The Trouble with “Transit Score”

If you want a quick assessment of the usefulness of transit at a location — say, a place where you’re planning to live, locate a business, or invest — what do you want to know?

The tool realtors know is Transit Score, a two digit number (like its elder sibling WalkScore) that supposedly gives you a quick hit of meaning about how good transit is.  Transit Score was invented by WalkScore.com, which has since been eaten by real estate giant Redfin.

But here’s how Transit Score is calculated:

To calculate a raw Transit Score, we sum the value of all of the nearby routes. The value of a route is defined as the service level (frequency per week) multiplied by the mode weight (heavy/light rail is weighted 2X, ferry/cable car/other are 1.5X, and bus is 1X) multiplied by a distance penalty. The distance penalty calculates the distance to the nearest stop on a route and then uses the same distance decay function as the Walk Score algorithm.

So in short:

  • For a product of WalkScore.com, Transit Score is awfully cynical about walking.  Your Transit Score goes down steeply if you have to walk further to the bus stop, even if wider bus stop spacing provides you with faster service, as it often does.  In short, Transit Score assumes that walking is bad for you, and that you hate walking more than you hate riding.
  • Transit Score assumed that the sexiness of transit technologies compensates for their objective uselessness.  For example, Transit Score assumes that you’d rather wait 20 minutes for a streetcar instead of 10 minutes for a bus, even though the two will have the same speed and reliability.
  • Above all, Transit Score is uninterested in how long it will take you to get anywhere.  It describes the transit around a site without evaluating where it goes.  Frequent transit that drove around in circles inside your neighborhood would score exactly the same as transit that went straight across your city and formed a connected network, accessing countless jobs and opportunities.

These problems arise from an unthinking real-estate world view in which transit is a feature of a site, like parks.  In fact, transit quality lies in a site’s position in a network, and it is the network, not the immediately proximate features, that delivers all valuable transit outcomes.

Imagine if an Auto Score were constructed like Transit Score:  It would give no value to average travel times to actual jobs around the region, but would be very interested in the square feet of paved roads found within a very short radius around the site.  It would also care about the aesthetic quality or “look and feel” of that pavement, and might give some weight the local speed limits.(1)

A few years ago, when Transit Score first rolled out, I discussed it ” WalkScore’s Matt Lerner, expanded on this very critique, and suggested a better (though computationally intense) approach.  It involves aggregating the content of travel time isochrones – effectively “maps of your freedom” — over all likely destinations from any residence, so the two digit score is actually a percentage, a composite of answers to the question “What percentage of jobs, retail, etc can you get to in __ minutes, on transit, from here?”   It needs refinement, but that’s the only truly factual measure of access that could be reduced to a two digit number — one that would actually mean something.

(1) Yes this is an inexact analogy.  Transit Score does begin with frequency, which matters a lot, and the impossibility of translating frequency into automotive terms is one of the main reasons it’s so poorly understood, especially in North America.

6 Responses to The Trouble with “Transit Score”

  1. M1EK March 27, 2017 at 7:07 am #

    “even if wider bus stop spacing provides you with faster service, as it often does”

    The most oversold thing in transit, bar none. Hardly ever faster _enough_ to be worth the extra time to walk.

    • Ilya Petoushkoff March 27, 2017 at 1:08 pm #

      Usefulness of farther stop spacing depends on from which distance you are increasing. Reasonable distance is about 400…600 m (depending of course on the urban context of particular area), from which any further increase indeed won’t make much sense.

      • Jarrett Walker March 27, 2017 at 2:54 pm #

        In the US context, stop spacing can be absurdly close — 150m in some cities — and widening to 400-600m makes a difference to speed if passenger volumes are high.

        • Ed March 27, 2017 at 4:11 pm #

          I already walk 500 m up to the main street with the transit route. My local streetcar stop was removed a couple of years ago. The next closer stop is the loop itself (never the direction I want to go), but it’s upstream and I have to cross on a traffic light and then make my way across a slip ramp. In the other direction, it’s a 200m walk.

          There is nothing quite as infuriating as walking for five minutes up my street, and then seeing the streetcar go by, that I could have caught if the stop wasn’t removed.

          IN a discussion locally, someone brightly informed me “It’s not a problem, you just have to leave a little eariler to catch the streetcar!” Which kind of negates the “extra speed” (of which there is….none….anyway).

  2. Dorian April 3, 2017 at 8:40 pm #

    The closest bus stops I know of are stops in NYC that can be around 20 m apart. That’s shorter than the crosswalk at the end of my street!

  3. vramin April 5, 2017 at 10:58 am #

    Actually, walkscore used to have an isochrone map. I used that to help planning my last two moves. They seem to have removed that feature from their website, for reasons I don’t understand.