Coverage overview.
How the public transit feeds in this site's current coverage are changing and whether the comparable covered corpus is getting better. This is not a census of any country or of the world. An absent scorecard is simply not covered yet.
Absolute score rankings and individual percentiles are not published. Named changes compare a feed only with its own prior check when the rubric, scoring profile, validator, and measured category set are unchanged. Corpus aggregates use rubric 1.3, scoring profile gtfs-scorecard-1.3, MobilityData gtfs-validator 8.0.1, reader archive profile raw-v1, and measured categories Correctness, Freshness, Rider experience; records with unresolved duplicate identities are excluded.
What changed since the last check
- 0 improved
- 17 slipped
This summarizes significant movers, not every quiet feed.
Most improved
- No comparable upward moves were recorded in this snapshot.
Needs attention
- Vy Express (Vy Buss) down 5.3 · D → F
- Bloomington Transit down 2.8 · C
- City of Jackson (JTRAN) down 2.2 · C → D
- Concord Kannapolis Area Transit (Rider Transit) down 2.2 · D
- KU on Wheels & Lawrence Transit System down 2.1 · D
- North Carolina State University (NCSU) down 2.1 · D
- RoadRUNNER Transit down 2.1 · F
- Rosemead Explorer down 2.1 · F
- Go-Ahead Ireland down 1.8 · C
- Autotransporta direkcija (ATD) national routes down 1.7 · F
- Autolinee Toscane (at) down 1.7 · F
- Embark down 1.7 · B
- PKP Szybka Kolej Miejska w Trójmieście (SKM) down 1.7 · F
- Broward County Transit down 0.4 · B → C
- Réseau urbain TCL down 0.4 · B → C
- TAD ligne Arrens-Marsous - Lourdes down 0.4 · B → C
- ふれあいバス (猪名川町) down 0.4 · C → D
Subscribe to changes (Atom)
to get grade drops and export changes in a feed reader or a webhook, with no
sign-up. Each agency also has its own feed at
/agency/<id>/feed.xml.
Is transit data getting better?
Across feed scorecards with one rubric, scoring profile, validator, and measured category set, the corpus average score slipped 0.1 points between 2026-08-06 and 2026-08-23 (now 68.1).
Covered-corpus average score
By date
| Date | Avg score | Feeds | Expired |
|---|---|---|---|
| 2026-08-23 | 68.1 | 1577 | 5.3% |
| 2026-08-22 | 68.2 | 1577 | 5.2% |
| 2026-08-21 | 68.2 | 1576 | 5.0% |
| 2026-08-16 | 68.2 | 1576 | 5.0% |
| 2026-08-15 | 68.2 | 1576 | 5.0% |
| 2026-08-14 | 68.2 | 1576 | 5.0% |
| 2026-08-13 | 68.2 | 1576 | 5.0% |
| 2026-08-12 | 68.2 | 1576 | 5.0% |
| 2026-08-11 | 68.2 | 1576 | 5.0% |
| 2026-08-10 | 68.2 | 1576 | 5.0% |
| 2026-08-09 | 68.2 | 1576 | 5.0% |
| 2026-08-08 | 68.2 | 1576 | 5.0% |
| 2026-08-07 | 68.2 | 1576 | 5.0% |
| 2026-08-06 | 68.2 | 1575 | 5.1% |
The series carries each agency's most recent score forward to each date and averages, so it is smooth even though agencies are checked on different days. It never changes a grade. The same data is at the trend API (trend.json).
In plain words: this page tracks the covered corpus at once, not any single agency. The most common problems page names the recurring fixes behind these numbers. Writing about this data? Start with the reporter's page.