Behind the scenes with MBTA data.

Can you identify an MBTA station by its boarding print?

Each MBTA station has its own boarding pattern generated by the land use around it, transfers from other services, and parking capacity. The pattern is most variable by time of day (and weekday vs weekend). But some stations have inconsistent boardings by season or proximity to large special event locations.    

The charts in this post show the distribution of station entries by 15-minute period for weekdays in Fall 2016 (September 1 through December 31). A “station entry” is a card or ticket validation recorded at a faregate. The number of entries at each station for each 15-minute period was found, and then the 10th, 50th, and 90th percentiles for each period was found to show variation between weekdays. From these charts, we can examine patterns of ridership and learn some interesting things about how stations work.

Most of the stations follow similar patterns. This post will examine what a few of the common typologies look like and the reasons for their patterns. Then, we have a quiz to test your knowledge of the system!

Many outlying stations have graphs like the following at Davis Square. Passengers living nearby or alighting buses that service the station enter these stations in the morning on their way to work. Then ridership is lighter for the rest of the day. We’ll call this the “AM Peak” pattern.

Station Print for Davis Station

Above: Davis Square

Here you can see a spike of entries in the morning as people living near Davis Square (or taking buses that drop off there) enter the station. For the period from 8:15-8:30, the median number of entries was 856 – enough to fill an entire Red Line train. After about ten AM, ridership is fairly steady throughout the rest of the day, at approximately 400 entries per hour.

Stations near job centers have the opposite pattern (We’ll call it “PM Peak”). Here’s State – entries are fairly low throughout the day, and then peak right at 5 PM as people get out of work.

Station print for State

Above: State

A number of stations located near middle or high schools have an additional peak pattern around the times that school lets out. Here’s Savin Hill. Note that the y-axis here is much smaller than the previous graphs, as this is a lower-ridership station.

Station print for Savin Hill

Above: Savin Hill

Finally, we noticed a number of stations that have both AM and PM peaks. In some cases this is because they are located near both centers of residential population and job centers (for example, Broadway), but more often this is because they see a number of transfers at one peak from people who are transferring from buses or commuter rail (or some combination of both). For example, Back Bay station has a peak in the AM from commuter rail riders, and likely has a good amount of walk-up riders as well.

Station print for Back Bay

Above: Back Bay

Other stations where non-traditional ridership occurs have much more interesting patterns. Here are the boarding prints for 5 mystery gated MBTA stations, see if you can identify them!

It might look hard at first so consider: the magnitude of boardings, the variance, the number and timing of peaks, and that this data is from Sept-Dec 2016. We chose stations with distinct and interesting patterns. You may wish to open the images in a new tab to view them at full size.

Give us your best guesses in the comment section below. Partial credit for explaining your thinking! We will post the answers next week. 


Mystery station print #1


Mystery Station #2


Mystery station print #3


Mystery Station #4


Mystery Station print #5