Final Score: Virginia 16, Georgia Tech 9
Model Prediction: GT by 5, GT to cover: Incorrect
Projected EPA (Offense and Defense) Margin of Victory: UVA by 8.8
GT Win Probability (Based on Success Rate, Yards Per Play, and EPA): 16%
Perhaps the best decision I’ve made this year was agreeing that Akshay would write Advanced Stats Reviews focusing on the offense, while I would focus on the defense. Because of that agreement, I didn’t have to write this and instead get to focus on the side of the ball where GT continues to impress and improve. Last Thursday night, Georgia Tech’s defense outscored its offense 6-3 and absolutely played well enough to deserve a victory. Alas, that wasn’t to be. We’ve put the offense to bed though, so let’s dig into what the defense did against Virginia.
Advanced Stats Comparison
When Virginia Had the Ball
Perhaps the most ignominious number from this game is that Georgia Tech held Virginia to a 0th percentile performance in Total EPA and still lost the game. That should be enough to clearly show which side of the ball was most responsible for the frustrating loss. Beyond that key top line metric, the defense held UVA to below average numbers almost across the board. We should note that Virginia’s offense has been well below average coming into the game, as we saw in our preview. Georgia Tech still managed to hold the Cavaliers below their season-long pace in numbers like EPA/rush, run stuff rate, and explosive play rate. In other areas, Virginia was right around its season-long percentile performance.
Georgia Tech absolutely shut down Virginia’s ability to run the ball. Without Brennan Armstrong’s 43 yard scamper in the fourth quarter on what was a called pass play, Virginia carried the ball 37 times for 112 yards, which is right at three yards per carry. Virginia continued to try running the ball to end the game, and Georgia Tech kept getting stops to give the offense one more opportunity to mount a scoring drive.
Overall, PFF graded the Georgia Tech run defense at an 89.2, by far GT’s highest mark of the year. Against FBS opponents, the previous high in 2022 was a 74.0 against UCF. At the individual level, Keion White led the way with an 81.9 grade, but it was a well-rounded effort. D’Quan Douse checked in at 76.8, Charlie Thomas at 75.5, Clayton Powell-Lee at 75.0, LaMiles Brooks at 72.7, and Zeek Biggers at 72.4. A few defensive linemen graded out around 60, but this was a very solid all around effort, holding UVA to 25th percentile or worse in all the key rushing categories.
Our key disruption metrics provide more encouragement about how this defense has developed. Looking from the UVA side of things, Tech’s defense forced its opponent into a 24th percentile performance in run stuffs allowed, 13th percentile in havoc rate allowed, and 25th percentile in pressure rate allowed. For the third straight game, the pressure rate for Tech was well above average. The defense would benefit from converting more of these pressures into sacks, but the pressure rate itself tends to be more predictive of future sacks than the actual sacks in a game. Keep getting pressure like this, and the sacks will come.
Ace Eley once again led the way there with 5 pressures, Sylvain Yondjouen had 4, and Charlie Thomas and Akelo Stone had 2 each. Thomas and Eley had the highest pass rushing grades on PFF, followed by Stone, White, and Yondjouen. The ability to get pressure from multiple defensive linemen and linebackers has been a very encouraging development for the defense this season.
Building on the success up front, the back end of the Georgia Tech defense continues to hold up well. This is now the fourth straight (and fifth overall) game holding the opponent to negative EPA/pass, which is something the defense did only twice in 2021. The CPOE showing was also excellent, which is a good indicator that the secondary is not allowing opposing quarterbacks to complete much down the field. The secondary performance was led by LaMiles Brooks, whose pass coverage graded out at 81.4. Of course, he scored GT’s only touchdown of the night, and he was the clear MVP of the game for Tech. Zamari Walton and Myles Sims continue to turn in solid performances outside, grading out at 72.5 and 70.5 respectively.
