What the Bus Line Refresh changes, and for whom

Two readings of one plan. Who gains and who loses coverage, measured at every populated census block in Allegheny County and weighted by who lives there — and which removals take the most riders, measured at the stops people board today. Read across all six equity questions at once, the plan's trade runs progressive — car-free, low-income and Black residents come out ahead of the county on nearly every tier. The exception is age.

Losing more is not the same as losing ground

Each dot is one group on one tier, placed by how fast it loses service compared with the county as a whole (across) and how fast it gains service back (up). Dot size is the group's population.

Only the shaded corner is losing ground: above the county rate on losses and below it on gains. Five groups lose net access to any bus at all, particularly residents aged 65 and over.

Where the trade landed

The same result in each group's own percentage points. Every group loses some plain coverage and every group gains weekend frequency; what differs is the exchange rate. Read the two columns together — sorted by the left one, the right one is close to its mirror image.

• Race · people
• Age · people
• Income · households
• Vehicles · households
• Disability · people
• Language · households

Where this happened

The losses are concentrated. Of Allegheny County's 1,062 census block groups, 182 lose coverage and 75 gain it; a hundred of them hold 90% of the loss. So this is a list rather than a map — and the list says something the county totals do not: the coverage being trimmed is suburban. Baldwin, Ross, McCandless, Mount Lebanon, Kennedy, Scott, Plum, Bethel Park. That is the same geography behind both the age finding and the white-residents finding above; they are one result seen through two variables, not two independent results.

Car-free households are the ones with no fallback, so they are the first table.

These 12 places hold 1,530 of the 2,376 car-free households that lost it; a further 843 are spread across 40 more places. Full list, block group by block group, in data/equity_places.csv.

And the plan does add buses to ground that has none today — most of all in McKeesport, and in West View, Brackenridge and Robinson. The fourth row is unnamed because it takes its name from the nearest bus stop and there isn't one: four block groups on Perry Highway around Wexford, where nothing stops within 3 km today and the plan puts ten stops, including one at AHN Wexford. They have no name to borrow for the same reason they are gaining a bus.

These 12 places hold 15,546 of the 20,095 residents that gained it; a further 4,549 are spread across 22 more places. Full list, block group by block group, in data/equity_places.csv.

The removals that take the most riders

Of the 6,284 locations a bus calls at today, 633 lose every one of them, and between them those locations carry 580 weekday boardings — 0.8% of the system's 73,408.

The largest single loss in the county is about seventy-four boardings a day, and 263 of the 633 removed locations board nobody at all.

Every one is listed below, ranked by the riders it takes and running down through the clusters that board nobody. A row is a group of removed stops within 150 m of one another — PRT splits a single corner into two stop ids and a corridor into a dozen — named by the street they share and by the municipality or neighbourhood whose boundary contains the busiest of them. Click a row to open it on the map.