Trimming the Fat
There’s a lot of fat in college football statistics. Schedules vary so widely across the country, and even within conferences as those continue to grow, that numbers like total offense, scoring defense, etc. frequently have less value than we assign to them.
Here’s an easy example. Nebraska’s offense ranks first in the Big Ten, Wisconsin’s ranks fourth. Easy enough to understand, both offenses are capable of moving the ball. Limit it to conference play only and the Huskers fall to third while the Badgers stay at fourth. What that raw rankings don’t consider is that, in conference play, Nebraska played against the eight best defenses it could have faced. Wisconsin got to face Illinois (10th in total defense), Purdue (11th), and Indiana (12th). That’s a basic example, but does it change how you view each offense?
It should slightly and there are all sorts of little statistical quirks when you look at this game. Luckily, there are some steak knives out there that help trim the fat by controlling for things like garbage stats, strength of schedule, and field position.[ad 6]
Let’s look at some of those numbers from a couple of the leading advanced stats sites out there. It’s not too rigorous. I’ll be as clear and concise as possible and I promise you’ll come away with a better understanding of this Nebraska-Wisconsin match-up.
First, for reference, here is where the Huskers and Badgers rank nationally and in the Big Ten in the “traditional” statistical categories:
Drive-based stats are the basis for most of the advanced analysis of football, college or NFL. The big difference here from traditional statistics is that FBSDriveStats.com removes any stats that come during blowouts and also removes any games against FCS opponents as those tend to cloud the actual numbers. What you get are numbers that reflect how well (or poorly) a team is doing when the game is still in doubt.
The biggest thing you see when comparing Nebraska and Wisconsin this way is the role field position has played this season. To do that, you can use net field position which takes average offensive starting field position minus average defensive starting field position. Offensively, Wisconsin starts drives nearly 6-yards closer to the end zone than opposing offenses do. They may not seem like a huge difference, but it ranks first in the Big Ten and 10th nationally.
Nebraska has had to make up yards on offense. The Huskers have started on offense 2.3 yards further away than it has forced the opposition to start. Again, a seemingly small difference, but that ranks Nebraska 91st nationally and 10th in the Big Ten.
Field position played a huge role in the first game. Wisconsin started two touchdown drives inside Nebraska’s 25-yard line thanks to a pair of Huskers fumbles. It’ll be vitally important again on Saturday so watch not just the turnovers, but special teams. If field position is close to average on Saturday – about the 31-yard line — the Huskers have shown so far that they’re better equipped to gain the necessary yards than Wisconsin is.
The Power Rank uses an algorithm that controls for strength of schedule and uses yards per play as the primary basis for comparison. Similar to Value Over Replacement Player (VORP) in baseball, this method gives you an expected margin of victory over an average team, which can then be used to come up with an expected margin of victory over any opponent.
In this game, the Power Rank projects a 1.5-point victory for Nebraska with a 53.4-percent win probability. In 11 Nebraska games this season, they’ve picked the straight up winner each time. For Wisconsin, the Power Rank was 9-2 straight up.
On the offensive and defensive fronts, the adjusted stats reveal a few things. One, Wisconsin’s defense is better than Nebraska’s offense. You could intuit that from the traditional total offense and total defense numbers, but the difference is smaller than those numbers show.
The Badgers rank 14th nationally in yards per play allowed but, when adjusted for who they played, they rank seventh allowing 4.49 yards per play. Nebraska’s offense takes a significant jump, from 29th nationally (6.26 ypp) to 14th (6.37) based on strength of schedule. Wisconsin still holds the overall edge, but it’s not as big as the straight per game averages indicate.
Nebraska has the advantage on the other side. Defensively, the Huskers are good (16th nationally) while the Badgers are almost dead average (63rd) using the Power Rank numbers. What you’re looking for here are gaps. While Wisconsin’s defense is good it’s a marginal difference of seven spots against Nebraska’s similarly good offense. Flip it around and Nebraska has a pretty significant advantage, 47 spots, on defense over the Badgers’ offense.
That’s diverges a bit from the traditional narrative of Nebraska’s offense versus Wisconsin’s defense headed into the game. The bigger difference is on the other side of the ball.
In theory, the ultimate statistical measure is a head-to-head match-up and we have that history here but one game is a small sample size. The question is how true was the first result?
In terms of score, it wasn’t very true. Nebraska was more than three points better than Wisconsin in September. The two short fields mentioned above resulted in 14 Wisconsin points. On the stat sheet, Nebraska dominated in almost every other category – 24 first downs to 17, 259 rushing yards to 56, 440 total yards to 295. On the score sheet, the difference was a field goal. (Incidentally, both teams missed one in that game.)
But the number that really jumps out is three. That’s the number of punts Wisconsin forced the first time around and Nebraska’s fewest for any Big Ten game. The Badgers forced two three-and-outs in the first half, but the only punt of the second half came after an eight-play drive where Nebraska bled 4:38 off the clock in the fourth quarter while maintaining a three-point lead. Neither team likely came away dissatisfied with that drive.
Overall, Wisconsin really struggled to stop Nebraska the first time out, which puts me in a unique position. In general, I’m cautious when it comes to Nebraska’s chances, perhaps overly so, but in terms of pure team strength I think the Huskers are about two scores better than Wisconsin.
Filtering that first game through some of the advanced statistics above shows that it will likely be a closer game than that. Nebraska might have marginally less success on offense and Wisconsin might have a little more. Also, football isn’t played in a vacuum. Sudden change plays – turnovers, blocked punts, special teams returns – can and do upset the statistical apple cart.
There’s no way to truly account for that. Three days away from the game, all we have is the numbers. Once you trim away some of the fat from those, I think there’s good reason for Nebraska to be very confident – and I get the sense that they are – headed into Indianapolis.