clock menu more-arrow no yes

Filed under:

Notre Dame VS Virginia Tech Analytics Recap

New, 8 comments

Analytics to Kind of Understand What Just Happened

It wasn’t always pretty or easy, but Notre Dame heads into their bye week at 5-1 having gone a respectable 2-1 through the hardest three game stretch of the schedule. The Irish certainly have a fair number of questions to answer before they play USC on October 23rd, and we’ll try to answer as many of those here as we can. For more information on the statistics we use, please follow this link to our analytics primer.

Brian Kelly and Tommy Rees chose to go with Jack Coan under center to begin the game, and early in the game he showed all the potential downsides of his play. Coan had suspect pocket presence once again, culminating in the third down play where he stepped up directly into a sack in a clean pocket that led to his benching. At this point, few Irish fans probably expected to see Coan back under center perhaps all season long. And yet, when Coan was called upon late in the game he answered the call and then some.

Coan’s two game winning drives drove his EPA/Dropback back up to a solid 0.14, far surpassing Buchner’s while showcasing borderline NFL level accuracy in the game’s highest leverage moments. He also had his highest PFF Grade of the season in this game, posting a 78.3 overall grade with an 80.8 passing grade. Like many, we have been quick to criticize Coan’s lack of movement and proclivity for getting sacked, but wow did he deliver in the game’s biggest moments Saturday after getting benched.

The Tyler Buchner experience was a lot more varied than some are saying. We’ll start with the positive: allowing Buchner to run the entire playbook opened up space for him to run, even though it was very much still a run-first offense. Buchner’s 0.32 EPA/Designed Rush was a breath of fresh air for this offense, and likely opened up space for Kyren Williams to have one of his better games of the season so far. Kyren was still underwater at -0.03 EPA/rush, but after some recent showings from this run game it was a sizable improvement. Buchner also flashed his arm talent with the deep ball to Kevin Austin, and showed great poise playing in a tough environment.

All of this said, Buchner was wildly inaccurate. According to Pro Football Focus he had three turnover worthy plays on 14 pass attempts. The two interceptions (the pick six especially) tanked his passing EPA to -0.38 EPA/dropback, and he posted just a 36.2 PFF passing grade on the day due to his inaccuracy, well behind Coan’s 80.8 mark. Buchner hasn’t had full starter’s practice reps yet and was missing Michael Mayer so some of this is understandable. He clearly is still working to improve as a passer, although the rushing upside does provide some room to breathe as he develops.

Whether to start Coan or Buchner is a difficult decision moving forward as both provide different strengths, but it’s likely better for the Irish program in the long run to allow Buchner to work through his growing pains on the field than have him sit behind Coan for the remainder of the year. On the other hand, this is hardly a throw-away rebuilding season as the Irish are at 5-1 and ranked in the AP Top 15 currently, with this college season possibly being the most wide open in recent years. My biggest insight into the quarterback battle is I do not envy being Brian Kelly and making this decision.

A note on Irish play calling: the run-pass choices were really predictable based on how Notre Dame aligned. They ran just 25.8% of the time when they lined up without a tight end and ran 87.5% of the time when they lined up with one. The Irish did their best rushing on the day into lighter boxes (admittedly small sample size here but this is something that holds true generally), so we’d like to see Tommy Rees consider spreading to run and letting Kyren Williams carry the ball into lighter boxes out of the 10 alignment (one running back, no tight end) in the future.

In addition to that, we really need to see more Play Action passing, especially when Buchner is in the game. The Irish used Buchner as a pseudo running back for weeks. At least capitalize on this by running play action to give him some more space to operate in the secondary. NCAA average play action rates are around 25-30% of dropbacks. The Irish used it just 13.7% of the time in this game and only twice on Buchner’s 14 dropbacks. When the Freshman is in the game and the offense is going to be very run-heavy, it makes sense to complement that with heavy Play Action usage to keep the defense guessing. This includes RPOs and rollouts where Buchner has the option to tuck the ball and run. It’s a very simple wrinkle that can be added to the offense and lead to fairly large gains in efficiency.

Virginia Tech was coming off of a bye week and this was an offensive gameplan you could tell they used the entire two weeks to formulate. They threw the kitchen sink at the Irish, featuring heavy motion, heavy play action, and a glorified wildcat package with Connor Blumrick among other wrinkles. The Irish play six opponents coming off of bye weeks this season, and between this and the current reputation of Notre Dame’s program, the Irish get everybody’s best shot.

It wasn’t always pretty, but the defense took away the things Virginia Tech wanted to do. Motion at the snap was ineffective for the Hokies, and their play action dropbacks were less efficient than their straight dropbacks, which has become a theme for Marcus Freeman’s unit. The Irish once again played light boxes, and Virginia Tech’s run game still could not get going. At the same time, the added help in the secondary completely took away deep passing with Virginia Tech only connecting on one deep pass a week after getting beat multiple times downfield. Star Hokies wide receiver Tre Turner saw a ton of volume, but was held to a manageable 0.34 EPA/play, not bad when considering his 11.6 aDOT should have seen more production. Keep in mind as well the 29 points factors in the Pick 6, which the defense had no part in allowing.

One last thing we want to touch on is Virginia Tech’s decision to punt on 4th and 1 with 2 minutes left in the game. This was a clear “Go for It” situation from an analytics standpoint, with the expected added gain in Win Probability far outweighing that of punting (over 14%). The reason punting is so detrimental here is that you set yourself up for a potential scenario where Notre Dame scores and you have little to no time left in the game to try and tie it, which is exactly what happened. The Hokies also had a 74% chance of converting which is almost as sure of a guarantee you’ll find in a game, and should they have failed they still had a few different paths to get the ball back. First, they could have stopped Notre Dame on the first set of downs, and they had two timeouts remaining so at most the Irish would have taken off about 50 seconds on the clock. They would then get the ball either down 3 or tied, depending on whether Notre Dame makes the field goal or not. Second, if they try to stop Notre Dame on downs but allow a 1st Down conversion, they could then let the Irish score a touchdown so they get the ball back and have an opportunity to respond. This is the exact situation that played out in the Toledo game that worked out in Notre Dame’s favor and also one the Browns utilized this past week against the Chargers. Or they could have immediately let Notre Dame score a touchdown, keeping their two timeouts for a potential game-tying or game-winning drive. There’s a strong possibility the Irish could be 4-2 if Justin Fuente chose to go for it, but he made a critical mistake that allowed Notre Dame to score the go-ahead field goal with little time left on the clock.

Overall, the Irish won a football game on the road against a solid opponent without its best offensive weapon in Michael Mayer and an extremely turbulent quarterback situation. While one can blame the coaching staff for allowing the quarterback situation to reach this point, this was a gutsy win the Irish coaching staff and players deserve credit for. Jack Coan, Tyler Buchner, and Brian Kelly all seem singularly focused on the team and winning, and that’s a culture that can only be achieved over time through hard work and attention to detail. Things can always be better, which is a lot of what we focus on in our articles, but they can always be worse too and these coaches and players deserve a world of credit.