AUSTIN, TX - NOVEMBER 5: Running back Joe Bergeron #24 of the Texas Longhorns gets a third quarter hit by cornerback Jarvis Phillips #21 of the Texas Tech Red Raiders on November 5, 2011 at Darrell K. Royal-Texas Memorial Stadium in Austin, Texas. Texas beat Texas Tech 52-20. (Photo by Erich Schlegel/Getty Images)
I've been bothered recently by valuations of Big 12 offenses that seem rather shallow in their methodology. For instance, a team that in conference play scored 35.8 ppg (5th in conference) and finished 2nd in the conference in turnover margin (+13) is routinely graded well behind other teams in estimations of the league's best offenses.
I think the error is in the confusion of the purpose and strategy for an offense. Typically if you asked someone, "What is the goal of the offense?" you would hear back "to score as many points as possible."
However, in reality that is merely a strategy for achieving the real objective: to score more points than the other team. If you score 40 points every game and yield 43 are you necessarily a better offense than the one that scores 26 but gives up 23? Maybe, or maybe not.
Consider the success of the Alabama offense against the 09 Texas defense in the Rose Bowl. On average, that offense scored fewer points per game than did the offensive unit from Norman, but their consistency, control of possession, and lack of turnovers had better results against Texas.
Or consider the Spanish National football team with their Tiki-Taka possession game. They don't always have scoring bursts as they did in this last Euro Final, but they yielded zero goals in the elimination rounds primarily as a result of their constant possession of the ball.
Through a quick survey of last year's conference game results we find that the spread passing offenses of the Big 12 are not necessarily having the impact that is guessed.
I was curious about the actual success of the spread passing game in this league after watching and breaking down Iowa State's upset over Oklahoma St. as well as Kansas St.'s early victory over RGIII and the Bears. In both instances running-oriented offense combined with superior possession of the ball resulted in victories over offensive juggernauts.
So I decided to examine every league game that took place last year and measure 6 features:
1). The winning percentage in games of teams with 400 yards of passing offense.
2). The winning percentage in games of teams with 200 yards of rushing offense.
3). The winning percentage of teams who out-rushed their opponent.
4). The winning percentage of teams who out-passed their opponent.
5). The winning percentage of teams who won the turnover margin.
6). The winning percentage of teams who had the greater time of possession.
The results were fascinating...
Teams with 400 yards of passing offense:
Record 10-6, .625
A team reached 400 yards of passing offense only 16 times in the 45 conference games last year. That's 17% of the time. Now, I doubt other conferences even reached 17% but it's interesting to note how rare this was.
While .625 is nothing to sneeze at, I found it remarkable that on 6 occasions last year a team managed to throw for 400 yards and still lose. Though it was very difficult to achieve, the marginal benefit was not as great as I would have guessed.
Teams with 200 yards of rushing offense:
Record 20-10, .667
Not only did teams manage to rush for 200 yards in a game 33% of the time (nearly twice as often as teams managed to throw for 400 yards) but when teams achieved that plateau they had a better winning percentage than the 400 passing yards team!
Had there been 30 games in which a team managed 400 passing yards those teams would have been expected to win 18.75 of those games. Pretty marginal advantage for the 200 rushing yard team but again, a much easier goal to manage.
Teams that out-passed their opponent:
Record 27-18, .600
Obviously throwing for more yards than your opponent is a very solid strategy for achieving victory.
Teams that out-rushed their opponent:
Record 30-15, .667
This isn't a perfect sample size to make these comparisons but it's all we have. And again we find that out-rushing the opponent offers the higher chance of winning the game than does the passing game.
While it would be a large conclusion from small data to conclude that a run-centric offense is better than a passing attack, you certainly couldn't look at this data and conclude that it's worse.
Teams that won the turnover battle:
Record 26-6, .812
On the 32 of 45 occasions in which 1 team turned over the ball more than the other, that team rarely won.
You couldn't quite say that it's impossible to win if you turn the ball over more than the other team, but the word difficult doesn't seem to capture the desperate nature of the predicament either.
