More Than One Slippery Slope

Ornoth’s been playing with data visualizations again, and as usual the results are pretty cool.

Climbing hills is how cyclists measure themselves. We roam around the countryside, testing ourselves against short, steep hills; long, steady hills; and especially ones that are both steep and long.

Ascending each hill dozens of times, we become intimately familiar with every detail, having discovered where the slope increases, where the opportunities to recover are, and whether a rider should attack it aggressively or work his way to the top more slowly and conservatively.

When cyclists get together, hills are a natural topic of conversation: complaining about them, reminiscing about them, and comparing them to one another. This hill is longer, but that one’s steeper. But the first one is steeper right at the start. Or is it?

The one thing that’s missing from our conversations is quantitative data that allow you to objectively compare one hill with another, or even a whole set of hills. Ideally, that data would all be summarized in one simple chart that you could read at a glance.

You’d think the interwebs would have created such a thing, but I couldn’t find one. Tons of sites will show the elevation profile of one hill, but I couldn’t find any that would show multiple hills on the same chart. So I went and wrote one myself.

If you go to this page, you can enter the URLs for up to thirteen Strava “segments”. The easiest way for me to identify hills (or any road segments) is by leveraging Strava, the cycling activity tracking site.

Once you’ve told me what road segments you’re interested in, behind the scenes my page will fetch all the elevation data from Strava, then build a chart for you that displays the elevation profiles of every segment.

If you click on the thumbnail image at the top of this post, you can view a full-size example, although it won’t show the interactive features of the chart: you can hover the mouse over any line, and a tooltip will display the slope of the hill at that point; you can show and hide each segment; and zoom in closer to see greater detail.

My only disappointment is that it’s only as good as Strava’s data, which isn’t always as good as you’d want and expect.

It can be a bit of a chore chasing around Strava to find segment URLs, so I’ve created some example charts for you to play with.

The first one compares some noteworthy hills near Boston.

The next one shows the thirteen hills in Pittsburgh’s Dirty Dozen ride.

In addition to comparing local hills, this makes it easier for me to compare Boston’s hills with those in Pittsburgh, both to satisfy my own curiosity as well as to share with my cycling buddies back in Boston. Here’s an example chart comparing some hills from Boston and Pittsburgh.

But to satisfy your own curiosity, go to the input page to use whatever Strava segments you care about, from your neighborhood or anywhere in the world.

I hope you enjoy it! It was fun to develop, and I think it carries really interesting and useful information that no other site provides.