I often travel to New York and, like many urbanists, find myself in unreserved awe at how busy the subway system is and how truly integral the subway is to daily life. While my city’s system (the Washington Metro) is the second busiest heavy rail (aka “subway”) system in the country, the number of people it carries isn’t even close to New York’s: the NY subway transports between 10-11 times more people per year than DC’s metro. Indeed, the NY subway carries 70% of all heavy rail trips in the US!
However, I always had an inkling that on a per-station basis, the stats are a bit closer. New York has an incredible number of stations: 422. While this is undeniably a strength of the system, it also means many stations are small and lightly used.
This is partly a result of the system’s age. Older systems tend to have more/smaller stations, reflecting many possible factors including a desire to saturate the market and a reflection of how much denser the urban cores of cities used to be (e.g. even with 1.6 million people, Manhattan is still below its 1910 peak of 2.3 million residents). Also, this is a result of the fact that NY’s subway was built by multiple private lines that often built competing stations near each other in popular areas. Some stations have been merged or closed in the decades since the system consolidated but many lighter used stations survive (often for good reasons).
By contrast, DC’s metro is a significantly newer system, expansion was planned slowly and deliberately, and its 91 stations tend to have one or two very large main entrances rather than many small entrances. As a user of both systems, I gained the impression that New York’s stations, on average, while busier, were not as busier in relative terms as the systems were overall.
The evidence appears to back this up:
While the New York Subway system as a whole carries 10.6 times more passengers annually than DC’s Metrorail, its stations are only about 2.3 times busier on average (mean).
I decided to expand this more broadly, looking at average station ridership across other large US heavy rail networks and two international systems: Tokyo’s metro (the world’s busiest) and London’s tube (the world’s oldest). ‘
Important: this is looking at simple mean average station use, simply taking overall system ridership and dividing by stations (see note below).
One conclusion is clear:
New York’s relative use is less pronounced on a per-station basis (although still quite high, especially in US terms). Although system-wide NY carries many times more passengers than any other US system , on a per-station basis, NY’s stations are about twice as heavily used as some major systems (e.g. Boston, San Francisco, and DC) and about three times as heavily used as the US (non-NY) average.
The data also suggests that not only is Tokyo the world’s busiest system by ridership, its stations are also very heavily used – about 2 times busier than New York’s already crowded stations. Having ridden extensively in London’s overcrowded system, I’m a bit surprised its stations are (on average) less crowded than New York’s however this may in part be due to design: many tube stations are exceedingly small compared to their current ridership needs.
Because of the nature of Boston and San Francisco’s systems, where their heavy-rail systems are complemented by and integrated with extensive light-rail systems, I thought it made sense to merge the two datasets and view average station ridership jointly:
Limitations, Caveats, and Notes
- Most obviously, this analysis ignores the distribution in each system which means it could mask the presence of outliers. While this is true, a cursory look of station ridership suggests there are no true outliers in the sense of stations would greatly skew the overall picture. Take NY for instance. While the least used station got only 305 average weekday users in 2015 (Beach 105 St.) compared to the busiest station (Times Square – 42nd Street) which had 206,247, there are actually quite a few stations that have few low-used stations; for example, 9 stations had <1,000 per weekday. Similarly, other systems have some low-use and high-use extremes and, overall that sort of balance out. That’s also part of the point: with a system that has 422 stations, there’s going to be some variation and let’s acknowledge that.
- Transfer stations (called “station complexes” by New York’s MTA) are where multiple lines intersect for the first/last/only time. While some systems count this as 1 separate station for each of the lines, for consistency (and logic), I did not. Therefore, while New York officially has 469 stations, if you remove duplicates, it only has 422 (the number I used). The one exception is for my analysis of joint light- and heavy-rail systems (Boston and San Francisco) where transfers between the two systems are typically considered two stations.