I was curious to see if teams that ran the ball more had fewer turnovers and found that the teams with the 4 highest totals of rushing attempts comprised the 2nd, 3rd, and 4th best turnover margins. However, the team that ran the ball the least finished first in turnover margin. The outlier team that managed to finish in the top 4 in rushing attempts but last in turnover margin? Our own Longhorns of course.
Top 5 rushing attempts
1. KSU (424)
2. Missouri (415)
3. UT (413)
4. Baylor (407)
5. Kansas (386)
Top 5 in turnover margin
1. Ok. St. +17 (254 rushing attempts)
2. KSU +13
3. Baylor +3
4. Missouri +1
5. Iowa St. -3 (6th in rushing attempts)
In my opinion, these numbers have significant predictive qualities for the coming season. Let's go ahead and note that despite running the ball often (thereby limiting TO's) and effectively, Missouri didn't kill it in the TO margins and were unable to win as many league games as this approach would have suggested.
Texas managed to utilize an offense geared towards avoiding TO's and still turned the ball over enough to finish -8 in league play. It should be clear where to lay the blame for this given how well our defense played down the stretch...another microcosm of how horrendous our QB play was last season. If winning the turnover battle results in victory 80% of the time then by merely improving from "all-time worst" to "average" at quarterback Texas could expect to improve by several wins. That's not completing more passes or throwing more touchdowns...that's just from reducing interceptions.
Baylor and Ok. St., inversely, had tremendous QB play. For all their success last year, the abilities of RGIII and Weeden to carry large offensive burdens and avoid turnovers was predominant in their teams' success. You think new starters Nick Florence and Wes Lunt will match that success?
Kansas St.'s offense, which was mostly Collin Klein, was extremely good at avoiding turnovers. He's back, and with more help at the skill positions. For them to match that high TO margin may not be likely, but it's not unreasonable to believe that they'll continue to be successful there and more successful than Baylor or OSU.
Teams that had greater time of possession than their opponent:
Record: 24-21, .533
This is where the impact of the no-huddle/up tempo offense is noted.
For Iowa State or KSU, who sought to control the clock and run the ball, TOP was a good indicator of winning or losing. But Ok. St. routinely stomped teams and then allowed them to run plays and clock while sitting with a comfortable lead.
Because the spread passing game tends to result in points without making a big dent in the play clock, controlling time of possession doesn't have the same impact. Another key factor; the limited time didn't stop OSU from running more than enough plays to drop huge point totals on foes. They could usually run 60 plays in about 20 minutes of possession.
All the data reveals 2 primary points to me:
Quarterbacks who can run spread offenses without racking up turnovers are Kings in this league (or likely any other): Any team that doesn't have an experienced starter is fairly suspect because in a league with this much parity you'd better believe that TO's will make the difference in most of the games.
The following teams have returning starters:
Texas: We'll see how much Ash has grown.
Oklahoma: Landry has coughed it up too much in the past, if he can develop some new safety blankets after Broyles you like OU a lot with him back though.
Texas Tech: Seth Doege. If they improve on D from atrocious to average there could be big trouble from Lubbock.
Geno Smith: He's coming off a 7 INT season, which is very good.
KSU: Collin Klein, arguably the best QB in the conference.
TCU: Pachall also avoided TO's well last year but it's not clear if his defense will be able to force any with all their losses.
Helps clear up the middle of the pack somewhat, no? As for the top, the presumed favorites have some of the following flaws:
WVU: Just lost every defensive player who would have been useful this year.
OU: Turnover-prone QB, still struggling to build a running game to help him.
UT: Long way to go at QB.
KSU: One-man show coming off a somewhat lucky season. If not for their large TO margin last year, their terrible ypp and ypp allowed numbers would have sunk them.
The other takeaway is my suspicion that teams with a system for building consistent running games are more likely to sustain success than others so long as they manage to avoid wetting the bed at QB. Unless I was an Air Raid disciple capable of implementing that system, if I were a Big 12 coach I'd be looking to install some kind of option or power-O offense.
There are statistics behind the classical approach. More valuable than racking up offense? Protecting the ball.