Strava heatmap updated — and improved?

[secstate]

The Strava team has posted a good discussion of the new heatmap. They’ve completely re-engineered it, making the map much easier to update in the future.

The map isn’t a complete substitute for getting the opinions of local cyclists or doing other research. For instance, you’d never know from the map that Beach Drive isn’t exactly cycle-friendly during the weekdays. Still, I’ve found the heatmap to be incredibly helpful in locating bike-friendly routes in unfamiliar areas.

It includes all activities from 2015 until 9/2017. It might be nice to have an option to filter by year. Good routes go bad!

https://labs.strava.com/heatmap/#7.00/-120.90000/38.36000/hot/all

Share any discoveries here!

[bobco85]

Awesome! At last, the new Anacostia River Trail (IMO one of the biggest pieces of bike infrastructure in the past few years as it makes the popular Anacostia Tributary Trail System fully continuous) shows up! I’ve been waiting for an update to the heatmap, so it’s really nice to see newer routes show up.

[lordofthemark]

Someone has been doing a lot of riding to and from the building I live in since 2015. I am pretty sure that is mostly me.

OTOH I have not dominated the heat map going into Fairlington at 3oth, as more people prefer Columbus over 30th to Abingdon.

Lucky Run Trail getting some love, despite more folks taking the lane on Walter Reed.

A bunch of other things, because of the level of detail.

[chris_s]

Wow, the even distribution across the Old Town Alexandria street grid is really amazing. Goes to show how a regular grid can really distribute traffic over an area.

[PeteD]

Interesting, you can tell where all the bike commuters live.

[Steve O]

Okay, so who’s been riding their bike in the express lanes over the 14th St. bridge? Badass!

[bentbike33]

@Steve O 167256 wrote:

Okay, so who’s been riding their bike in the express lanes over the 14th St. bridge? Badass!

I don’t know, but I’m sure whoever it is has the KOM at an average speed of 58 mph or so.

[Judd]

@Steve O 167256 wrote:

Okay, so who’s been riding their bike in the express lanes over the 14th St. bridge? Badass!

About 20,000 people on this year’s DC Bike Ride. A small piece of the Whitehurst should be similarly hot.

[dkel]

@PeteD 167253 wrote:

Interesting, you can tell where all the bike commuters live.

And work. There’a a nice line that shows my preferred route for zipping through my building’s parking lot at high speed.

[Brett L.]

It’s really cool seeing one’s own impact upon the heat map. If you look close enough, you can see individual rides, let alone the route I take every single day. Hot spot into my building that I guarantee only two cyclists ride.

[jabberwocky]

@Judd 167261 wrote:

About 20,000 people on this year’s DC Bike Ride. A small piece of the Whitehurst should be similarly hot.

They really should have an option that changes the weighting based on how many days a particular road is ridden instead of just overall number of trips, which would heavily reduce the impact of single day high participation events. Just for those of us who use the heat map to judge bike friendliness of various bits of road. In that regard, knowing that 100 people ride a road 300 days a year is a lot more valuable than knowing that 30k rode a road once a year and nobody does outside that one day.

There are some MTB trails that have the same problem (Rockland Farm out here in Leesburg, for example, which is closed to the public except for the Bakers Dozen race once a year, but still shows up glowing brightly because of the sheer number of people who do that event).

The existing format is still valuable though. I’d like to just have an alternate weighting algorithm that could be switched to.

[Steve O]

@Brett L. 167287 wrote:

It’s really cool seeing one’s own impact upon the heat map. If you look close enough, you can see individual rides, let alone the route I take every single day. Hot spot into my building that I guarantee only two cyclists ride.

The green circle is me and a few of my bikey friends who have come to visit. No one else.

The purple is sjclaeys
[ATTACH=CONFIG]15673[/ATTACH]

[Judd]

@Steve O 167318 wrote:

The green circle is me and a few of my bikey friends who have come to visit. No one else.

The purple is sjclaeys
[ATTACH=CONFIG]15673[/ATTACH]

The apartment complex circle on the right has some heat from Freezing Saddles sleaze rides, sick day sleazes, gear shakedowns and the few times I was at 99 miles when I got home.

[Raymo853]

I am very impressed with it. The Strava data/gis people did a lot of good research before updating this. Strava is the first large group I have seen understand something I have been worried about for years, smart phone’s enhanced location. Many smartphones enhance your location by snapping to known geographic features. That is great for finding an address in a city with poor sight lines to the GPS constellation, but horrible when you want to analyze the data. It also gives many people, including GIS professionals, a false impression of how accurate smartphones are at calculating location. I have seen many propose using smartphones as survey instruments in the rural parts of the World and defended the choice by citing how accurately their iPhone shows their location while standing at the corner of 14th and F St NW.

I did find one funny thing, not a problem with their work but I suspect someone’s device. There is this very long ride trace in the extremely isolated high arid regions of Yemen. No way did anyone ride here: the cliffs are tall and steep, the closest town, road, or evidence of people is 10’s of km away, the nearest water source even longer away. Probably in the city of Thamud.

[ATTACH=CONFIG]15682[/ATTACH]

Here is the location in the heatmap:
https://labs.strava.com/heatmap/#11.36/49.36777/17.18610/hot/ride

Here it is in the OSM:
https://www.openstreetmap.org/search?query=17.2358%2C%2049.3751#map=13/17.2659/49.3980&layers=C

[dbb]

You can clearly see the Bike DC earlier this year, particularly the turnaround on 395 south of the Pentagon

[Author]

Posts