An Interactive Display of Boston Subway Performance Data

For those who can’t get enough of charts showing the behaviour of TTC routes, I highly recommend a visit to Visualizing MBTA Data, a project of Mike Barry and Brian Card at Worcester Polytechnic Institute.

In their article, they show data for the movement of trains, for headways and for trip times between points, not to mention station usage counts. All of this can be explored interactively to view specific sections of the network.

(For more details about their project, please see this handout from a recent presentation.)

As they note, the idea of plotting train movements goes back to the 19th century. I certainly didn’t invent it. The biggest challenge has been to take masses of data from the TTC and convert them to a format that is digestible and illustrates various factors of transit operations.

An encouraging note: the TTC is now doing some of this type of analysis itself, and this will inform work on improved scheduling and better monitoring of service quality.

7 thoughts on “An Interactive Display of Boston Subway Performance Data

  1. Very elegant, especially the trip time overlays. It draws a lot on Tufte’s work in the field of Data Visualization.


  2. The link to the MBTA Meteor is also interesting and is, I presume, an official MBTA site. Of course, if the TTC were to try something like this it would be plus or minus 5 minutes and not based on headways! I note the “Entrances and Exits per Station” on the main site you reference, I assume the TTC has similar figures (from turnstiles) – do they make it public?

    Steve: The site is a subdomain of which is a software developer’s website in San Francisco, not an MBTA site, although the app will be using real time MBTA feeds.

    The TTC has turnstile data that are manually collected on, at best, a daily basis. They are not online. Of course, many entrances and exits do not occur through turnstiles thanks to our free transfer system between routes.


  3. The data visualization here is wonderful.

    It’s interesting to see that Harvard is the busiest station on the T. In addition to being a tourist hub, it has a large underground bus / trolley platform (or at least did 10 years ago, when I lived there) that serves points west and north-west of Cambridge (towns of Watertown, Belmont, Arlington, etc.)

    One thing worth noting for TTC users is that, compared to Toronto, headways on the T are enormous. Have a look at MBTA schedule frequencies for examples. On a weekend, especially early in the morning or late in the evening, waiting 15 minutes for a train to come is totally normal.


  4. That is absolutely beautiful work. You know great data visualization when you can appreciate the aesthetics of it independent of the data, but the data nonetheless shows with such clarity. And the responsiveness of the design to hover as a focus on a particular run or timeslice is amazing. It’s clean and elegant, but absolutely packed with data.



  5. Steve, why are you against PPP? The Canada Line was the best thing that ever happened to Canada. Also Waterloo is hiring a private firm to maintain, finance, build and operate their LRT.

    It’s all about trust.

    Steve: The problem is not trust, but that PPPs are abused simply to keep debt and future operating costs off of the government’s books. The Canada line was more than 50% financed by the public sector, and there are already complaints that some of the corner-cutting done by the private builder (possible because of wiggle room in the contract) compromises future operations.

    Here in Ontario, we are often told that we can have “free” transit expansion because the private sector will pay for it. That’s hogwash, but it plays well with the neocons. The problem is that we may get into long-term deals with future costs that are not understood or revealed up front, and the whole process will be shrouded in the cloak of “commercial confidentiality”. There is also the problem of sweetheart deals such as the sale of Highway 407 where the private company has a lock on any future tariff increases they want to impose.

    The point has been raised often that the public sector can borrow at more favourable rates than the private sector, and unless overall costs are lowered through magical “efficiencies”, the PPP costs more in the long run.

    In the case of the KW LRT line, there is an advantage because there is no base of expertise in the local transit system to build and operate something like this. This is not the same situation as one would find in a large city with an established transit system.


  6. On a related note: Jarrett Walker of Human Transit just wrote about this interesting new transit planning app called Transit Mix.

    Armchair route planners (and professionals) rejoice!

    Cheers, Moaz

    Steve: Of course, for anyone with an “M1” postal code, it should preallocate a subway that cannot be erased from the map 😉


Comments are closed.