Developed by Football Outsiders’ Bill Connelly, Havoc Rate adds up tackles for loss (including sacks and other tackles behind the line of scrimmage), forced fumbles, and passes defensed (including interceptions and pass breakups) and divides by the number of plays in the game. For example, if a team faced 100 plays, got 5 tackles for loss, 1 forced fumble, and 4 passes defensed, they would get a 10% Havoc Rate. It isn’t a perfect stat, as it is probably a little skewed towards passes defensed, which are far more likely than tackles for loss (and probably should include tackles for no gain, but that would be much more difficult to chart), but it still gives a pretty good sense of how well the defense creates chaos for the opposition.
How have the Bruins been doing? Given that the average 2013 defense had a Havoc Rate of 15.9%, the Bruins have, as the message board suspects, spent much of the season well below that level. For the year, UCLA has a mean Havoc Rate of 12.4%. The Bruins have only passed the average 2013 team in the Cal and Arizona games. Let’s go a little more in-depth:
Stats are pretty darn messy but we can see a few trends. The Havoc Rate has indeed crept up over the past four games, with a mean of 15.4% vs 10.2% over the first six games of the year. The highest havoc rates came in the Cal and Arizona games, and those were probably the two best defensive games of the season, as the Bruins held the Bears well beneath their season averages and dominated the Wildcats. The lowest havoc rates came in the mediocre defensive performances against Texas and Washington and the awful performance against Oregon.
Let’s dial in a little more: here is the Havoc Rate including its contributions from the defensive line (note: we considered Deon Hollins a DL), linebacker, and defensive back component rates:
The first thing that jumps out at us is that even though the Cal and Arizona games had similar percentages, they were made up of very different components. The Cal game was a relatively equal showing, with the defensive line leading the way, while in the Arizona game the defensive backs drove most of the havoc.
Separating out the components:
The defensive line has had two games in which they created a lot of havoc, Utah (against the worst offense in the league) and Cal. Their lowest havoc performances were the second half collapses against Colorado and Virginia and the all-around collapse against Oregon. It does seem that the average havoc rate has gone up since the introduction of Takkarist McKinley in the Cal game, as the defensive line managed a mean Havoc Rate of 3.9% before McKinley and 4.9% after.
The linebackers have created the least amount of havoc of any of the three position groups, though that might be different if we had counted Deon Hollins as a linebacker. The linebackers created a lot of havoc against Colorado, with a fair amount against Cal, Arizona State, and Virginia. They created no havoc at all against Utah (a problem against a team that basically ran the ball all game), and very little against Washington.
The defensive backs have created the most havoc on average of any group. Even though the Arizona game seems like a bit of an outlier, the DBs still have the highest median and mean havoc rate. That Arizona game really was a disruptive masterpiece, but the DBs also created a fair amount of havoc against Cal, ASU, Memphis, and Virginia.
So there we are—the Bruins create less havoc than the average team, with the defensive backs creating the most followed by the defensive line and the linebackers. Takkarist McKinley seems to have had a measurable effect on the defensive line’s ability to create havoc, and the defense as a whole has been creating more havoc after the Oregon game.
Questions? Comments? Love havoc and want to hear about exciting college football chaos scenarios? Meet us on the Premium Football Forum or tweet us @Bruinalytics.