The Grizzlies Crunch Time Stats Broke Logic

With their most recent 103-101 victory at home against the Detroit Pistons, the Memphis Grizzlies moved to 12-6 on the season in games where “crunch time” was played– as defined on NBA.com as games where the score was within five points in the last five minutes of the game. In just 69 minutes of play, the Grizzlies have outscored opponents by 44 points in crunch time, and their net rating of 25.3 ranks second in clutch situations through games of January 14.

That’s perfectly fine and there’s nothing particularly crazy about those statistics, after all, the Warriors have been even better than the Grizzlies and have outscored opponents by 39.2 points per 100 possessions in 74 minutes.

However, the Grizzlies’ crunch time prowess becomes really crazy when you look at the past five seasons of data. Since the 2011-12 season, the Grizzlies have outperformed their pythagorean wins (expected wins based on points scored and allowed), and are once again after just half the season has been played six games ahead of where they should be. The odds of hitting tails, if that’s your desired outcome, five times in a row in a coin flip is one in thirty-two, and the over the past five seasons the Grizzlies have hit that and much more, winning 20 extra games than would have been expected.

As shown by the graph above, there have been two outliers over the past 5 seasons in addition to the Grizzlies. The Nets have outperformed by 17 wins, the Timberwolves have underachieved by 19 wins– largely due to the best team of the Kevin Love era, which by schedule-adjusted net rating is the best team ever to not make the playoffs. Fun fact, the second best team ever was the Phoenix Suns in the same season.

Over the past 5 seasons, the Grizzlies’ net ratings for the season have been (starting from this season going backwards): -2.8, 3.1, 1.2, 4.2 and 2.1.

In crunch time, those numbers have jumped to: 25.3 (69 total minutes), 14.0 (195), 26.1 (166), 8.4 (171) and 15.1 (145)

And in plus-minus, the Grizzlies have ranked: 3rd, 3rd, 1st, 3rd and 5th.

Consider that the average position for every other team over a five year sample has been between 10 and 20. The Grizzlies on average rank third! Overall, the Grizzlies have a 129-70 record in those games for a 64.8 win percentage. In the last three seasons, the Grizzlies are 75-36 and have a 67.6 win percentage.

For the rest of the league, the distribution of crunch time performance is relatively random over samples of five years, which makes sense. If you’re a plus team, you’ll probably find yourself more often in the “ahead” part of  “ahead or behind by 5 points or less”, but in such a short period of time with small margins in point difference, the probabilities get squeezed and there’s only so much you can do to add to your win probability.

Taken as a random 50/50 coin flip, the Grizzlies winning at least at a 64.8 to 67.6 win percentage rate in the past three to five years, has roughly a 0.013 to 0.015 percent chance of happening– translating to at best one in a few thousand chance. A statistic so wild and improbable I actually find it offensive and to be violating of common sense.

Over the past year, I’ve been posting numbers on Twitter continuously about the Grizzlies unbelievable over performance in terms crunch time play and thus expected wins, and the first answer typically is that it might have something to do with the Grizzlies slower style of play and grind it out style.

I think there’s one thing we can be relatively sure about– that explanation cannot possibly be the correct answer, at least it can’t be largely responsible. There’s just no way to effect the amount of possessions in such a small period of time consistently in a meaningful way, and close games are close games, regardless of how slowly you play them. My guess would be that the Grizzlies have executed well and put themselves into position to succeed in crunch time consistently over the past five seasons, but the bulk of the out-performance just has to be an amazing and lucky coincidence that is impossible to replicate.

The Grizzlies have clearly slowed down this season, and by their net rating you would have expected them to fall already in some of the previous years in the standings. Yet they’ve still been amazing in close games, being 12-5 in games this season that were within 5 points during the last 3 minutes of play. If they ever fall to below .500 in close games, the grit-and-grind era for the Grizzlies may already be over, even more so than people realize.