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

Final Score: Miami 35, Georgia Tech 13

Model Prediction: GT by 0.9, Miami to cover: Correct

Projected EPA (Offense and Defense) Margin of Victory: Miami by 27.6

GT Win Probability (Based on Success Rate, Yards Per Play, and EPA): 18%

Following a road win in Blacksburg, Georgia Tech returned home as a slight favorite against a Miami team that looked to be reeling. Alas, GT did not handle its relative prosperity well, putting up another depressing offensive performance and letting go of the rope late to fall 35-14 to Miami. Miami started its third string quarterback, true freshman Jacurri Brown, and came in with a clearly run heavy game plan. Georgia Tech tried to duplicate much of last week’s plan with Zach Pyron under center, but things were glitchy and then completely fell apart once Pyron was hurt and Gibson had to come in. Tech’s four interceptions allowed Miami to run away with things in the fourth quarter. Let’s dig into the numbers to see more about how the defense performed against the Hurricanes.

Advanced Stats Comparison

The box score doesn’t initially look as lopsided as I expected, but many of the places where GT shows an advantage are more related to “game state” type things than performance. That is, Miami didn’t need to try and throw the ball, and when the Hurricanes did, they did so very conservatively because it was working. GT did hold a few small advantages but not in the most significant places. Success rates and yards were play were relatively close, but the massive EPA edge largely reflects the turnovers and GT’s inability to make big plays. The Completion Percentage and CPOE margins are jarring.

When Miami Had the Ball

Rushing Defense

Miami put up average efficiency on a very high rate of running. The most concerning metric from the GT side is once again the Opportunity Rate, as Miami was able to convert 55% of its carries into four or more yards. Allowing that much efficiency on that much volume (over 2/3 of the Miami plays were called runs) when it was clear what the plan would be with Brown at QB was disheartening. Key confirmed that assumption from the GT coaching staff in his presser, so it was more a matter of being out schemed and outmanned than ignorance of what Miami would do. The biggest things to jumps out to me is that Georgia Tech did not try a single snap with three linebackers. Either the staff didn’t think that anyone else was ready to play, or the staff wasn’t cognizant enough of what Miami was doing. Either way, it’s very concerning for the state of that position going forward, as Thomas and Eley only have two more games apiece in a GT uniform.

Looking at the individual side of things against the run, the PFF grades aren’t as bad as expected. GT actually had an excellent team grade (for us) at 73.4, but that looks to be skewed by some garbage time numbers, as several backups for GT were the highest graded players. Amongst starters and regular individuals in the front, Sylvain Yondjouen and D’Quan Douse led the way at 71, Eley was at 69, Scott at 64, White and Biggers at 62, Thomas at 61, and Kennard was all the way down at 48. Again, no other linebacker played a single snap, which is concerning in many ways. Finally, the team tackling grade was only a 46, which is horrible.

Defensive Disruption

The defense was not disruptive. For the day, GT posted one of the lowest havoc rates I’ve ever seen in a game between two evenly matched teams. The run stuff rate was below average. The pass rush did generate a 40% pressure rate, but it required a 60% blitz rate to get that. Again, we see evidence that the defense isn’t getting home enough on blitzes and is not on the right side of the risk/reward balance there. We’ve been warning that the turnover barrage would regress to the mean, especially for a defense with a below average havoc rate. We saw that this week, as the defense had no forced fumbles and no pass break ups, much less an actual turnover. Georgia Tech’s defense was not disrupting anything Miami wanted to do.

Pass Coverage

The pass coverage wasn’t bad, nothing like 2021 levels, but it wasn’t as good as it had been for most of the past six games or so. Per PFF, the team coverage grade was still a 68.9, but things weren’t quite as sharp, as evidenced by the bust on Will Mallory for the opening touchdown for Miami. Jacurri Brown’s CPOE of +0.07 is extremely good, but he achieved that number while completing incredibly short throws on average. The average target and completion depth for Miami were both in the 20th percentile range. The Hurricanes did grab 7.3 yards after the catch on average, but that’s understandable given how low the average completion depth was. Focusing on the starters in the secondary, Wallace graded at 76, Brooks at 70, Walton at 68, Clayton Powell-Lee at 67, and Sims at 58. The worst grade for a regular was Thomas at 51, which was heavily influenced by that opening touchdown bust.

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. 2.9 EPA – 42 yard completion from Pyron to McCollum from the GT 9

This is a short list because GT forced no takeaways and had virtually no explosive offense.

Most Hurtful Plays

1. -11.8 EPA – Zach Gibson 99 yard Pick Six
2. -4.3 EPA – Pyron interception on flea flicker. This was horrible play design and execution. There was no actual misdirection to cause defenders to take a false step, and the only downfield target was your injured tight end.
3. -2.9 EPA – Gibson interception on 2nd and 6 from the GT 18
4. -2.8 EPA – Gibson incompletion on 4th and 3, trailing by 14 by 6 minutes left

The four worst plays for GT all came when trying to pass the football. The defense once again did a great job of limiting explosive plays but was badly hurt by the 15.6 points of turnover luck generated by the offense’s foibles.

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 offense once again fell short of our goals, and the low CPOE number looks even worse in context of the low pressure rate allowed. This one was more on the QBs than on the line’s inability to give time. The defense allowed its worst defensive EPA/pass number of the season, although given Miami’s low volume of passing, that was heavily influenced by just a few plays. The defense also posted its worst havoc rate of the season by far.

Takeaways

1. The offense doesn’t have a reasonable quarterback option left.
2. The defense started poorly, came into form, and then let up some after the barrage of interceptions. Once again, they did a good job of limiting explosives, but the incredibly high rushing efficiency allowed isn’t good enough.
3. The final two weeks of the season are going to be rough.