Georgia Tech Football: GT Defense vs. Pitt Advanced Stats Review

Final Score: Georgia Tech 26 Pittsburgh 21

Model Prediction: Pitt by 21, GT to cover: Correct

Projected EPA (Offense and Defense) Margin of Victory: GT by 3

GT Win Probability (Based on Success Rate, Yards Per Play, and EPA): 35%

It’s Wednesday, and Georgia Tech fans continue to enjoy the feelings from an all-too-rare victory, especially a win over a ranked conference foe on the road. Heading into Pittsburgh with an interim coach and as a 22 point underdog, Tech stayed close during a first half where the offense really struggled and rode a string of second half turnovers from the defense and explosive plays from the offense to a five point win. Let’s take a deeper dive into the defensive side of the ball, the EPA highlights from the game, and progress on our season long goals.

Advanced Stats Comparison

Heading into the game, we identified advantages for Pittsburgh in 21 of the 23 metrics we compare in our previews. Post-game, GT exceeded that expectation, at least slightly, holding the numeric advantage in five of the categories from the game itself. That may not sound like much, but it’s a good indication that GT over-performed expectations. For a team that hadn’t done a lot of that recently, it’s a good sign. Let’s dive a little bit more deeply into what went mostly right on the defensive side of the ball.

When Pitt Had the Ball

Through five weeks, we’re seeing encouraging evidence that this defense is meaningfully improved over last year. The pass rush prevented what Kedon Slovis wanted to do on Saturday night, the coverage was mostly excellent, and the run defense worked by limiting any semblance of explosion from Pittsburgh and creating a couple of disruptive plays that were very impactful.

The overall offensive line for Pitt from this game is illuminating. The Panthers put up a relatively high success rate (64th percentile), meaning they were able to keep from getting too far behind the chains. However, Tech held Pitt to a below average yards per play output and a very poor EPA/play number. That’s where we see the impact of GT limiting explosive plays and forcing three turnovers. It’s worth spending a little bit of time putting those factors in context.

Yesterday, Akshay covered some of the ways in which the performance of the offense on explosive ways contributed to the win but isn’t necessarily predictive for future games. On defense, we see the mirror of this same occurrence. The defense held Pitt’s offense to a meager explosive play rate (7th percentile). Preventing those explosives meant Pitt had to work hard to convert lots of third down opportunities, and that largely didn’t happen. This was a very successful recipe for winning this particular football game, but it also isn’t a formula that should be relied on. Bill Connelly explains how efficiency (here, think of that in terms of success rate) is more predictive of future performance than explosiveness.

The Georgia Tech defense showed us a lot that we wanted to see in this game: pass rush, run stuffs, coverage, broken up passes, turnovers forced. That’s worth celebrating. It’s also important that we not ignore the efficiency the Panthers found following the first quarter.

Rushing Defense

For the Georgia Tech defense against the Pittsburgh run game, things weren’t boom or bust. They were more like boom or “solid 6 yard gain but nothing more.” That’s good! The run stuff rate was above average, and the number of explosive runs was minimal. The performance still does not look very good through the lens of line yards or opportunity rate, and that’s reflective of Pitt getting a lot of runs in that five to seven yard range. But the defensive line largely held their gaps, preventing offensive lineman from getting to the second level and opening the chunk runs that we saw over the last several weeks. At the individual level, Ace Eley, Jaylon King, Zeek Biggers, D’Quan Douse, Myles Sims, and Zamari Walton all had PFF rush defense grades over 70 (remember that 60 is considered adequately doing your job). Only 3 out of 21 graded players were below 60 against the run, and that’s a massive improvement from weeks prior.

Defensive Disruption

At a team level, the pass rush posted an exceptional pressure rate of 39%. I’ve seen some comparison to the defensive performance against North Carolina last year, but this was one was even better. Pitt’s offensive line is better than UNC’s was last year, and the GT defense still put up only a 37% pressure rate in that game from 2021. Looking at some individuals, Charlie Thomas had a pass rush grade of 87, and K.J. Wallace, Jason Moore, D’Quan Douse, and Keion White were all graded quite well in that category. Thomas put up four pressures in just one half, which is incredible. Further, we saw the defense post above average havoc and run stuff rates. For this team, after last year, that’s awesome.

