Declared against measured

Where we and TfL disagree

Two bands on one clock, for every line. The top band is the status TfL declared; the bottom is what we measured from the trains over the same hours.

Over the week since Mon 14 Sept, across all 13 lines, there were 13 days in which TfL was showing a good service while we were measuring a real delay — and 8 days in which TfL was reporting disruption that our measurements did not pick up. Both are on this page, because publishing only the first would make our own record look better than it is.

Neither band is a verdict on the other. This is two measurements of the same railway, and where they differ, one of them is missing something — the second figure above is the honest estimate of how often that is us. How this is measured →

Good
Minor delay
Delayed
Severe delay
Nothing measured, or nothing running

Every line, the last seven days

Beside each line is how much of the week the two bands disagreed, counted both ways: time we measured delays that TfL was still calling a good service, and time TfL was reporting disruption that we did not measure. A line with neither figure is one we and TfL saw the same way all week.

Bakerloo line status13 hours of delays TfL called a good service · 21 hours of disruption TfL reported that we did not measure

Central line status26 hours of delays TfL called a good service · 2 days of disruption TfL reported that we did not measure

Circle line status2 days of delays TfL called a good service · 10 hours of disruption TfL reported that we did not measure

District line status13 hours of delays TfL called a good service · 8 hours of disruption TfL reported that we did not measure

Hammersmith & City line status13 hours of delays TfL called a good service · 26 hours of disruption TfL reported that we did not measure

Jubilee line status28 hours of delays TfL called a good service · 17 hours of disruption TfL reported that we did not measure

Metropolitan line status31 hours of delays TfL called a good service · 12 hours of disruption TfL reported that we did not measure

Northern line status4 hours of delays TfL called a good service · 19 minutes of disruption TfL reported that we did not measure

Piccadilly line status19 hours of delays TfL called a good service · 8 hours of disruption TfL reported that we did not measure

Victoria line status2 days of delays TfL called a good service · 24 hours of disruption TfL reported that we did not measure

Waterloo & City line status2 days of delays TfL called a good service · 59 minutes of disruption TfL reported that we did not measure

Elizabeth line status2 days of delays TfL called a good service · 9 hours of disruption TfL reported that we did not measure

DLR status40 minutes of delays TfL called a good service · 3 hours of disruption TfL reported that we did not measure

A grey stretch on our band means we had no measurement for those hours — usually the small hours, when nothing is running to measure. On TfL’s band grey means they declared nothing was running.

Our band only changes once a condition has held for fifteen minutes, and the figures above count only stretches that lasted that long. A line sitting either side of a threshold would otherwise flicker every few minutes and turn this into a barcode, and we would be claiming disagreements too brief for anyone to have experienced. The trade is that a sharp ten-minute incident is absorbed into the hour around it; the score chart on each line page still shows those.