Data’s drawbacks: The best way to commute is not always the fastest

July 10, 2014

What’s the fastest way to get from point A to point B in a major American city? It’s probably a bike, according to a new joint project of the MIT Media Lab and the Social Computing Group (also at MIT). In the You Are Here data visualization, interactive maps show the fastest way to get from wherever you are in any one of 12 cities to other parts of the city. It’s a really fascinating way to visualize how cities in the US are laid out, and possibly has policy implications in helping to understand how important bicycles and associated infrastructure can and should be to the future of urban transportation.

DC Manhattan Brooklyn SF(Fastest transport mode by color: red is cars, yellow bikes, blue transit, and green walking; cities clockwise from top left: Washington, D.C., Manhattan, Brooklyn, San Francisco)

However, it’s also a lesson in the drawbacks of data visualization. In this case, the data is incredibly one-dimensional, and doesn’t necessarily accurately represent the best ways to get around the city. In addition to time, people take into consideration cost, what they need to carry, and what the weather conditions are.

This is what the fastest transit options look like at an old address of mine in East Harlem in Manhattan:


You’d think that anywhere north of Central Park I’d commute by bike, in Midtown and down the center of the city I’d take the train, and anywhere on the edges of the city I’d take a cab. But this doesn’t remotely reflect the way I transported myself when I lived there. Instead, I took the subway almost everywhere, unless I was going somewhere I could park my bike inside and didn’t need to bring a lock — like my office, in prime blue territory in the middle of Midtown — or if I was coming home late at night, in which case I’d always take a cab.

Biking is by far the cheapest form of transportation here in New York. The cost of a bike plus maintenance probably runs about $600-700 in the first year (and much less after that). That’s roughly half of a $112 per month Metrocard, and several times less than keeping a car in the city or taking taxis regularly. You also don’t have to have a ton of concern for timing, since you don’t need to worry about traffic or a train that never comes.

The downsides are the risk of the bike getting stolen (necessitating carrying around an uncomfortably heavy lock, if not two), bad weather, and the danger of riding on the same streets as multiple ton vehicles whose drivers rarely have any concern for your ultimate safety. For an extra $50 a month and a few more minutes a day, you can avoid those things altogether by taking the train. That’s worth it to a lot of people. It’s worth it to me most days, and I own four bikes.

Then again, take a city like Santa Monica, where transit is rarely a great option. If I lived here, I’d probably bike everywhere, even if I owned a car, which is probably the best option for getting other places in the Los Angeles area. The weather is nice enough that neither sweat nor rain are a huge consideration in taking your bike. Even if it’s technically faster to get to the beach by car when you start on the eastern part of town, actual experience in Santa Monica tells me I’m going to spend at least 10 minutes looking for parking (and then probably pay dearly for it).


Data visualizations are really cool, and oftentimes helpful, but it’s often just as enlightening to think about what is left out as what is put in.

No comments so far

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see