basketball.casa

A basketball world simulated from the body up. Watch only.

A good decision that did not work: the basketball replay shows the chance a player had when they chose

basketball.casa is a basketball league where every game is simulated. Nobody scripts the plays. Each one comes out of who the players are at that moment: how tall, fast and tired they are, what they can see, and what they decide. You can watch any game back exactly as it happened, and pause it to see why a play went the way it did.

Someone watching one of our replays asked a sharp question about a drive the page called "cut off", with a better than even chance to beat the defender: was that chance the estimate before the drive, or worked out afterwards from how the play finished? Keeping the timing clear, they said, is what lets you tell a reasonable gamble that failed from a poor decision. They are right, and it is worth a whole post.

The chance is the one the player had when they chose

The number is fixed at the moment the player decides to drive, before a single step of the drive is run. It is the chance the driver had then, against the defender in front of them as they were at that instant. Nothing that happens afterwards changes it. The page then shows, beside it, what actually happened.

A paused replay of a basketball game, score 10 to 14, shot clock 14.9, from a camera following the ball: a blue player has just let the ball go, with red and blue players spread across the court. A card at bottom left says the driver was cut off, getting 1.55 metres round the defender in the 0.90 seconds the defender needed to cut across, and a note at top right says they had a 54% chance to beat them.

The card names the race the drive was: how much room the driver could get round the defender, against the time the defender needed to cut across in front. Here the driver got 1.55 metres round in the 0.90 seconds the defender needed. That was a better than even chance, and this time it was not enough. The card also lists the driver's own top speed and how quickly they reach it, because those are what the race was run with.

Good decisions and good results are different things

A drive with a 54% chance fails almost half the time. When it does, the player did not choose badly; they took a better than even chance and lost it. A drive with a 15% chance that happens to work was still a long shot. Card players and sports analysts have a name for judging a choice only by how it turned out: resulting. The replay is built so you do not have to: the chance at the moment of choice and the result sit side by side, and you can read both.

Does the number mean what it says?

It should, and we checked. Over twelve full games we sorted every drive by the chance it had when it was chosen and counted how often the driver got by:

That is no accident: in this league the chance shown is the same chance the game plays the drive out with, so a number on the page is never a label added afterwards. Over many drives it comes true, and on any one drive it can go either way, which is exactly what makes a single play worth pausing on.

How to see it

Open any replay and pause on a drive. The note at the top says how it ended and the chance it had, and the card names the players in the race and the numbers they ran it with. If you want to see what a player believed about the defenders when they chose, our post on where the ball handler thought the defenders were shows that too.