Home / Sports / Football / Regression to the mean: what does it mean for (A)U?

Regression to the mean: what does it mean for (A)U?

Can Foot Lauderdale do it again?

The offseason is a time for research, and prompted by a question reader Luke, I decided it might be useful to take a sweeping look at that alternately cruelest and sweetest of mistresses: Regression to the Mean.

It’s a pretty simple concept: where a football team has gotten lucky in one season, it should be less lucky the year after, and vice versa. So where was Auburn lucky last year? Where were they unlucky? Where we could we expect random improvement, and where might be see random decline?

We’ll start with a set of generalized metrics and then move on to ones specific to Auburn in 2009-2010:


Turnovers: With certain exceptions (almost all of which are named “Pick-Prone Quarterback”), turnover margin appears to be almost entirely random and sees huge fluctuations year-to-year.

But this doesn’t much matter for Auburn this year: at a middle-of-the-pack +2 margin in 2009, Auburn’s not due for any big swings either way. Doesn’t mean it won’t happen, but whether said swing would be a positive or negative, there’s no telling.

The same goes for fumble luck–fumble recoveries are widely seen as random events and can swing from year-to-year, but Auburn recovered 22 of the 43 total fumbles in their 2009 games, almost exactly 50 percent. They were neither lucky nor unlucky.

Record in close games: A team like 2007 Miss. St. or Vanderbilt in 2008 that wins all of its games by the hair of their chinny-chin without the benefit of actually, you know, playing better than their opponent, is almost always due for some retribution from the gods of fate the year after.

But again, Auburn’s luck in this department in 2009 wasn’t extreme enough in either direction to mean much for 2010. The Tigers went 2-3 in one-possession games, edging Tennessee and Northwestern but losing to Kentucky, Georgia, and Alabama. (Note here that Tennessee may not have actually been a close game, what with the touchdown on the very final play of the game, but then again the West Virginia game was obviously a much closer contest than the 11-point margin. I figure it evens out.)

I do think Auburn’s more likely to edge back over .500 in this department this fall than fall below it again–having gone a combined 4-7 the past two seasons–but being a game (or even two, heaven forbid) under again wouldn’t be too much of a shock.

Injuries: Check out Phil Steele’s start chart for Auburn in 2009. (It’s a PDF, FYI.) The red blocks are starts missed. Count how many Auburn had outside of the two safety positions. Now, take out the two weeks Byron Isom missed via suspension. So how many missed starts are left?

One. Eltoro Freeman’s, against Alabama. Yes, Auburn should definitely see better health at safety this year, and yes, from McCalebb’s ankle to Coleman’s arm to Will Herring’s heel it’s not like the Tigers were totally devoid of injury trouble aside from McNeil’s leg and Etheridge’s neck. Still: when the 20 non-safety positions combine to see just one start missed for injury reasons, it’s safe to assume Auburn’s not going to be quite that fortunate again this season. No complaints, since the Tigers should be much better equipped to deal with injury losses than they were last year, but I think it’s safe to assume that new depth will be a bit more tested in 2010.

(Then again, the Tigers just made it through a spring camp where the only injury expected to last into the fall’s was to third-stringer Frenchy, so knock on wood, maybe the bug will pass by Auburn’s door again.)

Conversions: As I understand the current statistical thinking from people like Football Outsiders (and someone correct me if I’ve got it wrong), third and fourth-down conversion rate isn’t entirely random–obviously a bad offense that leaves itself with 3rd-and-7 over and over again isn’t going to convert at a high rate. But a team that has a generally healthy offense on first and second downs but doesn’t translate that success into conversions may be the victim of bad luck.

If that’s the case, it doesn’t seem to indicate anything vital for the Auburn offense on third downs–filter out the nonconference games and the Tigers finished seventh in the SEC in yards per-play and, sure enough, seventh in third-down conversions. (That Chris Todd finished ninth in the league in YPA against conference opponents may be the culprit.) However, Auburn did only convert 4 out of 14 fourth downs–the 12th-worst rate in the country. That doesn’t jibe with the rest of Auburn’s offensive accomplishment, so maybe we’ll see an uptick there.

Defensively, Auburn may have been a little fortunate; they finished 8th in yards-per-play but a robust second–behind only the Tide–in stopping opponent’s third downs. (Again, that’s SEC games only.) Even amongst what I firmly expect to be widespread defensive improvement, that ranking might drop a bit. Fourth downs are the opposite story, though–Auburn was worst in the league and 104th in the country in stopping fourth downs. (You ask me, that Auburn could be so good at third downs and so bad at fourth downs seems to be further evidence results on those downs can be random, no?)