Clayton Powell-Lee did not fare as well in his second start, trailing the rest of the group with only a 51 grade. We will likely see opposing passing games increasingly target him given the relative strength around him, and it will be a great challenge to hold up in coverage once that happens.
One interesting development is that the early season talk about getting more guys reps in the secondary has not manifested at all. The five starters in the secondary got almost all of the snaps in this one; those guys continue to get results, but there does appear to be a quality depth issue in the secondary.
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
- 6.30 EPA - LaMiles Brooks 37 yard INT return for a TD, GT’s only touchdown of the night.
- 4.11 EPA - Trenilyas Tatum forced fumble, recovered by Brooks on Virginia’s opening series
- 3.27 EPA - Jalen Huff blocks UVA punt, recovered by GT at the UVA 36.
Once again, we see an EPA highlights section that includes no offensive plays for Georgia Tech. This is a systemic issue, especially if the running threat of Sims is missing going forward. The combination of a woeful offensive line and a lack of playmakers is making it very, very difficult to gain much in the way of efficient or explosive offense.
Most Hurtful Plays
- -4.62 EPA - Jeff Sims’s interception in the end zone on 3rd and goal at UVA 8
- -4.10 EPA - Zach Gibson’s fumble on 3rd and 5 on the GT 9, setting up UVA’s final points mid third quarter
- -3.10 EPA - Brennan Armstrong 43 yard run on 2nd and 10 from the UVA 9 with 7 minutes left
- -3.08 EPA - Armstrong 44 yd TD pass to Wicks, which included several missed tackles down the sideline
On this side of the ledger, the two GT turnovers were especially costly in the context of where they took place on the field, and the defense’s only two busts of the day show up here as well. The pass to Wicks was an enormous touchdown for the struggling UVA offense, and the Armstrong run created a massive time and field position shift late in a tight game.
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.
GT Season Goals vs. FSU
|Metric||Season Goal||This Week||Season Long|
|Metric||Season Goal||This Week||Season Long|
|GT CPOE||>= 2%||1.00%||-4.50%|
|Pressure Rate Allowed||<=26%||39%||34%|
|Run Rate on 2nd Down and Long||<=40%||47%||56%|
|Average Depth of Target||>=9||7.7||9.1|
|Defensive Passing EPA/play||<= 0.08||0.43||0.03|
|Defensive Havoc Rate||>=18%||12%||16%|
|Defensive Pressure Rate||>= 27%||33%||30%|
We can clearly see here the bifurcated performance between offense and defense through the lens of our goals. We hit all of our defensive goals in this one, and two out of three are on track for the season, with havoc rate running just one percent behind. On offense, the first three goals were missed for the game and are failing for the season, and the manner in which the ADOT goal has been hit calls into question whether that is a valid goal for this particular team. Clearly, throwing the ball further down the field has not been effective thus far this season, and especially without the threat of Sims running, teams should have ample opportunity to cover up those deeper targets.
- Despite the offense’s failure, the defense continued to provide stout opposition against Virginia and has now put together four impressive performances in a row. Friend of the program Bud Davis puts together weekly opponent-adjusted EPA metrics, and GT now sits just on the cusp of the second tier of all FBS defenses (see the left graphic in the embedded tweet below). This continues to be an incredible turnaround story from last year.
Rushing & Passing Opp Adj EPA/play: pic.twitter.com/dywu27l4XW— Joshua "Bud" Davis, PhD (@JBudDavis) October 24, 2022
2. A defensive performance that holds the other team to 0th percentile EPA and includes four turnovers should be a winning formula. Unfortunately, the Georgia Tech offense seems to have run out of ideas to compensate for the terrible offensive line, the lack of playmakers, and the inaccuracy at the quarterback position. Even when Sims was involved in this game, his inability to run meant GT had little to chance on offense, and that little chance completely dried up when the back up came in. Health and availability at the quarterback position are obviously the key determining factors in whether GT will have a legitimate chance to win again this season.