Line 1T · Fruitvale Tempo platforms
8:15–10:15 PM
A qualifying gap appeared during this window on three of four nights. In all three cases, three to five scheduled arrivals fell inside a 45- to 62-minute gap.
Observed hit rate: 75%I have been spending a lot of time lately trying to understand the difference between a bus route that looks acceptable as an average and one that is actually useful when a rider walks up to a stop. A line can run most of its scheduled trips over a full day and still produce a 45- or 60-minute hole at exactly the time someone needs it.
So, I asked a practical question: if a reporter wanted to stand on the street and talk to riders while one of these failures was happening, where should they go? Using data from the AC Transit Report Card, I looked at actual arrivals from Monday, August 24 through Thursday, August 27 at central stops on AC Transit’s seven busiest routes. Arrivals were separated by route, stop, and direction; a bus going the other way cannot close the gap.
38gaps over 40 minutes on the seven busiest routes
26of those gaps began between 7 and 11 PM
4 of 4nights with a 40+ minute Line 40 gap
The seven busiest routes
The ranking below uses scheduled trips across the four-day period. For each route I picked a central, practical “screenline” where someone could actually observe both directions from nearby platforms. A qualifying gap begins after 5 AM and contains more than 40 minutes between consecutive observed arrivals in the same direction.
| Route | Scheduled trips | Screenline | 40+ min gaps | Days affected |
|---|---|---|---|---|
| 1T | 884 | Fruitvale | 11 | 3 of 4 |
| 40 | 788 | Foothill & Fruitvale | 12 | 4 of 4 |
| 51A | 696 | 12th Street BART | 2 | 1 of 4 |
| 6 | 692 | Telegraph & 40th | 4 | 3 of 4 |
| 51B | 692 | Downtown Berkeley BART | 1 | 1 of 4 |
| 57 | 604 | MacArthur BART | 0 | 0 of 4 |
| 9 | 584 | Hayward BART | 8 | 4 of 4 |
The problem is particularly concentrated in the evening. Line 40’s twelve gaps all began between 7 and 11 PM. Seven of the 1T’s eleven gaps and three of Line 6’s four gaps began during that period. By gap start time, the seven routes produced one AM-peak gap, three midday gaps, three PM-peak gaps, 26 evening gaps, four after 11 PM, and one beginning shortly before 6 AM.
The worst gap on each route, each day
The number in parentheses is how many scheduled arrival times fell between the two observed buses. Those trips may have been cancelled, short-turned, very late, or absent from the real-time feed. What the data can establish is that no bus was observed at this stop in this direction during the interval.
| Route | Mon. Aug. 24 | Tue. Aug. 25 | Wed. Aug. 26 | Thu. Aug. 27 |
|---|---|---|---|---|
| 1T | 39.5 min near-miss | 8:48–9:50 PM, downtown — 62.4 min (5) | 10:11–11:11 PM, San Leandro — 59.6 min (4) | 6:22–7:08 PM, San Leandro — 46.6 min (5) |
| 40 | 8:05–8:59 PM, downtown — 54.4 min (4) | 9:36–10:19 PM, downtown — 43.0 min (2) | 8:05–8:51 PM, downtown — 46.0 min (3) | 8:54–9:35 PM, downtown — 41.0 min (2) |
| 51A | — | 10:12–11:15 PM, Rockridge — 63.2 min (3) | — | — |
| 6 | 7:52–8:37 PM, Berkeley — 44.3 min (3) | 7:58–8:39 PM, Berkeley — 40.9 min (2) | 9:17–9:58 PM, Berkeley — 40.4 min (2) | — |
| 51B | 8:38–9:18 PM, Rockridge — 40.8 min (3) | — | — | — |
| 57 | — | — | — | — |
| 9 | 8:45–9:32 PM, Union City — 47.1 min (4) | 7:15–8:00 AM, Union City — 45.1 min (3) | 1:19–2:02 PM, Union City — 42.8 min (3) | 5:27–6:13 AM, San Leandro — 46.6 min (2) |
A dash does not mean the line ran well all day; it means that this particular stop and direction test did not produce a gap strictly greater than 40 minutes. The Monday 1T result is a good example of how arbitrary a hard threshold can be: its 7:57–8:36 PM gap was 39.5 minutes, missing the cutoff by about 30 seconds.
Where I would send a reporter
I tested fixed two-hour observation windows. A “hit” means that the visit would have overlapped at least 30 minutes of a direction-specific gap longer than 40 minutes. This is useful for deciding where to interview riders, but it is still just four days of history.
Line 40 · Foothill & Fruitvale
8:15–10:15 PM
A qualifying gap appeared on all four nights. The gaps lasted 41 to 54 minutes and contained two to four scheduled arrivals; two nights contained at least three.
Observed hit rate: 100%Line 6 · Telegraph & 40th
8–10 PM
A qualifying gap appeared on Monday, Tuesday, and Wednesday. The gaps lasted 40 to 44 minutes and contained two to three scheduled arrivals.
Observed hit rate: 75%Line 52 · North Berkeley BART
9–11 PM
A qualifying gap appeared during this window on three of four nights. Gaps ranged from about 59 minutes to the extraordinary 148-minute Wednesday gap.
Observed hit rate: 75%Those would be my first four choices. Lower-confidence options are the 51A at 12th Street BART from 9–11 PM and the 51B at Downtown Berkeley BART from 7:30–9:30 PM; each hit on only one of four days. Line 9 failed on all four days, but at wildly different times, so there is no honest fixed interview window to recommend from this sample. I would not send someone to Line 57 at MacArthur BART based on these four days.
The most striking result is not simply that a bus was late. It is that on some of the East Bay’s busiest lines, multiple scheduled buses can fit inside the space between two actual arrivals. For the 1T and 40, this was not an isolated middle-of-the-night event. It was a recurring evening pattern at stops where people should be able to depend on frequent service. This is really bad service, folks!