The idea of clutch may be the most divisive concept in basketball circles. It's the closest thing the sport has to religion. There are many who strongly believe in clutch and will bristle at any suggestion it does not exist, and others who don't think the science supports any evidence of its existence.
I've been in both camps and currently sit on the fence. We don't have much evidence that clutch exists. But we also don't have much evidence against its existence. There is also some sound theory as to why it might exist.
Call me Clutch Agnostic, with a slight lean toward "There Is A Clutch Unicorn Gene That Looks A Lot Like The Gene That Allows Us To Remain Calm And Focused Under Pressure".
Of course, like any theology, there are infinite debates within the debate. Like, what constitutes clutchness? There are so many ways to slice the data, both formally and not.
Take Kemba Walker. He hit his third game-winner of the season Wednesday night, a very difficult pull-up in traffic.
Three game winners in basically two months. If that's not clutch, what is?
Here's the problem: when you look at the data, it's hard to find evidence Walker is better in clutch situations, no matter how you define clutch. For perspective, he has an effective field goal percentage of .449 overall this season. (All data in this piece comes from NBA.com.)
In the final five minutes of regulation or overtime in a one-possession game, Walker has an eFG of .410. If you keep those time restrictions, but allow a margin of up to five points, he has an eFG of .357. In the final two minutes of regulation or overtime in a one-possession game, he has an eFG of .414. If you extend the margin allowable to five points, Walker's eFG is .378.
In the final minute of a one-possession game, Walker is 7-for-18 with one three for an eFG of .417. In the final 30 seconds of a one-possession game, he is 5-for-15 with one three for an eFG of .368. But in the final 10 seconds of a one-possession game, he's 4-for-10 with one three for an eFG of .450 -- which is basically his overall shooting clip for the season.
To find data from this season that supports the notion that Walker is clutch, you have to drill down to the thinnest slice and smallest sample size. That begs the question: if Walker shoots effectively in the final 10 seconds of a close game but shoots poorly in close games with one or two or five minutes remaining, is he clutch?
We need some context. In the last 10 seconds of one-possession games, NBA players are 67-for-234 with 22 threes for an eFG of only .333 this season. That's much worse than KWalker's .450 mark. In the final minute of one-possession games, NBA players are 222-671 with 46 threes for an eFG of .365. Again, that's much worse than Walker's .417. In the final five minutes of one-possession games, NBA players are 989-for-2,491 with 215 threes for an eFG of .440. That's substantially better than Walker's .410.
So again, it depends on how you define clutch. If we're talking about the final 10 seconds or minute of a close game, Walker has performed better than the average NBA player. If you're looking at the last five minutes of close games, his performance is underwhelming. If you're looking at games with 5-point margins, the shine really gets taken off for Walker. How you slice the data is the biggest determinant because we are dealing with really small samples on the individual level.
And this is exactly why I'm Clutch Agnostic while leaning toward the existence of some human attribute that allows some people to remain calm and focused under pressure. There's a hint that some players are better in these situations than others, but nothing like robust evidence.
That doesn't mean, of course, that I can't watch in amazement as Walker, Kobe Bryant, Monta Ellis or anyone else sinks a tough game-winner. That's the beauty of the game unencumbered. No matter what you believe, you still get to enjoy every second without prejudice.