After dismantling New Mexico in non-dramatic fashion, the Notre Dame Fighting Irish descended upon Athens, GA for one of the most anticipated non-conference match-ups of the year. The Irish fought hard and kept well closer than the gambling spread of fourteen points, but came up short in an early blow to their College Football Playoff aspirations. If this is your first time taking an advanced analytics look at the game, please follow this link back to the first installment of this series for a primer on the terminology that will be used. Let’s begin with the game’s win probability chart.
The Irish got a big gift from the Bulldogs early, recovering a muffed punt deep in Bulldog territory to take a 7-0 lead and eventually going into halftime up 10-7. While the first half was a back and forth affair, a nightmarish third quarter saw the Irish squander their lead and Georgia take control of the game. The Irish had a spike in win probability after the Bulldogs’ terrible punt late in the fourth quarter set the Irish up with great field position for their final drive, but the Irish could not take advantage and complete the comeback.
The Irish run game went from mediocre to awful in this game, with the coaching staff choosing to ignore it almost entirely. The team drew up twelve designed runs, with just nine going to running back Tony Jones. While Jones contributed little on the ground, this was actually one of his better games. He looked dangerous coming out of the backfield as a receiver and did an excellent job in pass protection, something that was critical facing that intimidating Bulldog front seven.
This pass protection was important because the Irish were simply overmatched up front. If the ineffectiveness of the inside run game wasn’t enough to see that, look at how infrequently the Irish threw down the field. It’s pretty clear the coaching staff wanted the ball out of Book’s hands almost immediately on every pass play, and the majority of passes did not travel over nine yards as a result. The Irish used motion and quarterback rollouts to get the ball to the edge away from the teeth of the Bulldog defense. It was a spectacular game plan, and one that turned a day that could have been really ugly on offense into one that was mostly fine.
And while the game plan did not lead the Irish to be particularly effective when passing, they did do far better through the air than on the ground. Cole Kmet and Chase Claypool were the two most dangerous receivers for the Irish all season and this was no exception for the pair of second round NFL Draft picks. Every pass caught by Claypool was successful, and he even recovered the muffed punt that set the Irish up with excellent field position to score their first touchdown. Claypool and Kmet were sensational in this one, and this game showed why they will be playing on Sundays this fall.
With the Irish offense playing below its season averages it was up to the defense to keep the Irish in the game, and they delivered. Throughout the offseason Irish fans feared what D’Andre Swift and the Bulldog run game was going to do to the inexperienced Irish linebackers. Clark Lea’s group performed well, holding Swift to under 100 yards on 18 carries, one of just three games last season where he was given over 15 carries and held under the century mark. However, his box score numbers are slightly deceptive, as Swift’s .23 EPA was very good considering he averaged .09 on the season. His Success Rate (44.4%) and First Down % (33.3%) were both in line with his season averages of 43.6% and 30.3%. The Irish run defense deserves credit for limiting the big play, holding Swift to a long of 15 yards, but he did plenty of damage consistently picking up gains of over 6 yards and was more successful than traditional yardage metrics would suggest.
These graphs are a great example of what a good pass rush like Georgia’s can do to a team. The play action game was rendered completely ineffective because by the time Book could get his eyes downfield from faking a handoff the pressure was already starting to penetrate the Irish line. This game featured the lowest percentage of play action in any game last season, so it appears the coaching staff figured this out as well. Given the prowess of Georgia’s front seven, Book was not under pressure too often in this game (25.5% of his drop backs) due to the emphasis on the quick passing game with shallow depth of targets. Georgia’s rush was very disciplined as well as dangerous, staying in their gaps and limiting Book to just two scrambles. This forced him to throw his way out of pressure situations, and he was only able to convert first downs on 8.3% of these drop backs and averaged 3.2 yards per play. These were well below his season averages and terrible compared to the efficiency Jake Fromm displayed in all situations. The Bulldog front seven won them this game with their pressure and run stopping, and these charts are just another reason why.
The Irish were most effective splitting Kmet out wide in this one, as he created all kinds of problems for Georgia down the seam and on intermediate routes. As usual the two tight end package was the least effective for the Irish, one of the reasons I almost spit my coffee out when I read Brian Kelly is considering using 13 personnel in non goal line situations this season. Similar to the two tight end sets not working, the Irish were terrible under center. Anything that pitted the Irish offensive line more directly against the Bulldog defensive line simply did not work, including these power formations on offense.