So: the offense’s rate should rise as the offenses gets better. The defense’s might decline, but maybe they can make up for it (amongst other ways) by not permitting so many fourth-down makes.

Yards-per-point. A Phil Steele original metric, as best I know. As you’d expect, teams that score a bunch of points without actually moving the ball regress towards needing more yards the following year, while defenses that give up a bunch of a points while still not allowing much yardage tend to have that balance out.

Which is where Auburn finds itself this year. Thanks to the Tigers’ red zone efficiency and defensive red zone inefficiency (about which more in a moment), both sides posted very low YPP numbers in 2009: 12.963 for the offense, 13.594 for the defense.

Meaning–as Steele would no doubt tell you–that the Auburn offense may have to work a little harder for scores in 2010, but that the defense should catch a few more breaks … and could stand to get a few more stops in the red zone.


Punt returns. No one’s claiming that punt return numbers are random accidents, but a major program like Auburn finishing 113th in the country in the statistic is the sort of outlier that truly can’t help but improve this season. (Look at the bottom 20: there’s only a handful other BCS programs down there, and Penn St.’s the only one besides Auburn with a winning record.) Finding a returner that can hold onto the ball will be a huge help, but since you don’t wind up that gawd-awful without some bad luck as well, Auburn should see some better results almost regardless.

FG accuracy. Wes Byrum’s remarkable 15-of-16 season remains as eye-popping as the day it was completed, but I wouldn’t expect him to come quite that close to perfection. After the sort of ginormous step forward he took last year, he may regress just a hair to his freshman/sophomore form

He should still be very, very good, one of the best kickers in America–but as he gets more attempts and the coaches trust him to try them at longer distances, you’ll probably see him at something like 19-of-23 or 21-of-26 this year. And that’ll be fine. In a related story …

Red zone efficiency. In terms of simply putting some kind of points on the board, almost no one in the country was better than Auburn last year–the Tigers scored on better than 95 percent of their red zone opportunities, third in the country. Touchdown percentage was a little weaker, particularly in conference, but still awfully strong. As long as Byrum doesn’t collapse and Newton doesn’t throw substantially more picks than Todd I wouldn’t expect those numbers to regress all that much, but it stands to reason Auburn might get a couple more bad bounces inside the 20.

Not on defense, though. Roof’s Tigers finished 11th in the SEC in raw red zone conversions and dead last in TD percentage. Why was that YPP mark so low? Because almost every time an opponent got inside the 20, they put six on the board. Better depth and ability to deal with long drives should help immensely … but Auburn could also be due for a few more breaks in red zone D, too.

Total fumbles. Did you know that Auburn fumbled 28 times last year–more than all but 9 other teams?

Auburn can’t really expect to recover a greater percentage of their fumbles or throw fewer picks (that being one of the few things Todd did very, very well … up until the final two weeks of the season, anyway) but the turnover situation might improve anyway, thanks to the fact that Auburn should be hard-pressed to put the ball on the ground as freaking often as they did a year ago.


Yes, Virginia, Auburn’s defense is going to improve. Most of that will be  the greater depth and familiarity with Roof’s schemes and (hopefully) healthy safeties, but some of that will be that last year’s D was a shade better than it looked; wobbly as it was, it was unfortunate in the red zone and wasn’t helped by the  special teams issues. (The horrendous kick return coverage should also have nowhere to go but up.) Even if that third-down percentage goes up a bit, overall scoring should go down.

Offensively, Auburn’s going to have to earn their improvement. That YPP number is almost certainly going to go up as a few more red-zone possessions go empty. Someone important is going to miss some time with an injury. Auburn’s third-down problems last year aren’t going to get solved by luck alone. This by no means means we should dial down the expectations … just that we should understand that random chance doesn’t look likely to save the offense if the expected step up from Todd to Newton and overall  second-year-of-the-system improvement doesn’t materialize.

Special teams-wise, very little other than Byrum’s FG percentage has the ability to do anything but improve.

All in all, it’s encouraging how many positives there are … but there’s enough negatives with the injuries and potential for turnover problems that no one should go getting crazy confident just yet.

Photo via.

About WarBlogEagle

Check Also

Auburn 2017 National Signing Day: No news is good news

It feels like we’ve seen this exact headline every February since Cyrus Kouandjio went rogue, …