I started to use Bestbikesplit to follow a power plan for a long gravel challenge (GBC500, Qc, Canada). I used Ride with GPS to upload the course, but D+ estimated by Ride with GPS is completly innacurate and underevaluate by 20-35%. For example, a training course with 1400m D+ estimate turned out to be 2000m D+ over 100km. It makes a huge difference for the pacing strategy. Any suggestions? Thanks!
I’m a big-time Ride-with-GPS user, but you’ve identified one of the downfalls of their system. On long routes, I believe they lower the “sampling rate” for determining up-and-down. This means it misses a lot of altitude gain along the route which as we all know, has a major effect on effort.
Can you upload the GPX into Strava, allow it to correct the data, and use that as your source for Bestbikesplit? I’ve found Strava to have better altitude data 99 times out of 100 for long routes.
D+ means elevation gain??? assuming that.
I have seen RWGPS and others get elevation wrong for a ride in some conditions, notably:
- bridges over deep ravines
- roads cut through hills
- tunnels through hills/mountains
In both of these cases the roads don’t follow the surrounding terrain and the estimates for the routes use elevation data for the ‘area’, and from what I have seen don’t use road-specific elevation data.
Even for routes where they will have every single ride show that the riders stay level over the bridge, this doesn’t seem to be fed back into their elevation data to correct it.
There are likely other ways for these estimates to be off, but these are what I have seen. You can edit a route to correct elevation, but you have to know where to do that.
Ehh, in this case it would show much more elevation than what the road actually does. My experience with RWGPS is that it smooths the elevation to a point where it’s much much lower than actual.
The other year I looked at RWGPS to determine if a route was going to be good to take my new-ish to cycling wife on for her longest ride ever up that point. RWGPS said that the steepest hill was going to be a couple miles at like 6% with a tiny blip at 9%. Okay, that’s not too bad so we go on the ride.
We turn the corner and it’s like 2km at easily 10-11% average and like 500m at around 15%.
And that’s not the only time it’s done that exact same thing. Strava is usually much more accurate IMO.
My one experience with BestBikeSplit wasn’t great. It reported 25% lower elevation gain than 3 other platforms based on the same GPX. From my notes:
Plot a route says 6190’, Garmin Connect says 6220’, Strava says 6205’. When I import into BBS, it says close to 4500’
I worked with their support for a bit, but they didn’t have a great explanation.