Dan Crawford is scheduled to be a referee in tonight’s Eastern Conference Finals Game 6 between the Heat and Celtics in Boston.
Given the fact that the refereeing has been a major topic of discussion in this series, the fact that Crawford will officiate tonight’s game might raise some eyebrows.
Because if you look at the numbers over the last several postseasons, the Heat seem to perform better in playoff games refereed by Crawford as compared to their other games in the span.
Last year, the fact that the Mavericks had a very poor record in playoff games reffed by Crawford was well-documented on ESPN.com (and other outlets) prior to Game 2 of their series against the Trail Blazers. It should be noted that the Mavs went on to win that game and their next game that Crawford reffed en route to an NBA title.
Heat in Playoffs Since 2004
In the nine postseasons since 2004 (the “Dwyane Wade era”), the Heat are 13-3 in playoff games reffed by Dan Crawford and 48-40 in all other playoff games in that span. To control for effects like home/road or playing better/worse opponents, it may be better to look at the expected win-loss for the Heat in those games given the pre-game point spread (betting line) for each game.
Historically, the spread has been shown to be an accurate predictor of what happens in a game – that is, teams are generally .500 against the spread over large samples of games. The Heat are 11-4-1 vs the spread in those 16 playoff games reffed by Dan Crawford since 2004, compared to 41-44-3 vs the spread in the other 88 games.
Going further, we can take the pre-game point spread for the Heat in each game refereed by Crawford and use it to calculate the probability that they should have won the given game (using the average win probability for an NBA team given that point spread). For example, the Heat were a nine-point favorite in Game 1 of the 1st Round against the Knicks (a game refereed by Crawford); given this, they had an 82.8% chance of winning the game.
Adding up their point spread-based win probabilities across all 16 games, the Heat should have won about nine games on average. Even more, given the point-spread based probability of winning each game, the chance that Miami would go 13-3 or better in those 16 games is 1.9%. Statisticians generally use a cut-off of 5% or lower for statistical significance, so this would be indicative that this trend goes beyond what would be reasonable by random chance.
Here is the breakdown of the Heat’s performance in postseason games refereed or not refereed by Dan Crawford since 2004. Notably, the Heat do not seem to get more fouls called or free throws attempted in their favor when Crawford refs, so the difference in performance can’t be directly attributed to foul calls, which is obviously something the referees contribute to significantly.
To be clear, we are not saying that Dan Crawford is biased toward the Heat, or that the NBA assigned Crawford to this game to increase the Heat’s chance of winning. The NBA assigns referees to each game prior to the series, not game by game. This is just some statistical evidence that may be worth considering given the importance of tonight’s game and the refereeing in this series.