Devon Yau

In today’s NBA, a hot topic of fan debate centers around officiating. Are they making too many calls? Are referees actually biased or just blind? Well, in the 06-07 season, veteran referee Tim Donaghy (pronounced DAH-nuh-ghee) was actually biased! Between December 2006 and March 2007, this 13-year NBA refereeing veteran quietly accepted cash payments to tip gamblers on games he officiated.

But did this biased officiating actually have an effect on game outcomes?

To address this question, I sifted through a play-by-play dataset of the 2006-2007 NBA regular season from sports-statistics.com to compare periods in which Donaghy was active versus inactive.

Note: Dataset has changed since original source date. See code comments for more details.

The black boxplot represents Donaghy’s inactive period, and the orange represents his active period. To see if foul calls were uneven between teams, the Y axis represents the absolute difference in fouls called for each team. Higher values represent games where one team had more fouls called against them—potentially indicating biased refereeing. Of course, foul numbers don’t necessarily matter as much as their context, but let’s take a look.

Initially, we don’t see a huge difference between the two periods! At first, it actually seems as if Donaghy called games more evenly than other referees. But if we take a look at the total number of games with unusually high foul differences in each period, the numbers tell a different story.

Games with unusually high foul spreads during Donaghy active vs. inactive periods
Active window Number of games with high outlier foul spreads
Donaghy Inactive 9
Donaghy Active 19

Donaghy had 10 more high foul spread games than the rest of the league’s referees! To be fair, there are 3 on court referees at the same time for each game, but this difference feels too high to be mere coincidence. Let’s take a closer look at the games with the highest foul differentials.

Games with the Largest Outlier Foul Spreads
Absolute Foul Spread Game ID Game Number Window Status Home Fouls Away Fouls
18 20600577 577 Active 34 16
17 20600725 725 Active 15 32
16 20600634 634 Active 14 30
16 20600670 670 Active 24 40
16 20601047 1047 Inactive 16 32

While we don’t have data about the teams’ seasons or records, a quick google search tells us that game 725, on February 7, 2007, was a particularly close game between the Toronto Raptors and Orlando Magic. Both teams were close playoff contenders coming into the game, but with Toronto shooting 38 free throws to Orlando’s mere 15 in this game, Toronto escaped with a 113-103 point win. Entering the final quarter, Orlando was down just 1 point but was buried in fouls as Toronto outscored them 36–27. In a close game that both teams needed to win to maintain playoff dreams, lopsided foul calling heavily affected the game outcome. Hm. Sound familiar, 2026 OKC Thunder?

Anyways. If I were a bookie in the 2000s and noticed that Donaghy was rigging games, which teams would I bet on? On the X axis below, we have total fouls called against each team (leading to free throws shot by opposing teams), and on the Y axis, we have the top 10 teams in this metric. Below, we see that Chicago and Atlanta lead the pack—penalized with 1275 and 1237 fouls called against each team respectively.

So—what can we learn from this? Fouls play a huge role in NBA game outcomes! (A code that the OKC Thunder seem to have cracked). Additionally, this case implies that corrupt officiating could be ongoing and still affecting outcomes, so us data-science-loving NBA fans must stay vigilant and keep referees accountable.

And above all, we learn to always take the under on the Atlanta Hawks! As a salty resident of Atlanta who hates the Hawks, I approve this message.