The NBA’s 0–3 Curse: Can it be Broken?
The Dallas Mavericks find themselves in a precarious position, down 0–3 against the Boston Celtics in the NBA Finals. While the odds of a comeback are slimmer than Derek Lively’s chances of winning a limbo contest, it’s not entirely impossible.
Game 3, played in Dallas, was a wild affair that could have gone either way. Assuming the Mavericks can hold their own at home in games 4 and (potentially) 6, they’ll just need to pull off an upset victory in game 5 at Boston to extend the series to seven games. It’s unlikely, but so is the idea of finding a parking spot right in front of your building in New York City.
Historically, no NBA team has ever come back from a 0–3 deficit to win a 7-game series, but a few have come tantalizingly close. In the Finals, we expect the two teams to be relatively evenly matched, but a three-game lead at the start suggests otherwise. Using a simple probability model that ignores home court advantage (because math is hard enough), we can estimate that the leading team has about a 61% chance of winning any individual game, like flipping a coin where one side is slightly heavier.
With a 61% win probability for the 3–0 team, the 0–3 team has about a 2% chance of making a comeback and winning 4–3. Out of 156 series that have started with a three-game sweep, we would have expected around 4 comeback victories. While we’ve seen none, our simple model that doesn’t account for factors like home court advantage or the moon’s gravitational pull seems close enough.
It’s also possible that the teams are actually closer to a 50/50 match, and the Mavericks just happened to flip tails three times in a row. It happens, just like finding a parking spot right in front of your destination. If the 0–3 teams were truly evenly matched but unlucky, we would have expected ten comebacks historically. Although this doesn’t align with the reality of zero comebacks, it suggests that a comeback isn’t an impossible feat, like convincing a cat to follow orders.
For the expected number of comebacks to be 0 (when rounding to the nearly whole number) from 156 series, the per-game win probability for the 3–0 team would need to be around 77%. It seems unlikely that a Finals team would be a four-to-one underdog in every game, unless they were playing against the Monstars from Space Jam.
So I’m sticking with my 2% chance for a Mavs comeback. They probably won’t do it, but I wouldn’t be surprised to witness such a comeback in my lifetime.
But please, basketball gods, never let it happen against my beloved Celtics.
Michael Bagalman is a data scientist and, until the Jazz return to their rightful home in New Orleans, a Boston Celtics fan.