Pass Coverage

There’s more to celebrate here, as the coverage on the backend was a big reason why Pittsburgh struggled so much to produce explosive plays. When Pitt did catch the ball, yards after catch were very limited. The only clear coverage bust came on the 26 yard touchdown that Pitt scored pretty late in the game. LaMiles Brooks is one player to highlight; he has been exceptional since getting the starting job a few weeks ago and added a couple more pass break ups on Saturday night.

Two things to note here going forward: it looks like Kaleb Edwards will be redshirting in hopes of finding a more expanded role next year, and Jaylon King is out for a while, perhaps the rest of the season, after undergoing ankle surgery following the win. We hope the best for both of those guys and look forward to seeing them back on the field next year.

EPA Highlights

EPA calculates the expected number of points added (or lost in the case of a negative number) on a particular play based on the down and the location on they field.

As always, we’ll take a look at the most helpful and hurtful plays for GT.

Most Helpful Plays

1. 6.63 EPA – Jaylon King forces fumble, recovered by Charlie Thomas and returned to the Pitt 17 in the middle of the third quarter
2. 4.85 EPA – Ace Eley forces fumble, recovered by Clayton Powell-Lee at Pitt 34 early in the fourth quarter
3. 3.72 EPA – Charlie Thomas interception on 3rd and 6 at the Pitt 49 at the end of the third quarter
4. 3.64 EPA – Hassan Hall 62 yard run to get GT down to the Pitt 10 late in the fourth quarter
5. 3.64 EPA – Georgia Tech forces 4th and 4 incompletion in the second quarter, takes over at own 37 after Keion White pressure

The balance sheet for GT here is reflective of some of the concern coming out of this one, despite the win. In these most helpful plays, there are no passes and only one offensive play. Turnovers, while extremely helpful for winning an individual game, are not sustainable at this level (15.2 EPA on these plays alone).

Most Hurtful Plays

1. -3.35 EPA – Jeff Sims incompletion on 4th and goal from the Pitt 1 in the middle of the fourth quarter
2. -3.06 EPA – Kedon Slovis 26 yard touchdown pass late in the fourth quarter to make it 19-14

On this side though, we once again see just how well the defense did at limiting anything explosive. There’s only one Pittsburgh offensive play here, and that came on the coverage bust we mentioned above. This kind of containment and discipline on the defensive side is great to see.

Tracking Season Goals

*I set these goals for the 2022 season in some of my offseason preview work. We will be tracking them as we go this year.

The goals that weren’t hit in this one largely hang on the quarterback and the offensive line. As Akshay noted yesterday, QB play for GT is at the bottom of the P5 right now. That’s not being helped by the struggles on the offensive line, but they go together. Sims creates some of the pressures with a lack of pocket awareness. It may seem counter intuitive to throw more on second and long with that kind of bad QB play, but calling the game that way at least increases the chance of avoiding the difficult 3rd down situations that are killing the offense right now.

On the other side, we see again the amazing turnaround that has happened with the Georgia Tech pass defense. To date, this unit has gone from last in the FBS in 2021 to above average. That’s worth celebrating.

Takeaways

1. Beyond the offensive and defensive numbers we’ve looked at, we would be remiss if we didn’t mention the unit now under the direction of Jason Semore. For the first time this season, Georgia Tech was on the positive side of the Special Teams EPA battle, posting 1.8 total EPA compared to the -4.5 EPA of Pitt’s Special Teams. Hats off to Gavin Stewart, David Shanahan, and the punt protection team for the clear improvement. Beyond that, Georgia Tech benefitted from the absence of any clear coaching gaffes. As Brent Key told us, there’s very little schematically you can change in week 1 as a team’s interim coach, but you can remove the costly mistakes, and this staff did a good job of that.
2. There are massive issues in the passing game which are being exacerbated by poor offensive line play. Georgia Tech sits at 100th nationally in EPA/pass. It’s getting close to time to give someone else a series.
3. Overall, Saturday night was full of joy. The team and coaches looked and sounded as if a weight had been lifted, as if they were now free to play and coach to their potential. There are still limitations of course, but seeing a team be free of what had become an anvil is exciting and provides much more hope as Tech enters the heart of its conference schedule. There’s still a lot to show up, play, and cheer for. Let’s go.