn't matter at all, but I think it's nice to know what I actually climb on rides I know so I know what to expect when I go places I haven't been. A lot of this is because I am fat, slow, and stupid, but still.
Back to my point:
- I did a new-to-me ride last week that Strava said had 2092 feet of gain but Trailforks claims 2685.
- A local loop is 2550 according to Strava, but over 3005 according to Trailforks.
- Another local loop is 3420 according to Strava but 4073 according to Trailforks.
Does anyone have any more info on this? I suppose the logical follow-up is that if Trailforks is at least consistent, then a local ride that climbs 3500 should have more or less the same amount of climbing than some ride I'm looking at that it also says has 3500, right?
- I did a new-to-me ride last week that Strava said had 2092 feet of gain but Trailforks claims 2685.
-
08-01-2018, 11:05 AM #2Registered User
- Join Date
- Oct 2007
- Posts
- 12,664
Always go with the higher number. Duh.
-
08-01-2018, 11:24 AM #3
What are you using to record? A cell phone's GPS is less accurate than a garmin or something, especially when it comes to determining elevation, and especially when you're in the mountains and forest where it can't see as many satellites, especially ones that are not right above you.
A dedicated gps usually uses a barometric pressure sensor to help determine your elevation, because with a couple of satellites above you it is pretty hard to determine your exact elevation. This is fine, except that the actual barometric pressure often changes which would result in the gps device thinking you are changing elevation.
Strava/garmin/trailforks etc can also try to correct your elevation based on topo data, but that is only as accurate as the map data and your gps track, so there is also a lot of potential for variation there.
So the answer is that your ride elevation data is never going to be "exact." I don't worry much about it and it usually seems close enough. Still, it's a little annoying when I'm on a mountain or whatever where I know the elevation and my garmin is reading hundreds of feet different. Like last weekend I was up on point 6, elevation 7940, and my garmin says I was only at 7624. Still the overall gain for the ride seemed to be in the right ballpark.
-
08-01-2018, 11:24 AM #4
My rides usually include a few trails linked together so Trailforks isn't as useful for that. I usually import or draw in tracks into other software and link them into a route if I need to get stats. I find they can all be off depending what base map is being used to pull elevations from. Accuracy of tracks loaded into TF will vary too especially if it comes from a smartphone app, and on steep terrain this can exaggerate errors. I only use my Garmin for recording rides. It overstates distance but elevation is usually about right. Never bothered comparing it to TF and I don't use strava.
-
08-01-2018, 11:25 AM #5Registered User
- Join Date
- Aug 2008
- Location
- Central VT
- Posts
- 4,808
I have riding buddies who have Garmin GPSs that link to their Strava. I’m told those are most accurate for vert and mileage.
Strava is all over the place when I’m on a group ride and we compare numbers afterwards.
Sent from my iPhone using TGR Forums
-
08-01-2018, 11:25 AM #6
What are you using to record? A cell phone's GPS is less accurate than a garmin or something, especially when it comes to determining elevation, and especially when you're in the mountains and forest where it can't see as many satellites, especially ones that are not right above you.
A dedicated gps usually uses a barometric pressure sensor to help determine your elevation, because with a couple of satellites above you it is pretty hard to calculate that accurately. This is fine, except that the actual barometric pressure changes due to temperature and weather which results in the gps device thinking you are changing elevation even if you stay in the same place.
Strava/garmin/trailforks etc can also try to correct your elevation based on topo data, but that is only as accurate as the map data and your gps track, so there is also a lot of potential for variation there.
So the answer is that your ride elevation data is never going to be "exact." I don't worry much about it and it usually seems close enough. Still, it's a little annoying when I'm on a mountain or whatever where I know the elevation and my garmin is reading hundreds of feet different. Like last weekend I was up on point 6, elevation 7940, and my garmin says I was only at 7624. Still the overall gain for the ride seemed to be in the right ballpark.Last edited by jamal; 08-01-2018 at 03:19 PM.
-
08-01-2018, 11:28 AM #7
Yes, a phone. But I have had three different phones over the last three summers and the same rides are essentially identical on each phone as tracked by Strava. My understanding is that neither Strava nor Trailforks use GPS elevation data, "and instead generate elevation for each lat/lng point in your ride." https://www.trailforks.com/help/view/90/
-
08-01-2018, 11:28 AM #8
-
08-01-2018, 11:31 AM #9
If you really want to know, here is a good blog on the discrepancies with recording devices and the data averaging that they do:
http://regex.info/blog/2015-05-09/2568
-
08-01-2018, 12:04 PM #10
Strava doesn't register a change in elevation unless it's (I believe) 30 feet or more. Trailforks registers elevation changes of (I believe) .5 feet. So if you drop the same GPS file into each program, they'll show different amounts of climbing.
