Data and techie stuff thread for BAFS 2016

[Vicegrip]

Is not a big thing all in all but can we find Mark P and get a elevation entry cleanup? I am missing my Fav chart.

[Raymo853]

Got a tech question. Is there a way to force a refresh of the data? There is a weird and incorrect outlier in the elevation data. Mark P had a ride that showed a very high and false elevation gain that was corrected later. (How was it corrected? Stava runs such elevation profile fixing algorithms in the background. Using digital elevation models in place of GPS or smart phone collection data.)

He is shown in the BAFS2016 records as having 123,889 feet of gain over just 152 miles. Or an average of about 15% grade the whole time. His own Strava profile does not show that, but only a gain of 4,524 feet.

[ATTACH=CONFIG]10428[/ATTACH]

[hozn]

The data should automatically refresh (every night?). How long ago was this corrected? It may be that our “is-it-different?” algorithm is not accounting for changes in elevation data only.

[Vicegrip]

Been orbiting all weekend.

[hozn]

I will clear it out!

[Raymo853]

@hozn 131785 wrote:

The data should automatically refresh (every night?). How long ago was this corrected? It may be that our “is-it-different?” algorithm is not accounting for changes in elevation data only.

He has been an outlier on the graph for many days. At least since last week.

[jwetzel]

I see that the FS site reports avg temp for my rides. Is there an easy way to calculate and report (perhaps under pointless) an average temperature for all of a users rides? Give each rider an average riding temperature?

[Vicegrip]

@hozn 131801 wrote:

I will clear it out!

You rock, Thanks.

Hmmm What is the chance that you can set Eric and Subby to report only 3/4 actual miles? I might be able to hook you up with some babysitting via my now certified babysitter daughter.

[hozn]

Done & done.

[Vicegrip]

@hozn 131801 wrote:

I will clear it out!

Major Tom is back. Will give him credit he has gained even more altitude this ride.

[hozn]

I probably need a bounds check on the import to just reject unrealistic elevations.

[hozn]

I am (slowly) building a new user interface for FS and was playing around with a new leaderboard idea that visually broke down the contribution by team members. (Note: this is not current data.)

I don’t know that I’ll actually implement this, as the only way I’ve found to achieve this using Highcharts is to create a massive matrix of athletes and their contributes to *each* team. In practice this means there are 265 data series but for any one “category” (the teams) most of those series have a value of “0”. This has some side effects when you hover over to see who is accruing the points.

Anyway, I thought it was interesting. Perhaps most interesting was simply that the data seemed less notable than I might have expected (I thought there might be bigger discrepancies between contributions of the “main players”).

Also the static image is not as useful, as the one I have where I can hover over to see that, for example that Jamie W is a disproportionately large contributor to Team 8’s points.

I have another concept I’d like to incorporate related to how “hot” (or “cold”) someone is riding — compared to their typical point accrual patterns. E.g. that Jamie W guy is riding a lot less this week than previous weeks; that might be information worthy of a visualization (table) on the homepage.

Finally, I’ll probably make a development version of this interface available (maybe beta.freezingsaddles.com) for commentary on these other visualization ideas.

[dkel]

Ouch! What did Jamie ever do to you?

[hozn]

@dkel 133113 wrote:

Ouch! What did Jamie ever do to you?

Maybe bad example I am sure Jamie is riding a little less this week …. but then aren’t we all.

[Rod Smith]

Good I don’t show in the snowiest ride leaderboard. It wouldn’t be fair for the others.

[Author]

Posts