The One More Run Problem, Measured Rather Than Complained About

Why declare the intention first

Everybody knows the sensation of playing longer than planned. Almost nobody can tell you afterwards by how much, because the plan gets quietly revised while the playing happens. The only way to measure the gap is to fix one side of it before the game starts.

So before each evening a figure is written down: how long we intend to play. It is never revised. When the session ends, the actual figure goes beside it, and the difference across twenty evenings is the overrun.

The table

Four further games sit between eleven and twenty-one minutes of overrun and are omitted from the table because they add nothing to the pattern.

The variable

Overrun does not track total playtime, difficulty, genre or our own score. It tracks one thing: whether a reward is visible before the current unit of progress ends.

In the four games with overruns under five minutes, the next unlock, upgrade or reward is not shown until it has been earned. In the four with overruns above ten minutes, it is displayed continuously — a progress bar, a next-tier preview, a streak counter or a percentage.

Why this is not the same as stop quality

The other log on this site grades what leaving costs. This one measures how far past your own intention a game carries you, and the two are related but distinct.

A game can be cleanly leavable and still produce a large overrun, because nothing prevents you from choosing to continue when the next reward is dangling in view. Conversely the driving game has a four-minute overrun with forty-minute routes, because although the units are long, nothing is being shown to you that you have not already earned.

Stop quality asks what leaving costs. Overrun asks how hard the game works to postpone the thought. A game can pass one and fail the other.

The intended figure moves too

One finding was unexpected and slightly uncomfortable. Our declared intentions for the high-overrun games grew over the twenty evenings — for the tower defence game the first week's declarations averaged thirty-eight minutes and the third week's fifty-five.

We had adjusted our expectations to the game rather than the reverse, without noticing, and only the written record showed it. Because intentions are never revised retrospectively, the figures in the table use the declaration made on each night, which means the recorded overrun for that game understates what it was doing to our evenings early on.

What this log needs next

Twelve games is not enough to separate reward visibility from unit length, because in this sample the two mostly move together. Isolating them requires either a game with long units and hidden rewards or one with short units and a visible progress bar, and neither has yet turned up in the file.

A patched game would also serve. If any title here adds or removes a visible progression display in an update, it will be measured again immediately and the result reported here.