Why ridership counts are always wrong.

Ridership counts are always wrong in the small, useful sense that no measurement captures every passenger perfectly. An automatic passenger counter can miss a boarding or count a movement incorrectly. A farecard tap records a transaction, not always a person’s complete journey. A manual count is a snapshot made by someone trying to watch doors, remember numbers, and keep their balance as the bus pulls away.

I am Ines Halvorsen, a former transit planner. I have worked with numbers that looked crisp in a spreadsheet and knew exactly how many assumptions were holding them up. The answer is not to stop counting. It is to understand what each count means and use it for the question it can answer.

What the equipment sees.

Automatic passenger counters use sensors at doors to estimate boardings and alightings. They can provide valuable information across many trips, but they need calibration and quality checks. Crowded doors, mobility devices, strollers, or equipment issues can affect readings. Agencies compare counts and investigate patterns rather than treating each sensor record as unquestionable fact.

Farecard taps are another partial view. A tap may represent a boarding, but fare rules, transfers, cash payments, passes, and fare evasion all complicate the relationship between transactions and people riding. Some riders do not pay through a tap at all. A transaction dataset is not a perfect passenger census.

The count is not the city. It is a tool for noticing how the city moves.

There is also a difference between boardings and linked trips. If a rider takes a bus, transfers to a train, then boards another bus, the system may record several boardings. From the rider’s point of view, it may be one journey. Counting boardings helps agencies understand vehicle use and route activity; linked-trip measures try to describe complete journeys. Neither measure is inherently wrong, but they answer different questions.

A missed fare can mean a passenger is absent from fare data while still present on the bus. Conversely, a transfer may create another boarding record without representing a new person beginning a new journey. This is why planners combine sources where possible: counters, fare transactions, surveys, service data, and observation.

Trends, not false precision.

Why rely on imperfect counts? Because repeated measures can show direction. If the same route, time period, and counting method are compared over time, the pattern can reveal whether use is growing, falling, or shifting. Planners can examine loads by segment, compare service periods, and identify where a change may deserve a closer look.

The comparison needs care. A new fare system, route change, sensor upgrade, or unusual event can break the series. Agencies should document methods, note uncertainty, and avoid pretending that a small difference is meaningful when the measurement itself varies. A number is strongest when it comes with its definition.

When you read a ridership report, ask what is being counted: boardings, linked trips, passengers on board, or fare transactions? What time period and routes are included? Did the method change? Those questions do not undermine the data. They let you use it honestly—to see trends and make better decisions without mistaking a count for the complete story.

— Ines