College Football: Notre Dame Fighting Irish VS Purdue Boilermakers Analytics Preview

After two extremely tight wins Notre Dame is back in action at home this weekend against the Purdue Boilermakers. Both teams are unbeaten, and the 7 point spread indicates a closer game than many Irish fans are hoping for, while the first glimpses of Tyler Buchner last week only add to the intrigue. This edition of the analytics preview will provide you a new look at the quarterback controversy, what Purdue brings to the table, and much more! For a better understanding of how these stats work, please follow this link to our analytics primer.

Despite the up-and-down start the Irish still have a strong passing game that ranks a solid 28th in college football in EPA/play and 22nd in success rate. Reminiscent of how Javon McKinley and Ben Skowronek emerged last year, Kevin Austin and Braden Lenzy have stepped up into larger roles while Michael Mayer is one of the best Irish tight ends in recent memory. All is well with the pass-catchers.

However, the passing game appears to have a fatal flaw that could doom the Irish offense: Jack Coan is getting sacked too much. Many have placed blame on the offensive line, but we believe Coan himself is actually deserving of a fair amount of the blame. The Irish offensive line ranks 19th in pass blocking according to Pro Football Focus, and is giving up relatively few pressures so far. The problem is that 42% of pressures on Jack Coan end in sacks, the third highest conversion rate of pressures to sacks nationwide and far above the average of 18%. Here is an article to a study PFF conducted on who really controls the rate at which a quarterback gets pressured and sacked. They’ve found the year-to-year correlation is much stronger with the quarterback and that it is the most important variable in the models they’ve built trying to predict future pressure rates. https://www.pff.com/news/pro-z-quarterbacks-in-control-a-pff-data-study-of-who-controls-pressure-rates

Essentially while the line may not have five future NFL stars on it, Coan’s inability to run and tendency to hunt the deep ball when pressured is causing a much higher rate of sacks / pressures. When he’s in a clean pocket he has an aDOT of 9.4 but when pressured it jumps to 13.5. While an offense will need to accept the risk of a sack for the potential payoff of the deep ball, Coan is basically offsetting any gains made passing with his 11 sacks taken (2nd in the nation).

Coan has been one of the lower blitzed quarterbacks so far at 27.7% of drop backs, making designed screen passes a tenuous solution to the problem. This isn’t to say Kyren Williams and Chris Tyree shouldn’t be involved in the passing game, as they’ve been just as efficient as some other pass-catchers on the team. Notre Dame has also been hyper efficient on screen plays as they’re 8th in YPA on such plays. It’s more that throwing more designed screens when the opponents are barely blitzing would be trying to fit a square peg into a round hole, and that frankly the solution is Coan just has to play better and get the ball out quicker when under duress, and allowing Williams and Tyree to run more routes into the flat could aid in this area.

Another way to fix this would be to play Tyler Buchner who would have Ian Book level ability to move out of sacks and run when plays fall apart. We’re not advocating for this yet, but if the Irish can’t find a way to get Coan comfortable checking the ball down to avoid sacks and it begins to seriously hamper the offense, moving to Buchner may become an inevitability.

Note: There is a problem with Purdue’s 2020 season EPA statistics on ESPN’s website that resulted in some of Plummer’s plays being dropped from the dataset. Please take his 2020 statistics with a grain of salt.

The Boilermakers have an experienced quarterback who has posted impressive success rates over 50% in back to back seasons. Purdue also helps Plummer out massively by calling play action on 48.4% of his drop backs. Plummer is generating 13.7 yards per attempt and has completed 82.1% on play action, compared to just 5.6 yards per attempt and 68.8% completions on straight drop backs.

While Marcus Freeman’s softer fronts provided Florida State some running lanes they took advantage of, this strategy of lightening the box has been effective against play action. While this is a good thing, one would also expect this strategy to limit big plays and the opposite has actually happened. If Brian Kelly is correct that the root of the problem is just tackling, then hopefully this should be something that can be corrected more easily than a significant flaw in the scheme.

Purdue’s most dangerous passing game weapons are wide receiver David Bell and tight end Payne Durham. Bell is a chunk play receiver, gaining 15.31 yards per target despite a modest of aDOT of 9.6, showcasing his ability of making plays after the catch. Durham is the opposite, already having hauled in three touchdowns for a gaudy 2.05 EPA/pass. While the small sample size and touchdowns mean that Payne Durham is probably not the greatest collegiate tight end of all time as that efficiency number might imply, it does mean the Irish need to keep an eye on Durham near the goal line and be sure not to lose him in play action.

Moving to the run game, Purdue running back Zander Horvath is set to miss the next several weeks with a fibula injury, setting the stage for a committee headed up by King Doerue and Dylan Downing. The pair have been more efficient than Horvath so far by the metrics we use, but were perhaps less trusted by the coaching staff since Horvath has been heavily involved in the offense since he was a sophomore in 2019. It will be interesting to see how their workload is divvied up and how the pair perform in Horvath’s stead.

Jack: I have now been wrong two weeks in a row betting on Irish covers, but I am going right back to the well this week. The Irish had really bad turnover luck and a few other high variance things break against them last week. If the Irish limit the big plays on defense and get the ball out of Jack Coan’s hands this week, I think they win by double digits. If neither of those things happen, then losing this game straight up is certainly in play.

Cooper: Having taken a commanding lead over Jack on the season, I too will be picking a Notre Dame win and the Irish to cover. This game has been marketed as a trap game and it’s easy to see why. Purdue has a good offense while Notre Dame has struggled against a couple of inferior opponents. The reason I’m rolling with the Irish is because I think Tommy Rees will begin to lean even more into the pass and Marcus Freeman’s defense is set up well to defend against Purdue. A heavy play-action team with a good quarterback would have been trouble against Clark Lea, but given Freeman’s proclivity to give the offense light boxes and incentivize the run will limit the damage Plummer and Bell will do through the air.