The case in three lines. The Green Gazette names Route 7 the Most Improved Bus Route because it went from 61% on time in March to 84% in April, the month a new depot manager started. Over the whole year the route is usually about 80% on time, before the change and after it; March was an unusually bad month, the month the ring-road works ended. The old manager, moved after March, writes in, and Tara has to write her last corrections box.

One question to ask. "What should the box say?" Then: "Did the new manager do nothing?" Let them argue both sides.

One thing to do at the table. Think of a time someone was praised for a great result and did worse next time, or told off for a bad one and did better. Ask: was it the praise or the telling-off, or was the first result just unusual?

The grown-up layer. The case is about regression to the mean: after an unusually bad month, the next one is usually more ordinary, so measuring "improvement" from the worst point always finds some. It gives the new manager her strongest point, that she really did change the rotas and the drivers say so. It gives the regular passenger his: the bus is on time, and that matters to him. The habits it models are looking at the whole season, asking what else changed, and comparing the usual with the usual.

The case leaves the line blank on purpose. Ask for it before you offer yours.