Points per post-up is the wrong number to keep

Play-type efficiency splits credit at the wrong boundary and makes good actions look like bad ones.

The tagging convention is the problem. A possession gets labelled by the action that ends it, so a post entry that draws two defenders and produces a corner three is filed under spot-up shooting. The post-up gets none of the credit. Then someone publishes a chart showing post-ups are the least efficient play type in the sport, and a coaching staff somewhere quietly cuts an action that was working.

I do not think anyone is being dishonest. It is just that the unit of measurement was chosen for convenience. Ending actions are easy to tag from video. Causal chains are not.

Consider what this does in the other direction. Transition scoring looks spectacular in every play-type table, because a possession only gets tagged as transition when the break actually materialised. The failed breaks, where the outlet was late and the offence resets into a half-court set with fourteen seconds gone, get tagged as something else entirely. So the number tells you what happens when running works, not whether running is a good idea. Those are different questions and only one of them is useful.

The fix is not complicated conceptually. Measure at the possession level. Ask what a possession containing a post entry is worth, on average, against a defence of a given shape, compared with a possession that did not contain one. You lose granularity. You gain the ability to answer the question a coach is actually asking.

There is a reason it does not happen much. Possession-level splits shred your sample size. A team runs perhaps eight post entries a game against a switching defence, so a season gives you a few hundred, and the confidence interval swallows the effect you were trying to find. Play-type tables are popular partly because they produce numbers that look stable, and stable numbers are persuasive whether or not they mean anything.

My rule of thumb is that any statistic which assigns full credit to the final action of a possession should be treated as a description, not an evaluation. It tells you what happened. It does not tell you what caused it.

If you take one thing from an efficiency table, take the shape of the distribution rather than the average. The average post-up is mediocre. The good ones and the pointless ones are not the same play, and lumping them together is how you end up coaching to the mean.