A three-point team can be better and still lose the series

Building an offence around high-variance shots raises your ceiling and hands a worse opponent a real chance over seven games.

Shoot forty threes a night and your bad nights are worse than anyone else's bad nights. That is not a criticism of the strategy. It is the invoice attached to it.

The maths of the shot is settled and no serious person argues against it. A shot worth three, taken at a decent rate, beats a long two at a better rate, and any team refusing to accept that is choosing to score less. Fine. What gets ignored is that the same trade changes the shape of the distribution and not only its middle. Threes come in streaks and droughts because the misses are noisier and the makes come in bunches, and a team that gets a large share of its points from them will produce more extreme results in both directions.

Over eighty-two games this is irrelevant. Variance averages out and the better team accumulates wins. Over seven games it is decisive, because seven is a small number, and a small number is exactly where noise beats quality.

So the honest version of the argument is that a three-heavy team maximises expected points per possession and simultaneously buys itself a worse chance of surviving a specific short series against a specific opponent who can defend the arc. Both things are true. Front offices tend to say only the first one out loud.

What I would want from a contender is not fewer threes. It is a second way to score that does not correlate with the first. A post player who can get a bucket against a set defence. A guard who lives at the free throw line. Something that keeps producing on the night the shots are rattling out, because in a playoff series you do not get to average across the bad night, you just lose it.

The teams that win titles almost always have this, and they are then praised for their toughness or their culture, when what they had was an uncorrelated scoring source.

Efficiency wins seasons. Redundancy wins series. Those are different problems and it is strange how often the same roster is asked to solve both.