Georgia Tech Football: Advanced Stats Review - GT vs. Pittsburgh

Final Score: Pittsburgh 52-21

Model Prediction: Pitt by 0.26, GT to cover: Incorrect

Projected EPA (Offense and Defense) Margin of Victory: Pitt by 30

GT Win Probability (Based on Success Rate, Yards Per Play, and EPA): 16%

What does GT’s win over UNC mean in light of the crushing loss to Pittsburgh on Saturday afternoon? I’ve been wrestling with that question for 48 hours. Joey and I both wrote last week about the UNC win demonstrating proof of concept for Collins and his staff. I still think that is true, but I worry that the kind of consistency necessary to compete in the ACC will continue to elude this staff. Well coached teams don’t have performances this polarized.

Let’s dig into what all went wrong.

Success Rate Comparisons

Success rate is the baseline metric for efficiency. As a reminder, a successful play gains 50% of the needed yards on 1st down, 70% on second down, and 100% on third or fourth down.

If you told me that GT would post a 47% Success Rate against Pitt, I would have guessed we either won or lost by no more than one score. But if you told me they would post 52% before garbage time started, I would have known GT was in trouble. The Pitt rushing attack- non-existent before coming into Atlanta on Saturday- hit for 46% success. And the passing game was an otherworldly 55%. Even with Jeff Sims playing an outstanding game through the air and with his legs, GT couldn’t come close to keeping pace.

Advanced Stats Comparison and Positional Breakdowns

Quarterback

As expected, Jeff Sims got the start. Although there was some clamoring for Yates to be inserted after Sims threw two interceptions on his first two throws, the coaches were right to stick with Sims. The two early interceptions- which, seemed to be about 80% the fault of the offensive line- cost Georgia Tech about 13 expected points. After that, Sims earned back about 9 Expected Points before garbage time started. His CPOE of +12% was a career best, and his EPA when not under pressure was outstanding.

Let me say it loud and clear: Jeff Sims was excellent on Saturday. His PFF grade has risen from 39 to to 54 to 76 over the past three weeks. He was not the reason Georgia Tech lost, and he should clearly be the starting quarterback moving forward. Yes, the turnovers were brutal. Yes, ball security is paramount. But Jeff Sims is about the only reason this game wasn’t 52-0 on Saturday.

Rushing

The top two running backs for Georgia Tech had zero successful carries between the pair. That is a recipe for trouble. The repeated efforts to run inside zone on first down in the first half and in key short yardage situations were fruitless. The offensive line could do nothing to generate seams inside the box. Most of the Jeff Sims rushing production came on called pass plays, but he also generated one 23 yard carry on a called run on GT’s first offensive play from scrimmage. Dontae Smith produced - as he has every single week this year - but the running game never got out of the blocks.

The most stunning number in this department is the 39% run stuff rate that GT conceded; Clemson had put up the previous season high against GT in stuffing runs. They stuffed 19%, less than half of what Pitt did. Pitt sold out to stop the run, and it worked. Coach Patenaude did well to call passing plays on 62% of first down snaps before garbage time, but he accompanied that with 50% run plays on 2nd and long, which set these guys up for failure.

Receiving

Kyric McGowan had been the best GT option in the passing game, and he missed Saturday with an undisclosed injury. The bulk of the receiving responsibility ended up on the shoulders of one senior and three freshmen — Malachi Carter, Jahmyr Gibbs, Malik Rutherford, and Nate McCollum.

It was an encouraging performance. Four guys posted 50% success rates on their targets, and the other two leading receivers ended up at 40%. Rutherford, Carter, Sanders, and especially Gibbs all posted monster after the catch numbers. The future of the passing game looks bright, even as the 2021 leader had to miss the game.

Offensive Line

The offensive line was missing Kenny Cooper on Saturday, and according to various reports, several of the guys who did play were dealing with varying degrees of injury. Still, it was a bad day. The 39% run stuff rate allowed was not just the worst of the 2021 season but was worse than every game last year save the Clemson debacle. The pressure and havoc rates that Pitt generated were both right around average, but the timing and severity of those plays had an outsized impact on the final outcome.

The left tackle position is solid. The center position is adequate. The rest of the line is a rotating mess. Between four incoming transfers and eight linemen signed in the last two classes, more improvement should be seen by now. The talent acquisition and development piece is not where it should be, especially when this position coach makes more than double any other on staff.

Defensive Disruption

It looks a lot more like the disruption performance against UNC was owing to their offensive line issues rather than some kind of new normal for Georgia Tech. Pitt’s offensive line is probably just above average, but there is no reason that GT should have fallen back to a havoc rate in the mid single digits. The pressure rate ended up looking good, but much of that came in the third quarter before garbage time had technically kicked in. At halftime, that pressure rate was only 22%, and that is much more indicative of how ineffective the GT rush actually was.

