Stats are for Nerds: Rice Week

We live in a fantastic era for tracking sports, especially for those of us who live far from our favorite teams.  The internet puts video and commentary about even obscure teams at our fingertips.  So when I wanted to find out what Rice looked like, I tried some google searches.  "Rice football highlights" turned up a video of Robert Griffin III and Baylor lighting them up.  "Rice highlights": would you like Ray Rice, Jerry Rice, Glen Rice, Sydney Rice, or Tony Rice?  "Rice football blog": LOL NO (though LTP showed up on the second page on the strength of having "football blog" prominently featured.  Nice search optimization there).  Eventually, I was able to figure out two things: first, you can view their games in their entirety on the same crappy CBS video player I had to deal with for the Army game.  Second, CBS actually does have highlight videos available for free, but they do their absolute best to ensure that you can't find them unless you really work at it.

So, after fighting through layers of poor web design, I discovered that Rice runs a balanced-looking spread offense.  The defense was featured more in their opponents' highlights than in their own; it seems to be from the 4-2-5 family popularized by TCU.  Since I've had about enough of real football, let's get to the stats. 

Last week, Northwestern survived playing a sloppy game on the road, picking up the win in spite of turning the ball over three times thanks to great line play, a huge Ebert touchdown, and good game management by the coaches.  Bill Connelly, the man behind the S&P+ rankings, has a statistical breakdown posted at the SBNation Football Study Hall blog.  The story his numbers tell is unsurprising: Nebraska won the field position battle (Northwestern started every first half drive from at or behind their own 20), but Northwestern consistently stayed on schedule (73% leverage; 50% success rate) and was both consistent and explosive through the air.  Nebraska's run game was bottled up, and though they passed effectively they didn't get many big plays.  Before factoring in turnovers, Connelly's numbers predict a 3 point NU win; after, they point to a 1 point UNL win.  Fortunately, football isn't played on the stat sheet, and NU picked up the win.

 Onward to see how this effected the thing we all know REALLY matters: advanced stats and automated rankings!

This week, I will be looking at the same three measures as last week: S&P+ (play based), FEI (drive based), and SRS (point based; similar to Sagarin's predictor ratings).

Getting a win over a good team helps in the SRS for obvious reasons, boosting NU to 60th with a rating of 40.2.  Rice, coming off a win over a less good team, sits at 96th with a 30.3; three points for homefield predicts a 12.9 point NU win.  Currently, the SRS divides the Big Ten into a few tiers: Indiana and Minnesota are clearly at the bottom, Purdue and Northwestern are solidly ahead of those two but behind the rest of the conference, 3 (Nebraska) to 8 (Illinois) are all close together, Michigan solidly clears those teams, and Wisconsin stands alone at the top.

On to the more detailed systems.  FEI was seriously impressed with the win, as Northwestern jumps 9 spots from 54 to 45.  I'm not sure whether to be more impressed by the offense (16th in the country) or the defense (a shockingly high 80th).  Interestingly for a unit notorious for giving up big plays, the defense has given up more drives of 10 or more plays than of 10 or more yards per play.  This suggests a modest victory of the bend-but-don't-break strategy that I thought had been replaced by a BREAK! BREAK! BREAK! tendency.  Still, the bottom line problem remains that the defense is not particularly good more than its strategy.

S&P+ was less impressed; NU moves up two spots from 62 to 60, and the defense is a more believable 98th.  This was driven by Taylor Martinez's excellent day passing, as S&P+ now rates only New Mexico as a less efficient pass defense than Northwestern.  S&P+ also has the NU offense at 16, driven by exceptional balance: 20th passing, 21st rushing.  The only worrying factor is that NU is heavily dependent on staying on schedule offensively; the offense is 9th in the country on standard downs but drops to 61st on passing downs.

Rice fares just as badly by these measures as by the SRS.  FEI has them as a balanced team: 94th overall with an 88th ranked offense and 92nd ranked defense.  S&P+ puts them at 99th with the 94th offense and 95th defense.  They aren't particularly good defending the pass (65th), but rushing is their real weakness (96th).  Unsurprisingly, they are awful at knocking teams off schedule (90th on normal downs) but somewhat better when they do (67th on passing downs).  Their offense is similar: awful running and on standard downs, merely bad passing and on passing downs.

Overall, I can't find anything particularly interesting about this matchup in the stats.  Rice is very bad at everything, but Northwestern is bad enough at defense to give them hope.  Still, if both teams play the way their stats suggest, Northwestern will put up a lot of points, Rice will put up some but not nearly as many, and Tyris Jones might get another chance to carry the ball.  Considering how thin NU's margin for bowl eligibility is, I would like nothing more than a nice, boring start to my Saturday.