Monday, January 29, 2007

Penalties – Nearly Irrelevant

Penalties, in college football, are irrelevant, at least when it comes to winning games.

The Sunday Morning Quarterback has been doing an in-depth statistical analysis of every major conference this season with the view towards determining which statistics are most relevant to winning. Most recently, he did an analysis of the Pac 10.

I have seen it suggested –repeatedly in fact - that the commission of a great deal of penalties is “poor coaching”. Well, statistically speaking at least, if penalties are "poor coaching", then a lot of poor coaches are winning a lot of football games.

In every analysis done by the SMQ thus far, penalty yards have been at or next to the bottom of the least significant statistic in winning a college football game. In fact, teams with more penalty yards are more likely to win than lose in every conference he has analyzed so far.

I have long suspected that certain types of penalty yards are more often than not a result of complicated offensive schemes. I have also suspected that these types of penalties, like delay of game’s and illegal motion’s, while frustrating have little effect on the game outcome. In this way of thinking, penalties are in fact merely an irritating symptom of an overall effectively operating offensive machine.

After reading the SMQ’s work, I decided to take a look at the Gator’s 2006 MNC season to see if the same held true.

The following list is the games in which Florida was more penalized than their opponent, and the outcome (team, yards penalized, score)–

Florida 102 – LSU 30: 23-10
Florida 90 – Western Carolina 35: 62-0
Florida 75 – Georgia 50: 21-14
Florida 71 – UCF 30: 42-0
Florida 71 – Kentucky 66: 26-7
Florida 67 – Arkansas 25: 38-28
Florida 65 – FSU 27: 21-14
Florida 65 – Tennessee 34: 21-20
Florida 53 – Alabama 26: 28-13
Florida 53 – Vanderbilt 33: 25-19

There were 2 games in which Florida was the least penalized team –

South Carolina 85 – Florida 49: 17-16
Auburn 40 – Florida 33: 17-27

So the Gator’s least penalized game this year was their only loss. The second least penalized game came to Spurrier’s Gamecocks, in a close win.

Games with huge penalty yards were generally comfortable wins.

(In the BCS title game the Gators and the Buckeyes both had exactly 50 penalty yards).

Certainly Steve Spurrier’s offensive juggernauts of the 1990’s were among the most penalized teams on an annual basis in the SEC. Many of those visor tossing incidents were a direct result of some delay of game penalty on a complicated play that the offense couldn’t get into motion in the allotted time.

I also suspect, though I have not fully run the stats, that penalty yards are closely linked to time of possession, especially in teams with complicated offensive schemes. Certainly this was true in the Auburn game, where Auburn won the possession battle 36:43 to 23:17.

In other words, I wish the Gators had be given the time to have committed more penalties in that game.

Certainly in the SEC, and elsewhere this year, having few penalties in no way corresponded to success, as the most penalized were Florida and Arkansas.

The least penalized team in the nation?

4-8 Northwestern.

If penalty yards are an indicator of “poor coaching”, then the best coached (fewest penalized) teams by major conference are –

ACC – Virginia
Big 10 – Northwestern
Big 12 – Iowa State
Big East – Connecticut
Pac 10 – Stanford
SEC – Vanderbilt

Now, anyone can name or recall some game where some ill timed penalty led to a situation where the other team won. Likewise, one can recall games where missed extra points have cost a game, or turnovers, or any other conceivable game situation relevant to college football. However, the incontrovertible statistical proof indicates that penalty yards do not hurt the team, and that the team with more yards penalized wins more often than not.

In some ways, this makes perfect sense. Take Florida’s most penalized game this year (LSU), and the 14 penalties committed. Of the 14, 5 were defensive, 2 special teams, and 7 on offense. The offensive penalties were 3 delay of games, 3 false starts, and 1 illegal shift. In the game, Florida had a longer possession time than LSU (32:35-27:25), and thus more opportunities to commit the type of small offensive penalties that complicated offenses seem to do so often. In Florida’s least penalized game (Auburn), the lack of possession time equaled few penalties.

Now, I am in no way “advocating” the commission of penalties as some sort of game winning strategy. However, it is pretty clear that highly penalized teams are neither poorly coached nor prone to losing.

And if you want the type of offensive strategies that come with a coach like Spurrier or Meyer, you had better get used to the penalties (South Carolina committed the second most total penalties, behind Florida, in the SEC this year).

I, for one, am willing to trade those annoying little penalties for titles.


Matt said...

I think the biggest differentiation to make here is the difference between offensive and defensive penalties. I'm going completely off the top of my head here, but most offensive penalties that UF committed over the course of the season resulted in the lowest possible penalty: 5 yards, no loss of down. This makes it slightly harder to get a first down, but doesn't take away your chances to do just that.

However, defensively, the penalties are much more hurtful. Most of the penatlies against the defense are probably personal fouls (15 yards, first down), holding (10 yards, first down) or pass interference (15 yards, first down). These have a huge impact, as put the opossing offense signficantly closer to endzone and give them signficantly more chances to get there.

Henry Gomez said...

I agree with the idea that defensive penalties are more hurtful to you chances of winning. If we were to look at the ratio of offensive to defensive penalty yards for the Gators in 2006, I'd have to guess that offensive penalties made up the majority. We also had a lot of special teams penalties in the form of illegal blocks to the back and holding on kick returns that cost us points and field position.

Anonymous said...

I think that psychologically, referees are more likely to penalize stronger teams, particularly when the disparity is great. I have no data, but it would be interesting to check what trends exist when a top 10 team is playing an unranked team. (Current rankings, not end-of-year rankings, of course.)

Anonymous said...

Matt - The total number of penalties doesn't matter in my analysis or Mergz's; it's the total yardage. One 15-yard pass interference is the same as three 5-yard offsides fouls.

I don't notice any trend going through these games favoring stronger teams being penalized more, but I do notice the final penalty numbers tend to be pretty close - you don't see many games where one team has 85 yards, and the other 20. They're usually in the same vicinity of one another, though your numbers on Florida (I haven't looked at the SEC yet) don't show that except against Georgia, Kentucky, Auburn and maybe Vanderbilt. Anyway, I think penalty yardage is largely the result of how officials are calling the game rather than how the teams actually play it, and they might have an incentive to "even it up" a bit if there seem to be a disproportionate flags going in one direction, because that situation when it happens never comes across as fair.

Again, your numbers with Florida don't bear that. And I remember Texas Tech consistently had ridiculous penalty yardage, regardless their opponent's total. So that's not an observation I necessarily expect to stick if it was looked at more closely.


Anonymous said...

All of this proves nothing but the law of probability. The Gators have gotten screwed since Galen won the league in '84 when we were recognized as a threat to the rest of the SEC. We were soon-after labeled as "cheaters" and were penalized accordingly. 23 years later, nothing has changed except for the ridiculous number of games that we have lost, directly or indirectly, as a result of crappy calls and/or non-calls. You can show all of the coincidences and trends possible, but it all boils down to calling penalties both ways. Until that happens, you can bet your ass that refs can and do affect the outcome of games.