Behind the scenes with MBTA data.

The MBTA Performance Dashboard displays the MBTA’s ridership metrics through the most recent month for which we have complete data. This post explains how data collected by MBTA faregates and fareboxes is processed and turned into our ridership counts – first reported here, then reported yearly to the National Transit Database.

Ridership is reported as the number of trips taken by passengers on our system for the period specified. We collect ridership data using “interactions” with a farebox or faregate. Every time you tap your CharlieCard or pay with cash, your trip is added to the count. 

Ridership is reported here for an average weekday on bus and subway only, where we have automated fare collection (AFC) equipment.  Bus ridership includes local or express bus, trackless trolley, and Silver Line ridership.


What are unlinked trips?

In accordance with transit industry convention, the MBTA reports “unlinked” passenger trips as its ridership metric. Each time a passenger boards a transit vehicle, this is reported as one unlinked passenger trip. If a rider transfers from bus to subway, or from one subway line to another (even if they don’t have to tap their card again), it is counted as two unlinked trips. The reason for counting this way is a practical one: before smartcard fare payment technology, agencies had no way to know that the same person was paying at multiple places for the same trip – they would simply receive cash or a token that was not tied to a specific person.  

Why isn’t the data more current?

The MBTA would love to provide daily ridership reports, but there is a 1-2 month lag in uploading and processing the data from buses and light rail fareboxes. Unlike subway gates, fareboxes are not hard-wired to the MBTA’s central server, and the data they collect is uploaded when the vehicles are serviced.  Data from most vehicles is uploaded within a few days, but in some cases it can take several weeks. 

How is the data collected and processed? What’s the NTD?

The data is collected by the MBTA’s automated fare collection (AFC) devices. Whenever you tap your CharlieCard, insert a CharlieTicket, or put cash into a farebox, the transaction is recorded. These transactions are tallied and adjusted to account for the small number of people who board a vehicle or pass through a gate without interacting with the fare equipment. These “non-interactions” include people who “flash” a pass to the driver as well as our best estimates of fare evasion. Non-interactions also include people who are allowed to ride the T for free, such as children under 12. There are also adjustments for transfers behind the faregates; for example, from the Red Line to the Green Line at Park Street.

Flowchart describing the process of turning passenger trips into ridership data

Each year the MBTA collects manual counts to calculate the non-interaction factors by mode. At the close of each fiscal year, the ridership for the year is adjusted by the latest non-interaction factors and reported to the National Transit Database (NTD). These become the final ridership numbers for the year. The NTD is a nationwide database with ridership and service statistics on U.S. transit service providers going back decades. The Performance Dashboard reports numbers adjusted by the previous year’s non-interaction factors for the current fiscal year and the final adjusted NTD numbers for previous fiscal years. 

Why do we show weekdays only?

The dashboard displays ridership for an average non-holiday weekday over the month specified. This is not to exclude weekend riders (we count them too!) but to provide better comparisons, both between the T and other systems and between the T at different points in time. 

Where’s commuter rail?

Ridership data is shown here for MBTA-operated bus and rapid transit service. Commuter rail, ferry, THE RIDE and other private carriers are not shown because they do not use the AFC system, and most of the data is less reliable and less timely. The MBTA and our commuter rail operator Keolis hope to be able to provide detailed commuter rail ridership in the future as data collection improves.