Coach Thacker largely stuck with the 3-3-5 look that had brought so much success the past two weeks. But when it was clear early on that GT was not pressuring Pickett and that Pickett would pick apart the secondary with that kind of time, the defensive staff made no significant schematic changes. That blows my mind. Was it stubbornness? Lack of creativity? Failure of preparation?

Pass Coverage

Single digits are supposed to mean something in this program. On Saturday, they meant defensive backs getting scorched. The six single digit guys who I charted as being targeted on Saturday all gave up +15% or more CPOE and a combined 60 yards after the catch.

Kenny Pickett had put up 0.40 EPA/pass entering the game against a weak schedule. Georgia Tech let him go for 0.59 EPA/pass. The safety play defies words. At the very least, Jaylon King must start the rest of the season. That won’t be enough, but Carpenter simply cannot cover anymore. He looks lost more times than not. Give Sims some more burn. He might not be better than the other guys at corner, but at least he hasn’t proven that yet. The amount of regression and lack of development from what is on paper tied for the most talented area on the team is astounding.

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.

The EPA totals for this game leave us with a 30 point projected win for Pitt. Numerous fans are pointing to the magnitude of the two interceptions (and they were significant) but they accounted for only about 40% of the final EPA margin. Pitt beat GT soundly.

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

Most Helpful Plays

1. Jahmyr Gibbs’s 71 yard catch and run on a beautifully executed screen play to get GT inside the Pitt 5. We won’t talk about what happened next. 4.95 EPA.
2. Jeff Sims’s 44 yard touchdown pass to Nate McCollum. A wonderful throw after McCollum dusted the corner. 3.69 EPA.

These two explosives once again showed the potential of the GT offense. Unfortunately, one of them led to a red zone disaster, and there simply weren’t enough explosives on the ledger. Also noticeably absent is any significant play by the defense.

Most Hurtful Plays

1. Jeff Sims’s second interception, as he was pressured and hit before the ball was intercepted and returned for a touchdown. -7.18 EPA.
2. Jeff Sims’s interception on the second play of scrimmage, as an over the middle throw on an RPO was tipped by an untouched rusher. -5.22 EPA.
3. Kenny Pickett’s 55 yard touchdown throw to give Pitt the 28-7 lead early in the second quarter. -4.82 EPA.
4. Brent Cimaglia’s missed 40 yard field goal after GT had initially moved the ball to the Pitt 7. -3.35 EPA.
5. Georgia Tech’s failed 4th and 2 play from its own 46 late in the second quarter. -2.99 EPA.
6. Georgia Tech’s failed 4th and goal play from the Pitt 4. -2.88 EPA.
7. Kenny Pickett’s 20 yard completion on 4th and 4 to set up Pitt’s 5th touchdown of the first half. -2.86 EPA.

Offensive turnovers, failed short yardage plays, and big Kenny Pickett completions pretty well tell the story of the game.

Tracking Season Goals

*I set these goals for the 2021 season in some of my offseason preview work. We will be tracking them as we go this year.

The quarterback play continues to provide encouragement and represents the possibility of GT going on a second half run. The offensive line play, the lack of havoc, and the poor pass defense work against that possibility.

GT is in an interesting spot. 2-3 is disappointing but not devastating. 3 of our 6 season long goals are on track. The worrying thing is just how far off the pass defense and havoc rate are right now. The defense has gotten relatively consistent pressure but not been able to convert that to havoc plays.

Takeaways

1. Jeff Sims was far from flawless, but he once again threw the ball with a high degree of accuracy and executed explosive passing plays when they were there. The intersection of offensive line and quarterback play can sometimes be hard to parse, as evidence by the two early interceptions. The offense gave up an average pressure rate, but the damage on those pass plays was extreme. This is where the consistency piece for Jeff Sims has to come. Can he minimize the damage against pressure and continue to perform at an outstanding level without pressure?
2. The secondary looks lost. Yes, they faced a difficult task on Saturday, but they didn’t come close to meeting the challenge. The group posted no pass breakups, and I counted 8 passing plays where they simply let a Pitt target run free. There’s a failure of understanding responsibility to go along with poor coverage technique. The deployment of the roster on the back end is baffling.
3. The defensive front could not have had a more polarized experience between the UNC and Pitt games. The run fits were bad against a team that has not been able to run the ball on anyone. The pressure rate when the game was still in the balance was anemic, and the group almost got shutout on the havoc front. Coach Thacker had shown some real flexibility the past two games, but it seemed as if his creativity ran out this week.

The two early interceptions put GT in a hole, but the offense battled back and produced enough to make this the kind of toss-up game we expected going in. The defense let go of the rope. With every chance to seize the goodwill of the fanbase, the GT staff fell on its face once again and still have not managed to produce back to back wins in their now two and a half years on the Flats. Pitt is better than expected, no doubt. But it’s well past time to be getting blown off the field by teams with comparable or worse talent. It’s well past time to look lost and outcoached by a team who came out and did exactly what they’ve done every game this season. Last week, Georgia Tech proved concept, but will they ever prove consistency?