Your pick and roll data is full of handoffs

Tracking systems log dribble handoffs and ball screens in the same bucket, and the two actions ask different questions of a defense.

A handoff is not a screen, whatever the data says. The ball moves. That single difference changes the defensive problem so much that filing the two together is close to useless, and yet they land in the same bucket in most of the numbers you will read this season.

Think about what a defender has to solve. On a ball screen, the ball is already in the hands of the man he is guarding and the threat is continuous. On a handoff, there is a moment where the ball belongs to nobody, which creates two live options that do not otherwise exist. The receiver can refuse the handoff and cut. The passer can keep it and turn the whole thing into a drive.

Defensively, that means top locking is far stronger against a handoff than against a screen, because you are denying a catch rather than denying a route. It means switching is more attractive, since the exchange happens in a smaller space and the mismatch is briefer. And it means the big man's timing problem is different. He is not defending a rolling teammate, he is defending a player who was just holding the ball and now has to convert into a roll from a standing start.

None of this survives being averaged. When a defensive rating tells you a team allows a certain efficiency on pick and roll ball handler possessions, some meaningful share of that number is handoffs, and the mix varies enormously by opponent because some clubs build their whole half court out of them.

I am not asking for anything exotic. Split the bucket. Report the two actions separately and a lot of confusing player evaluations become clear, particularly for centres who look poor in one context and fine in the other.

You can hear the difference in how coaches talk about them. Nobody says drop coverage about a handoff. They say show, or they say jam it, or they say switch the exchange, which is a different vocabulary for a different problem. The people who play the game separated these two things a long time ago. The people counting the game have not caught up.

Until then, be suspicious of any argument about ball screen defense that rests on a single season-long percentage. Somebody averaged two different sports together.