Sunday, April 29, 2012

2012 SEC SDPI

In our final BCS conference SDPI examination of the 2011 college football season, we come to the SEC, home of the last six national champions, and the home to both participants in the 2011 BCS National Championship Game. To see last year's post on the SEC, click here.

This first paragraph will explain how SDPI is calculated. So if you want the meat of this article skip on down. In the 2011 SEC regular season, conference play only, championship game excluded, the average SEC team gained and allowed 2612.33 yards. The standard deviation for yards gained (offense) was 511.33 yards. The standard deviation for yards allowed (defense) was 633.43 yards. Arkansas gained 3217 yards and allowed 2912 yards. Their offensive SDPI was 1.18 = [(3217-2612.33)/511.33]. Their defensive SDPI was -0.47 = [(2612.33-2912)/633.43]. Their total SDPI was 0.71. This number ranked 5th in the SEC.

Here are the 2011 SEC standings.

Now here are the 2011 SEC SDPI standings. The standings are sorted by division by total SDPI with ranking for each category (out of 12 teams) in parentheses.

Statistically, Alabama was head and shoulders above the rest of the league. Unfortunately, their kicking errors and lack of success in the red zone allowed LSU to win the SEC championship. Fortunately for them, enough upsets occurred elsewhere in college football to allow them a rematch with the Tigers in the BCS National Championship Game. And the rest, as they say, is history.

So Who Was Better Than Their Record Showed?
Vanderbilt qualified for a bowl game in James Franklin's first season as head coach, but their 2-6 SEC record was not indicative of their play. The Commodores were a hard-luck 0-4 in one-score games, losing close contests in Nashville to Georgia and Arkansas and on the road to Florida and Tennessee. Vanderbilt was also exceptionally dominant in their two league wins, beating Ole Miss and Kentucky by a combined 53 points.

So Who Was Worse Than Their Record Showed?
The easy answer here is LSU as the Tigers did not have the statistical profile of an unbeaten team. However, the correct answer is the other Tigers from Auburn. Auburn's four SEC wins came by a combined 39 points. Their four SEC losses came by a combined 125 points. Team's that post Auburn's statistical profile typically win about two of eight conference games. All things being equal, they should have switched records with Vanderbilt in 2011.

Conference Superlatives:

Best Offense: Alabama 1.37
Even when they posted back-to-back 8-0 SEC records in 2008 and 2009, the Crimson Tide only finished fifth in offense in the conference. The 2011 offense was incredibly consistent, gaining between 366 and 437 yards six times in their eight league games.

Worst Offense: Kentucky -1.49
After losing quarterback Mike Hartline and receiver Randall Cobb from the third ranked SEC offense in 2010, the Wildcats plummeted to the worst offense in the conference. The Wildcats were held to ten or fewer points in six of their eight league games!

Best Defense: Alabama 1.84
The Tide were uber-dominant on this side of the ball. Five of their eight SEC opponents were held below 200 yards and no team gained more than 239 yards against them!

Worst Defense: Ole Miss -1.40
Facing the top-four offenses in the conference (Alabama, Georgia, Arkansas, and LSU) didn't help their numbers, but the Rebels were bad anyway you slice it. Remember, they gave up 30 points to the worst SEC offense (Kentucky).

John L Smith: The 'L' is for Loser
As you may have heard, the Arkansas Razorbacks have experienced a change in their head coaching position. I won't dwell on the circumstances surrounding the ouster of Bobby Petrino, but a cursory Google search should lead you to the story if you've been out of the loop. However, I will focus on his replacement, one John L Smith. Arkansas marks Smith's sixth destination as a head coach. Smith previously coached at Idaho (when they were IAA), Utah State, Louisville, Michigan State, and Weber State (more on that later). I am of the opinion that Arkansas could not have done much worse in their search for a replacement for Mr. Petrino. I will let you know up front, what follows is a hugely biased hatchet job, not unlike the Fox News special report on gas prices that aired a few weeks ago. However, unlike Fox News, I admit my biases. Why do I have such disdain for Mr. Smith? Primarily because I feel that much like Greg Norman at a golf tournament, John L Smith finds a way to lose. I believe that his teams constantly underachieved relative to their talent level. How can one go about determining whether or not a team under or overperformed? One way is to look at a team's Pythagorean Record. A team's Pythagorean Record is their predicted record based on the ratio of the points they scored to the points they allowed. First developed by Bill James for use in baseball, a team's Pythagorean Record can give us a pretty good approximation of a team's record. With that in mind, I looked at the IA teams coached by John L Smith in the 12 seasons from 1995-2006. I took each of his team's record against IA foes and compared it to their expected record based on the Pythagorean Theorem. For fun, I also calculated each team's record in one-score games. The table below summarizes the results.

In 12 seasons, Smith's teams finished a little more than six games worse than where we would have expected them to finish based on their points scored and allowed. Is this a huge difference? I don't know. To answer this question, we would need to calculate how other coaches have performed relative to their Pythagorean Record over an extended period of time. Perhaps one day I shall embark on that brave undertaking. For now though, we can say that his teams underperformed their Pythagorean Record over 12 seasons and 137 games (a relatively large sample). His teams drastically underperformed during his four seasons in East Lansing, finishing over four games below where they would have been expected to finish based on their scoring ratios. In addition, over a relatively healthy sample size of 50 games, his teams won just 40% of their one-score contests (20-30 record).

In addition to the Pythagorean numbers, I also examined the results for each game Mr. Smith coached over those 12 seasons to see if I could find any head-scratching results. I found a few. In 1995, his Utah State Aggies lost to Boise State when the Broncos were still a IAA team. In 1997, his bowl bound Utah State Aggies lost to a North Texas team that finished 4-7. His second Louisville team lost to a 3-9 Army squad 52-59. Army scored just 225 points that whole season and were shut out twice. His first Michigan State team lost at home to a 5-7 Louisiana Tech squad in 2003. His 2004 team missed out on a bowl because they lost the season opener to Rutgers. Rutgers finished 4-7 that season and lost the very next week to IAA New Hampshire. That same Spartan team also provided one of the few bright spots in Tyrone Willingham's final season as coach of Notre Dame as the Irish beat Michigan State in east Lansing en route to a 6-6 record. Michigan State also lost to Penn State in 2004. As you may not remember, that Penn State team went just 4-7 and prior to their 37 point outburst against the Spartans, the Nittany Lions had averaged just a shade over 9 points per game in their first seven Big 10 games. 2005 featured this historic meltdown that can't be put into words.

Finally, in his Michigan State swan song, his charges blew a huge lead at home against Notre Dame and also lost by 25 points to an Indiana team that finished 5-7.Of course, I'll admit, if you look hard enough, you can find some fantastic performance from his teams, like the greatest comeback in college football history. On the other hand, what are you doing trailing Northwestern by 35 points in the first place?

Finally, while I obviously don't know Smith personally, his dealings with his employers have been less than forthright. He allegedly accepted the Michigan State job at halftime of Louisville's 2002 bowl game. He also left Weber State hanging by accepting the Arkansas job before coaching a game for the Wildcats.We'll see how Smith does with the great talent on hand at Arkansas, but consider me extremely skeptical of this coaching hire.