The more consistently uphill the trail is, the closer the outputs will be. But particularly on flatter sections and rolling trails, trailforks registers way more vert just because a lot of those rolls aren't big enough to register on strava. With such small amounts of elevation change registering, trailforks would theoretically register a jump as a small climb (although I'm sure the accuracy of that varies based on your gps device).
I find both to be reasonably accurate compared to themselves (i.e. if I do the same ride a few times, the numbers are generally consistent). So for myself, I just track my vert with strava because it's more convenient. But if I was going to brag about it on tgr, I'd obviously use trailforks.
Sent from my SM-G950U using Tapatalk
-
08-01-2018, 12:06 PM #11
Trailforks is not even close to accurate. I've checked it against my altimeter and topo maps. They always overshoot.
-
08-01-2018, 12:08 PM #12
-
08-01-2018, 12:16 PM #13
I don't have a response to your altimeter comparison but I am not sure a 40' topo map provides the level of detail necessary to compare. Of course, this also raises the quesiton about how Trailforks uses topo maps to calculate elevation gain/loss in the small increments toast is noting.
-
08-01-2018, 12:24 PM #14
Strava auto corrects elevation to topo data unless your device has a barametric altimeter (most Garmins do). Some people turn off this correction (you can in settings) because it typically reduces your "elevation gained" on a ride. So if your friend went into the strava settings and turned this off, they'll show inflated elevation numbers
I wouldn't trust anything from trailforks.
-
08-01-2018, 12:36 PM #15Registered User
- Join Date
- Dec 2010
- Posts
- 3,936
This makes sense. One particular downhill trail has about 30 seconds of climbing midway in and the rest is downhill. Trailforks says the trail is 300' of climbing, while strava registers 30' of climbing. Sounds like trailforks is measuring every little pop, and rise, root and rock haha. I guess that adds up.
I bet according to trailforks Crabapple hits is close to a climbing trail haha.
-
08-01-2018, 01:00 PM #16
I tried using the Trailforks ride tracking feature once and it dropped the track a bunch of times. It was a route I'm unlikely to repeat so I was pretty bummed. Never again.
Toast's explanation makes a lot of sense. The TF file for the BMT trail at Snowbird, for instance, says there's over 800' of climbing. There's zero feet of actual climbing.
-
08-01-2018, 01:15 PM #17Registered User
- Join Date
- Feb 2008
- Location
- Donner Summit
- Posts
- 1,251
-
08-01-2018, 01:19 PM #18
I find the opposite.
I record exclusively with my Garmin.
Mrs Hubbs records exclusively with her iPhone.
When I compare the data from trails or loops we repetitively do together our results always vary from each other.
My garmin data recorded is very consistent to the data of the same trail on previous rides, where as her iPhone produces varied data every time on the same trail.
Some days her mileage is 15% more than me, some days 15% less. And elevation is really all over the map when she records by iPhone.
-
08-01-2018, 01:26 PM #19Registered User
- Join Date
- Dec 2009
- Posts
- 1,703
My Garmin also has settings to how often it records data (older Garmin 500), if I remember correctly, the settings are every 2sec vs every 5 or 10sec (for best battery savings)
-
08-01-2018, 01:30 PM #20
I tracked a ride over the weekend with Strava that I had last recorded 3 years ago with MyTracks on a much older phone and the total recorded vert was within about 100 ft. Virtually no difference in mileage, either.
-
08-01-2018, 01:51 PM #21
Those of you checking vs. topo, are you adding in the down and reclimb of the same vert vs the start elevation and top elevation?
I just use strava on the iPhone, and it's OK... and I know that and don't need something gnats ass.www.dpsskis.com
www.point6.com
formerly an ambassador for a few others, but the ski industry is... interesting.
Fukt: a very small amount of snow.
-
08-01-2018, 01:53 PM #22
Yesterday, I beasted out my stiffest climb of this season. I use a TomTom watch to record. It dumps to the TomTom app. The TomTom app is linked to Strava. I usually pay attention to Strava and ignore the initial output from the watch. It was 3,800 feet vs. 2,700 feet when I checked today!
3,800 sure makes me feel better about the fact that I felt like I was going to puke, but now I am just perplexed. Some of the ride was on rolling terrain on Mid Mountain and Keystone, but could that have been a 1k difference?
-
08-01-2018, 02:30 PM #23
You can likely turn on elevation correction with your TomTom app. I know you can with Garmin even if it does have an altimeter. If using strava, it's probably best to just use elevation correction across the board. If everyone did this, it would all be the same for the same ride...for the most part.
-
08-01-2018, 04:43 PM #24Registered User
- Join Date
- Aug 2008
- Location
- Central VT
- Posts
- 4,808
-
08-01-2018, 05:42 PM #25
"A man with a watch knows what time it is. A man with two watches is never sure." (link)
Bookmarks