Some Statistical Calculations

[HUSKERCR]

I finally had some time (without my boss noticing) to do a little number crunching on our favorite topic, Big 12 referees

during NU games. Here's what I found:

 

 

Ave. Number of NU Penalties/Game, Ave. Number of NU Penalty Yds/Game

All Games but Texas and A&M

 

7, 62.1

 

Ave. Number of NU Penalties/Game, Ave. Number of NU Penalty Yds/Game

Texas and A&M

13, 119.5

 

Note both the Texas and A&M games were officiated by the same Big 12 crew.

 

Based on a One Way Analysis of Variance (ANOVA), the chances of this variation

being due to random chance is .34% (1 in 300) for number of penalties, 1.9% (1 in 50) for total yards.

 

In other, non-technical terms: THE FIX WAS IN!!!!!!!

 

Opponent number of penalties is also lower with this crew by 37%, total opponent penalty yards 21% lower.

 

This "homer" Big 12 officiating crew doubles the number of NU penalties and penalty yardage while

calling significantly fewer penalties and fewer penalty yards on our opponents.

 

 

I found it interesting.

[Chucktown Husker]

What is the P value? Is this statistically significant?

Just a little stat humor.

[Sparker]

hrmmmmm....

[HUSKERCR]

What is the P value? Is this statistically significant?

Just a little stat humor.

 

 

The P-values are given.

 

Want the F-values as well?

[HUSKERCR]

Personally, based on this, I'd like to give

those homer refs a t-test with a nutcracker....

[Hujan]

Nice job, HUSKERCR!

 

Three questions:

 

1) Why did you not include the Iowa State game? It was the same officiating crew.

 

2) Does anyone know if this was the same crew as last year's CCG against Texas?

 

3) Can you run the numbers to see what the variation is in each individual game by that crew compare to the other games, and then see what the statistical likelihood is that you would randomly see that variation each time the crew officiates?

 

In other words, if that crew is off just once, may not be a big deal. But for them to be so out of sorts on multiple occasions is like one in a million. Make sense?

 

Edit: I have also heard a stat that our opponents have not been called for holding in the last 5 games. I am curious how rare that is and what the likelihood is that that is by random chance. Any takers?

[HUSKERCR]

Nice job, HUSKERCR!

 

Three questions:

 

1) Why did you not include the Iowa State game? It was the same officiating crew.

 

2) Does anyone know if this was the same crew as last year's CCG against Texas?

 

3) Can you run the numbers to see what the variation is in each individual game by that crew compare to the other games, and then see what the statistical likelihood is that you would randomly see that variation each time the crew officiates?

 

In other words, if that crew is off just once, may not be a big deal. But for them to be so out of sorts on multiple occasions is like one in a million. Make sense?

 

Edit: I have also heard a stat that our opponents have not been called for holding in the last 5 games. I am curious how rare that is and what the likelihood is that that is by random chance. Any takers?

 

 

1. Don't know.

2. I didn't include the Iowa State game because I didn't know the same crew was there.

It wouldn't affect the results much since it was one game out of 9 averaged.

3. I would do the crew analysis if someone can dig up which crews worked which games.

 

 

If I could find a listing of specific penalties/game I would also do the probability on the holding calls.

[bball_backer]

You're the man HUSKERCR. I don't mind crunching a few numbers and figuring avgs., but that probability stuff makes me loopy trying to do. Thanks for your work!

[Hujan]

Nice job, HUSKERCR!

 

Three questions:

 

1) Why did you not include the Iowa State game? It was the same officiating crew.

 

2) Does anyone know if this was the same crew as last year's CCG against Texas?

 

3) Can you run the numbers to see what the variation is in each individual game by that crew compare to the other games, and then see what the statistical likelihood is that you would randomly see that variation each time the crew officiates?

 

In other words, if that crew is off just once, may not be a big deal. But for them to be so out of sorts on multiple occasions is like one in a million. Make sense?

 

Edit: I have also heard a stat that our opponents have not been called for holding in the last 5 games. I am curious how rare that is and what the likelihood is that that is by random chance. Any takers?

 

 

1. Don't know.

2. I didn't include the Iowa State game because I didn't know the same crew was there.

It wouldn't affect the results much since it was one game out of 9 averaged.

