The Good, the Bad, and the Rock Fights: Cal at North Texas Football

I think we all know which category this one will fall into...

Photo credit goes to photographer extraordinare Rob Hwang

Welcome to The Good, the Bad, and the Rockfights, our weekly endeavor to put the most recent Cal game into one of four distinct categories. If you want an extended introduction to the series, check out our inaugural edition from last season. Otherwise, here’s the short intro: for each game the folks at Pro Football Focus (PFF)* grade the Bears across thirteen categories:

• OVER: Overall performance

• OFF: Offense

• PASS: QB passing

• PBLK: Pass protection

• RECV: Routes run by receivers

• RUN: RB performance

• RBLK: Blocking for the rushing attack

• DEF: Overall defense

• RDEF: Run defense

• TACK: Tackling

• PRSH: Pass rush

• COV: Defensive coverage of receivers

• SPEC: Special teams

Using a machine learning algorithm to find clusters of similar games across the Wilcox regime, we’ve found four types of games:

• The Good: everything goes well on offense and defense (e.g. the 44-11 Big Game win in 2021)

• The Bad: everything goes terribly (e.g. the 7-38 blowout loss in Seattle in 2017)

• The Rockfights: stellar defense, awful pass offense and decent run offense (e.g. the 27-16 win over Ole Miss in 2017)

• The Pillowfights: strong pass offense, decent run offense, mediocre defense (e.g. the 24-14 win over USC in 2021)

Each week we take the most recent PFF grades and figure out which bucket the game falls into. First, let’s review the grades.

PFF Comparison

Below I have highlighted the grades from the North Texas game (gold dots) compared to the rest of the grades from the Wilcox era. I use boxplots to display the data. In case you’ve forgotten everything from Stats 2, here’s how boxplots work: there is a box that captures the middle 50% of the data, or those data points between the 25th and 75th percentiles. The horizontal line represents the median, or the exact middle of the data. The vertical lines (sometimes called whiskers) capture data near the edges of the distribution. And singular blue dots represent extreme outliers.

PFF grades from the North Texas game (gold) compared to the rest of the grades from the Wilcox Era. The box represents the middle 50% of data and the blue horizontal line represents the median grade.

Saturday’s game fared quite well in every offensive category (except run blocking, oddly enough)—good job Coach Spav! Defense was mostly positive: the overall category was considerably higher than normal, tackling earned the highest scores of the Wilcox era, coverage fared typically well, and pass rush, usually the lowest defensive rating, earned solid ratings. Run defense was the only category that finished below normal, which is odd considering North Texas averaged 1.5 yards per carry and didn’t have a single double-digit run. Overall, it was an excellent day on the gridiron.

Cluster Analysis

Now for the fun part: I fed the grades into the clustering algorithm to figure out which type of game it was. Unsurprisingly, this one fell into The Good. The plot below visualizes all games since 2017 and the clusters they fall into; the North Texas game is highlighted with black rectangle.

Four types of games under Wilcox. Games closer together are more similar while games farther apart are more dissimilar. Each cluster is defined by its color-coded boundary.

This one falls comfortably into the blue cluster representing The Good. In that plot games closer together receive more similar ratings while those farther apart received more dissimilar ratings. Saturday’s matchup against North Texas has nearby neighbors Idaho State (2018) and Washington State (2019), both relatively high-scoring affairs in which the Bears held the opponent to about 20 points. Can the Bears achieve another entry in The Good next week against Auburn? Check back on Tuesday to find out!

*Programming note: Like all our pieces that utilize PFF data, future editions of this series will be behind the paywall for the remainder of the season. Become a paid subscriber to gain access and to help us pay for our annual PFF subscription, which enables content like this and other in-depth analyses utilizing their extensive data. Thanks for reading and Go Bears!