Initially, I intended yesterday's Sadistical Analysis piece to be a stand-alone post focusing on this season's disastrous offensive meltdown.
Alas, the esteemed editors of ¡El Carnaval del Descortezamiento Fantastico Gigante! demand fairness and balance. And so I took a quick dive into the complementary defensive statistics to present a similar analysis of the defense's performance so far.
First, a word of warning. I initially balked at blogging about the defensive statistics because I believe them to give a skewed impression that is overly favorable to the defense. For numerous reasons, the defense is giving up a good chunk of points while yielding exceptionally low yardage per-play. One of these reasons is that the offense and special teams are shockingly bad, and have consistently gifted the opposing offense exceptional field position. Another thing to remember is that while the defense grades out worse on points-based measures than it does on yardage-based measures, it's better in terms of PPG than most of us realize.
But there is still a mismatch between yards and points allowed. I suspect that there is another statistical quirk at play - namely, I think a careful analysis of the drive charts (something I don't have time to do at the moment) will reveal that this defense is unusually susceptible to rare big plays and/or long drives. At this point, my theory is purely anecdotal. And that's why I hesitate at this point to judge the defense, for good or bad. The stats simply don't match what we have perceived from actually watching the team play.
YARDS PER PLAY
With that out of the way, let's talk about the statistics. I evaluated the offense in terms of yards-per-play, and it's only fair to apply the same metric to the D. Here are some key points:
The defense has outperformed the average YPP against 8 of 10 offenses we've faced.
In the two subpar games, the defense just barely missed the mark, yielding 7.25% more than KSU's average and a miniscule 0.02% more than OSU's average.
On aggregate, the D has done 22.3% better than the opponents' average opponent.
In 7 games, the D has held the opposing offense to more than 10% less than their average YPP.
In 5 of the 10 games, the D has held the opposing offense to more than 20% below their average YPP.
The defense ranks #11 nationally in YPP allowed.
As you can see, the defense has consistently performed as well or better (usually much) than its opponents' average opponent. At least in terms of YPP. It's also important to note that, on average, Texas's opponents rank #63 in YPP gained. That's below average nationally, but only by 3 spots (in contrast, the offense faced defenses ranked, on average, in the 70's).
SCORING DEFENSE
In addition to YPP, I also looked at scoring defense (in terms of PPG). Honestly, I expected this to be much worse than it was. The defense ranks #54 in (unadjusted) PPG. That's only slightly above average, but remember that that particular statistic is not adjusted to factor out quality-of-opponent. For three reasons, I suspect that Texas is significantly stronger than the unadjusted PPG rank indicates.
First, Texas has played a schedule of high-scoring offenses. The average PPG rank of Texas' 10 opponents is an above-average #53.5.
Second, Texas has done well in terms of differential PPG allowed. In 10 games, Texas has held the opponent to less (typically much less) than their PPG average in seven. We had a huge hiccup against UCLA (77% above its PPG) and fairly sizeable shit-the-bed moments against ISU and KSU (+20% and +26%, respectively). In the other games, Texas held its opponents between 10% and 60% below their PPG average. On aggregate, Texas yields 22% fewer points than its opponents average.
Third, the last time Huckleberry ran the numbers (Nov. 1, immediately following the Baylor game), there was a big discrepancy between Texas' unadjusted PPG allowed rank (32) and its adjusted rank (17). In the meantime, we've played by far our highest-ranking opponent on PPG (OSU at #3 nationally) and another opponent that ranks well above our average opponent (KSU at #35). In other words, our strength-of-schedule has improved since Huck's last number-crunchathon and, thus, the difference between our unadjusted and adjusted PPG rank should be even more substantial now.
In sum, Texas's defense is very good on a YPP basis and average-to-good in terms of PPG. Since the latter measure is negatively affected by the complete ineptitude of our offense and special teams, it's legitimate to assign it less weight when judging the defense.
THE OFFENSE AND DEFENSE, COMPARED
Is this really necessary? Well, I'm a stats geek and the post would feel incomplete to me if I didn't wrap it up with a quick comparison of the O and D. But, I won't blame those who want to skip over this dead horse beat-down. Here's a quick executive summary for you: the offense sucks louder than possibly any offense in the history of offenses and its cancerous suckitude is killing our team and making a good-but-not-great defense look much worse than it actually is.
There. Now on to the pedantry.
The defense consistently does its job much better than does the offense. In only one game has the defense performed worse than the offense on the "differential YPP" measure. That game was OSU. The offense gained 1.46% above OSU's average YPP allowed, and the D allowed .02% more than OSU gained on average. Essentially, both units performed at the statistical mean. In every other game, the D significantly outplayed the O.
Want to see that in pictures? The following bar chart illustrates, on a game-by-game basis, the differential amount (in percentage) by which each unit outplayed its opponent's average opponent in terms of YPP. I've adjusted the numbers so that a positive number is good for both offense and defense.
That chart encapsulates this season. The defense has been consistently good, with a couple of off games. The offense has been...
You know what? I've run out of adjectives.