By Shaun K.
A couple weeks ago, fresh off of the Jayhawks’ loss at Mizzou Arena, Terez Paylor, the Tigers beat writer for the Star, attempted to tackle the issue of foul calls in Big 12 conference play. It was a valiant effort by Paylor to try and find out which team has the best home court advantage in terms of getting calls, but he’s a much better reporter than he is a statistician. The problem with the article was a faulty methodology that made it appear Mizzou gets the most favorable treatment by officials – or at least that was the conclusion reached by commenting Jayhawks fans still sour from the game-changing charging calls at Mizzou Arena.
A couple weeks ago, fresh off of the Jayhawks’ loss at Mizzou Arena, Terez Paylor, the Tigers beat writer for the Star, attempted to tackle the issue of foul calls in Big 12 conference play. It was a valiant effort by Paylor to try and find out which team has the best home court advantage in terms of getting calls, but he’s a much better reporter than he is a statistician. The problem with the article was a faulty methodology that made it appear Mizzou gets the most favorable treatment by officials – or at least that was the conclusion reached by commenting Jayhawks fans still sour from the game-changing charging calls at Mizzou Arena.
Here’s the problem with the data as it was presented – it said Mizzou had the biggest advantage because no other team at home had a larger gap between home fouls and away fouls. Sounds logical, right? But consider this – what if you’ve got a team that goes only seven deep, only two men who can guard the 5, and the other players are guards with quick hands? Not only could you not afford to get into foul trouble, but you’ve also got a team that’s made of guys who can pick pockets and are going to have a harder time creating enough contact to get called for blocking. In short, Paylor’s data doesn’t account for a team’s own inclination to foul; it makes an assumption that all teams foul the same and then based on home court, some teams get officials in their pocket. But I’ve already convinced you why this isn’t right.
So if each team has a different disposition towards fouling, then the only way to measure their home court advantage is to know how that team fared on the road. Missouri, for instance, while getting the least fouls called on them at home, are still only seventh in the conference for getting fouls called on them as an away team – they just don’t foul. This gap between home calls and away calls is the true home court advantage – a team that gets called for the same amount whether at home or away, thus has no advantage. To get a more complete picture, we have to compare not only the gap between each team while at home and away, but their competition too. Each of these two gaps gives us total home court advantage. So we add them together and compare. The results tell us that while Mizzou is near the top, they’re only third, while Texas is number one, and Kansas is number two.
One other interesting point – with the exception of Texas and Baylor, each team’s ranking in home court advantage is very close to where each team is in the standings. Baylor’s lack of home court advantage is likely due to them playing mostly a zone defense, in which they draw much less contact for reach-ins and blocks. Even over-the-backs are reduced because of the spacing when going up for rebounds. Texas, though, is a bit more mysterious – Deloss Dodds, perhaps?
| ***The more positive the Diff the greater the home court adv. | ||||||
| a | b | c | d | e | f | g |
| Fouls | Opp Diff | Rank | Team Diff | Rank | Total Diff | Rank |
| Baylor | 3.14 | 4 | -1.67 | 10 | 1.47 | 10 |
| Iowa State | 6.67 | 1 | -0.5 | 9 | 6.17 | 5 |
| Kansas | 2.76 | 6 | 4.53 | 4 | 7.29 | 2 |
| Kansas State | 3.64 | 3 | 3.12 | 5 | 6.76 | 4 |
| Missouri | 0.9 | 9 | 6.24 | 1 | 7.14 | 3 |
| Okla. State | 2.38 | 7 | 2.29 | 7 | 4.67 | 7 |
| Oklahoma | 1.29 | 8 | 2.81 | 6 | 4.1 | 8 |
| Texas | 3.14 | 4 | 5.5 | 2 | 8.64 | 1 |
| Texas A&M | 3.9 | 2 | 1.47 | 8 | 5.37 | 6 |
| Texas Tech | -1.9 | 10 | 4.57 | 3 | 2.67 | 9 |
| Paylor's Data | ||||||
| Home Games | ||||||
| Team | Opp. fouls | Fouls | Diff. | |||
| Baylor | 19 | 18.67 | 0.33 | |||
| Iowa State | 20.67 | 16.5 | 4.17 | |||
| Kansas | 18.33 | 15.33 | 3 | |||
| Kansas State | 23.14 | 19.71 | 3.43 | |||
| Missouri | 18.57 | 11.43 | 7.14 | |||
| Okla. State | 18.67 | 18 | 0.67 | |||
| Oklahoma | 17.29 | 16.86 | 0.43 | |||
| Texas | 23.14 | 17 | 6.14 | |||
| Texas A&M | 19.33 | 17.67 | 1.66 | |||
| Texas Tech | 18.43 | 18.43 | Even | |||
| Road Games | ||||||
| Team | Opp. Fouls | Fouls | Diff. | |||
| Baylor | 15.86 | 17 | -1.14 | |||
| Iowa State | 14 | 16 | -2 | |||
| Kansas | 15.57 | 19.86 | -4.29 | |||
| Kansas State | 19.5 | 22.83 | -3.33 | |||
| Missouri | 17.67 | 17.67 | Even | |||
| Okla. State | 16.29 | 20.29 | -4 | |||
| Oklahoma | 16 | 19.67 | -3.67 | |||
| Texas | 20 | 22.5 | -2.5 | |||
| Texas A&M | 15.43 | 19.14 | -3.71 | |||
| Texas Tech | 20.33 | 23 | -2.67 | |||
Thanks to Terez Paylor (or a Star intern) for compiling the foul data that I needed to do this analysis. The data is missing a few games since it’s a couple weeks old, but there’s a large enough sample size that we’re okay.
Thanks for that. I thought the same thing reading his article. I'm not sure you can compare teams inside of individual games at all. I think this has to be a comparison of one team's home game fouls versus their away game fouls.
ReplyDeleteIf an individual team varies their fouls little from their season mean at home versus on the road (assuming away other variables such as injuries or a reffing crew known to call more fouls than most or let teams play more than most, rate of drawing charges, end of game fouls which would have a great deal to do with this statistic I am sure, etc etc), then that team likely has very little home court advantage. Conversely, if a team has a wide variance from the season mean when looking at home versus road, then they would enjoy a large home court advantage. Either way, absent some huge regression taking into account every factor we could come up with to remove bias from the equation, any attempt at this is severely flawed and for Paylor to suggest otherwise is asinine.
Just for fun, I took a look Mizzou and KU in their first 3 home and away games (6 total) to see what my oversimplified comparison says in a much to small sample size.
Mizzou's variance from the mean is +/- 2.5 while KU's is +/- 1.667 so I would say that Mizzou enjoys the better home court refereeing advantage (gets away with more fouls) than does KU.
I wish I had an intern so I could have him or her look at the written game logs to remove what are essentially intentional fouls from the stats so we can get a better idea of what sort of home court advantage exists without the fouls in the final minute of a game that have little to do with judgement on the part of an official. Wouldn't a team that wins most of its games be more likely to be fouled more in the last 2 minutes of the 2nd half?
ReplyDeleteAbsolutely. That has to be one of if not the biggest factors creating bias. That may help explain the Texas bias. In Austin, Kansas came back in the last few minutes against them as did Missouri if I remember right. Then, Texas beat ISU by 7 at home, OU by 8, A&M by 10 and OSU & KSU by 11. That is 7 of their 9 home games where intentional fouls were likely to have been committed; whereas, Mizzou and KU have largely blown opponents out of the water at home. Too bad you don't have a team of statisticians to get a better look at those trends.
Delete