3 ms·
> it varies a lot by time of day etc. That _may_ explain the difference. The officially reported percentage is per train, not per passenger. Passenger onboardi
by Someone 5y ago
> it varies a lot by time of day etc.
That _may_ explain the difference. The officially reported percentage is per train, not per passenger. Passenger onboarding can cause delays (for example, a rural station may, some days, have a school class enter a train. If you make room for that in the schedule, customers will ask why you stop for 5 minutes. If you don’t, the train will be late at its next stop, if that’s only a few kilometers away), so chances are trains with fewer passengers are more punctual. If so, the per passenger percentage of on time trains will be lower.
It also is measured only at final destinations. That percentage almost certainly is higher than that measured at each stop (a train that’s late half-way through can try to make up for lost time, but it can’t make up for being late at stops it already had)
Finally, there the question how cancelled trains are measured. Possibly, they aren’t counted as late, but do delay passengers.