3. I would do the crew analysis if someone can dig up which crews worked which games.

 

 

If I could find a listing of specific penalties/game I would also do the probability on the holding calls.

 

Alright, nevermind. It looks like last year's CCG crew was an "all-star cast." Although there are some common names, the crew was mostly different:

 

2009 CCG Refs: http://espn.go.com/blog/big12/post/_/id/7868/officials-revealed-for-big-12-championship-game

 

Suspect Officiating Crew: http://www.huskerboard.com/index.php?/topic/48925-officials/page__view__findpost__p__707031

 

Anyway, I'd love to see your calculations using the Texas/Iowa State/Texas A&M games. (Although, of the three, I don't really remember a lot of officiating BS during Iowa State, with the exception of the meaningless flag on Legate, which, was probably not unjustified.)

[schuhbdoo]

I finally had some time (without my boss noticing) to do a little number crunching on our favorite topic, Big 12 referees

during NU games. Here's what I found:

 

 

Ave. Number of NU Penalties/Game, Ave. Number of NU Penalty Yds/Game

All Games but Texas and A&M

 

7, 62.1

 

Ave. Number of NU Penalties/Game, Ave. Number of NU Penalty Yds/Game

Texas and A&M

13, 119.5

 

Note both the Texas and A&M games were officiated by the same Big 12 crew.

 

Based on a One Way Analysis of Variance (ANOVA), the chances of this variation

being due to random chance is .34% (1 in 300) for number of penalties, 1.9% (1 in 50) for total yards.

 

In other, non-technical terms: THE FIX WAS IN!!!!!!!

 

Opponent number of penalties is also lower with this crew by 37%, total opponent penalty yards 21% lower.

 

This "homer" Big 12 officaiting crew doubles the number of NU penalties and penalty yardage while

calling significantly fewer penalties and fewer penalty yards on our opponents.

 

 

I found it interesting.

 

 

Send that to Sean Callahan. I have felt the Big 12 can't stomach NU winning the title this year. It only takes a game or two to throw a loss our way. Great work.

[zoogs]

So you're saying the fix was on because the penalty distribution wasn't random?

[TripleOption]

This is pretty cool -- Thanks! But, I think the default position would be the penalties are not random events. You've got to do something against the rules to incur one (Osborne's roughing call excepted apparently). So you could say that this proves that the events are not completely random, and one possible explanation is that Nebraska committed more penalty worthy infractions in these two games. Or it could just be that this crew was out to get us. But in fairness, we did earn a lot of those penalties...

 

Ideally, you might find every illegal action in every game, by both teams, and then see how many of them were called by different crews on each team. This would be more statistically valuable I think (but admittedly all but impossible). I think, at minimum, you should exclude false starts, illegal formation, delay of game and penalties like that which clearly aren't random.

 

Man, I love stats. Great post.

[Glendower]

This isn't really the interesting statistic-- I think it would be more interesting to look at the calls made by that officiating crew or ALL opponent penalties vs against NU; or all penalties by the teams played the very penalty-lopsided games vs their penalties against NU.

 

Looking at penalties called against NU is a sketchy way to go; even in the discussions about the A&M game, most nu fans had to admit that most of the 16 calls were legit. Now, it's hard to go back and verify non calls, but it is possible to show how statistically anomalous some of our opponents have been in terms of penalties called compared to all of their other performances.

 

I guess I'll just have to call to reactivate SPSS tomorrow (I hate that they adopted an activation scheme there)

[huKSer]

As some author from some state of east of Kansas said:

1) Lies

2) Damn lies

3) Statistics

[TripleOption]

This isn't really the interesting statistic-- I think it would be more interesting to look at the calls made by that officiating crew or ALL opponent penalties vs against NU; or all penalties by the teams played the very penalty-lopsided games vs their penalties against NU.

 

Looking at penalties called against NU is a sketchy way to go; even in the discussions about the A&M game, most nu fans had to admit that most of the 16 calls were legit. Now, it's hard to go back and verify non calls, but it is possible to show how statistically anomalous some of our opponents have been in terms of penalties called compared to all of their other performances.

 

I guess I'll just have to call to reactivate SPSS tomorrow (I hate that they adopted an activation scheme there)

 

Good call. I like the idea of looking at our opponents penalty total versus NU as compared to their other B12 officiated games.