by Ken Pomeroy on Thursday, February 28, 2013
With 23 seconds left in Tuesday’s Memphis-Xavier game, the Tigers’ Chris Crawford took an open three-pointer that would have given his team a two-point lead. He missed and Memphis went on to lose. The shot ended up being a referendum on the quality of Memphis’ 17-game winning streak. Had they won, this would have been a plucky team that comes together in tough times and does enough to win, and we would have seen a rise in the polls. But Crawford’s shot didn’t fall and so Memphis became a fraud in the minds of some. Perhaps no shot this season was more important to a team’s perception than that one.
Of course, if you’re familiar with my work you know where I stand on this. If your opinion of Memphis was significantly influenced by whether a 35% three-point shooter made a three-point shot, you’re doing it wrong. Based on how both Memphis and Xavier had played over the last two months, one might have expected the game to come down to the last minute. And the fact the Memphis couldn’t win the game says little about their character or their ability to win close games in the future. Hey, it might even reflect positively on their character since they got a really nice look to take the lead. The execution was fine and you can’t expect a 35% shooter to make 100% of his shots.
In an effort to test this idea, I thought I would run a simple experiment. When a team won its first close game of the season, what was their record in future close games? I only used conference games here to try and keep teams on relatively equal footing. For example, Xavier’s first A-10 game decided by three points or less was a 66-64 win over Saint Bonaventure. Since then, they’ve played one such conference game and lost it (73-71 to Richmond). Do this for every team since 2000 and you end up with a lot of data.
For instance, a team that won its first close game went 2770-2805 (49.7%) in future close games. By comparison, a team that lost its first close game went 2745-2750 (50.0%) in future close games. If the only information you had on a team’s performance in close games was that it was that clutch performance that one time you saw them, it’s pretty meaningless.
I’ve repeated this exercise for all three-point ranges up to 30. So in the table below, a team that won the first game it played that was decided by 10-12 points, won 57.8% of future games decided by 10-12 points. The results are fairly intuitive. Teams have more skill at winning games by large margins than small ones. Teams involved in a 30-point win are not likely to find themselves losing a 30-point game later in the season.
1-3 4-6 7-9 10-12 13-15 16-18 19-21 22-24 25-27 28-30 W 49.7 51.7 54.7 57.8 60.6 64.0 65.5 76.4 77.9 87.0 L 50.0 48.2 46.2 41.9 38.7 34.4 29.1 25.8 21.0 19.2
One close-game win isn’t enough to demonstrate a team’s mettle under pressure, but what if they won their first two close games? How did they perform in future close games after that? Obviously, the sample size is going to decrease, but we’re still on solid footing in that regard. Teams that won their first two close games went 914-850 (51.8%) in close games after that and teams that lost their first two close games went 870-881 (49.7%). So we do get a little more information that maybe a team possesses some small skill in close games. The table below contains similar data for other ranges.
1-3 4-6 7-9 10-12 13-15 16-18 19-21 22-24 25-27 28-30 W2 51.8 51.2 60.2 63.0 69.9 72.0 73.2 87.5 87.9 * L2 49.7 47.2 41.1 35.9 29.2 25.5 18.1 11.9 12.5 *
* fewer than 20 cases
Let’s keep going. When a team won its first three close games, they went 278-228 in future close games (54.9%). Teams losing their first three close games went 229-252 (47.6%).
1-3 4-6 7-9 10-12 13-15 16-18 19-21 22-24 25-27 28-30 W3 54.9 52.0 60.2 68.0 67.4 73.3 79.4 * * * L3 47.6 43.2 40.1 31.3 22.7 20.8 22.2 * * *
* fewer than 20 cases
OK, so if a team won its first four close games, surely they would be clutch and mentally strong. While the team losing its first four close games would be condemned to being choking dogs. In future close games, we’d just expect those teams to lose. We’re getting short on cases here, but the data says we might want to change the way we think. The winners of their first four close games went 75-71 (51.4%) and the losers went 57-67 (46.0%).
1-3 4-6 7-9 10-12 13-15 16-18 19-21 22-24 25-27 28-30 W4 51.4 48.9 61.0 70.5 69.8 * * * * * L4 46.0 42.1 24.1 34.2 6.1 * * * * *
* fewer than 20 cases
There are few teams that won their first five close games and played at least one more, but those that did went 17-18 (48.6%) and those that lost their first five went 7-19 (26.9%).
1-3 4-6 7-9 10-12 13-15 16-18 19-21 22-24 25-27 28-30 W5 48.6 45.2 68.2 80.8 * * * * * * L5 26.9 40.9 20.8 18.8 * * * * * *
* fewer than 20 cases
If a team that had won five close games in a row won a sixth, people would lavish all sorts of praise on their ability to perform under pressure. Even if you were skeptical about close-game powers, you’d look pretty foolish trying to provide an opposite viewpoint. Yet, as we’ve seen, each additional close win provides extremely little information about a team’s performance in its next close game. (For what it’s worth, the teams that won their first six close games went 4-1 if there was a seventh. Or eighth, ‘08 William and Mary is the only team to win its first seven close games against conference teams. They got to the CAA semis on the strength of two of those wins and beat VCU by three before losing by nine to George Mason in the CAA title game.)
This is not to say there aren’t teams with the ability to consistently win close games. It’s just that their record in close games tells you almost nothing about that skill. Simply identifying a team that has made a bunch of last-second shots (or seen opponents miss) and declaring them a bunch of winners is accurate from a reporting standpoint, but assigning the members of said team some special amount of intestinal fortitude is, in the vast majority of cases, going overboard. Despite what their coaches may say, their intestines are not all that much different from the team that loses a bunch of close games. If you want to judge a team’s ability in close contests, it’s best to look at their execution. The problem is, this might not be consistent with the outcome of the game.
Of course, this won’t stop people proclaiming that a team that wins close games is truly clutch. And they are going to look brilliant slightly north of 50% of the time after that team’s next close game. You’ll have no choice but to buy in. But that doesn’t mean you’ll be right.