urbanMSP

I just did a search for Strava on this site, only to find two references to it -- one by myself.

I think Strava Heat Maps would be a useful tool for evaluating where bike lanes and bike paths are needed, as well as improvements to pedestrian safety. It is based upon users uploading their rides to the Strava app (or using the Strava app directly while cycling or jogging).

Here is the Strava cycling heatmap of our core cities: http://labs.strava.com/heatmap/#12/-93. ... /blue/bike

Strava is going to get you a lot of recreational riders and a lot less commuters. I commute by bike, but do not have Strava, as it doesn't really provide any value to me.

I use it (both commuting and not) because I like to know how much I'm riding. Also for the National Bike Challenge.

I felt like I posted this link a while back, but a year ago was this story about forward-thinking city planners using Strava.

https://www.theguardian.com/lifeandstyl ... acking-app

Strava does have an option to tag your activity as a commute.

They also sell deep-dive kind of data to cities or whoever is interested in seeing this stuff at a more granular level.

I think LakeCharles is on to something though - if you're riding to exercise or to train you're going to be really eager to track every activity. If you're riding to commute, you're less likely to do so. It doesn't mean people *don't* use it for commuting, but it means that its data will disproportionately represent non-commuting uses over commuting uses. And I can tell you for a fact that most people don't ever bother changing the activity type from the default - which means that using the "commute" tag would gain you very little. Controlling for time of day and for regularity of patterns would help, but if one can't accurately classify activities as commuting or not, then using the data could result in actively bad decisions. One last point to consider is that using online tools such as Strava will exacerbate the already-significant class and race divides of the bicycle-commuting population, since it will mostly reflect wealthy, young white people's behaviors and under-represent other groups.

In theory, strava would have much finer granularity in the data for someone to make use of. That should be part of the paid data service they provide. You should be able to at least identify routes that get more used in poorer areas even when they might look minor compared to some routes in areas more popular with the strava target demographics. You could also look at trip density based on time of day to help identify commuters.

One objection to the 4-3 + bike lane conversion on Rice Street is that "nobody bikes on Rice Street". But if you look at Strava, you can see that many cyclists use Rice Street -- they are just being ignored. I doubt that many people biking down Rice are doing it for recreation, they are doing it because they need to get somewhere. At the very least, Strava provides good data to overcome objections.

Ha yes, since Strava is way more likely to under report than to over report, it can always be used to prove that a corridor is used. Interesting point, actually.