clock menu more-arrow no yes

Filed under:

Georgia Tech Football: Advanced Stats Review Part 2 - The Offense

New, comments

We know the winning formula: GT couldn’t put those pieces together in 2021, but 2022 holds a fresh opportunity

NCAA Football: Georgia Tech at Notre Dame Matt Cashore-USA TODAY Sports

I had high hopes that Georgia Tech’s offense would take the leap in 2021. Jeff Sims had a full year of experience and development under his belt. The offensive line made more sense on paper. Jahmyr Gibbs was an obvious threat to score at any time. Kyric McGowan looked like he could separate differently than the rest of the receiving corps. Patenaude had his quarterback and should have had a good idea how to best deploy him.

I wasn’t expecting Oklahoma, but average seemed well within our grasp. During this same offseason series last year, we provided 4 ways for the GT offense to cement some gains in 2021: “eliminate offensive line weak links, find a go to receiver, improve quarterback play, and call more first down passes.”

What happened in 2021? Let’s look to the numbers to help us tell the story of the season and get ready for an offseason of paramount importance in this program.

Start Up Front

Georgia Tech fans widely saw offensive line as the unit expected to improve most in 2021. Three transfers (Cochran, Pendley, and Kirby) promised immediate assistance, and the staff talked up at least ten guys as ready to contribute. Coming off of two seasons of putrid line play, competence sounded so appealing. Last year, we introduced an idea made popular by Eric Eager, who identifies some positions in football as “And” positions where you need strength across the board to be effective, while some positions are “Or” positions where a singular star can make the difference. Offensive line is an “And” position. The personnel additions last year gave the appearance that at least three, if not four, spots on the GT O-Line should regularly hold up.

Georgia Tech’s 2021 Offensive Line Grades, per Pro Football Focus

That hope proved fleeting. Above, you can see how Pro Football Focus graded and ranked the nine players who ended up seeing the most snaps across GT’s OL this past season. Only three of the nine were in the top 50% nationally, and five of the nine were in the bottom 25%. Chaos reigned across the offensive line, as injuries and poor performance from expected contributors led to raw freshmen getting snaps. “Positional flexibility” looked more like guys playing out of position. This unit failed to consistently protect or open up holes. What kind of impact did the offensive line weakness have on the overall fortunes for GT in 2021?

Impact of Pressure Rate on Passing Efficiency, 2021 Georgia Tech
Impact of Run Stuff Rate on Expected Outcomes, 2021 Georgia Tech

Above, you can see a game by game look at two metrics that tell us a lot about OL play: pressure rate allowed and run stuff rate (% of runs that go for 0 or negative yardage) allowed. In the first graph, you can see a strong negative relationship between the pressure rate allowed and the EPA/pass that GT put up in a particular game. Overall, this relationship has an R Squared value of 0.57 - meaning that 57% of the variance in GT’s EPA/pass can be explained by pressure rate allowed. That’s an enormous figure for what is typically pretty noisy data.

In the second graph, we are looking at the relationship between run stuff rate allowed and total EPA margin, as allowing a high share of stuffed runs often points to a high level of physical mismatch between the two teams. Sure enough, we see that in GT’s 2021 season, 42% of the variance in total EPA margin can be explained by the run stuff rate allowed on offense. As the physical deficiencies along the offensive line were magnified by stronger opponents, GT’s passing effectiveness and overall efficiency margin plummeted. GT’s 2021 improvement had to start up front, but the offense too often got stuck in the mud because the line couldn’t hold.

Once again, hoped for improvement for this fall largely rests on arrivals from the transfer portal. Paul Tchio and Pierce Quick both struggled to earn playing time at their respective blue-blood schools, so we don’t have much established support for the theory that they will be cures for the OL woes. Turnover is good for a position group that struggled, but it’s hard to predict improvement until we have some on the field evidence.

Throw to Win

Impact of Rushing Success Rate on Expected Outcomes, 2021 Georgia Tech
Impact of Passing Success Rate on Expected Outcomes, 2021 Georgia Tech

Last year, we made this interesting discovery that a team’s passing success rate has a much higher impact on playing winning football than does the team’s rushing success rate. For 2020, we saw that GT’s game by game passing success rate explained 47% of the overall variance in total EPA outcome, while the equivalent number for rushing success rate was a paltry 0.2%. Looking at the same numbers nationally gave us R Squared values of 0.43 and 0.29, respectively. We understandably came to the conclusion that GT’s offensive hopes for 2021 rested on significant improvement in the passing game. Once again, our hopes were dashed.

As you can see above, the relationships we found last year, though not quite as starkly contrasted in 2021, held true. The problem was that meant GT’s offense could only tread water this season. From 2020 to 2021, the rushing SR dropped from 48% to 44%, and passing held steady at 41%, while the national average for passing SR has risen all the way to 44%. In this day and age, you must be able to pass efficiently to win, and GT continues to lag behind.

Above, you can see a visual representation of the Completion Percentage over Expectation for the quarterback with the most attempts on each P5 team in 2021, as well as the same metric for every FBS team from 2020. As a reminder, Completion Percentage over Expectation measures something simple but extremely valuable: given where and how far down the field I’m throwing the ball, am I completing more or less passes than expected? It’s quickly become one of the gold standards for evaluating QB play.

After Georgia Tech started a true freshman at QB in 2020, I firmly believed that Jeff Sims would be moving his way well to the right in this significant metric. He put up +9% CPOE against both UNC and Pitt; I was almost sure that we were set at QB. And then GT (including the end of season starts by Yates) posted negative CPOE numbers in every game the rest of the season. In a day when you must be able to throw to win, GT simply couldn’t do it.

While last year’s incumbent starter finished his season at -1% CPOE, Georgia Tech made a recent addition to the QB room from the portal who posted a +16% CPOE number last season, albeit in a relatively small sample size. Chip Long was clearly not content to stand pat with what he was inheriting, and he managed to bring in an experienced player who has demonstrated effectiveness, even if at a lower level of competition. This can only help: Gibson could well win the job, or Sims could put an injury-ridden season behind him and make his leap. Either way, the floor for QB play in 2022 should be higher.

Throw the Ball Down the Field

We’ve seen now that GT’s passing game didn’t take the needed steps forward to produce winning football in 2021. Let’s go one step deeper in trying to understand how GT’s passing potential could be better realized.

Impact of ADOT on EPA/Pass
Impact of ADOT on CPOE, 2021 Georgia Tech

Above, you are seeing a game by game look at how Georgia Tech’s average depth of target on passing plays related to CPOE and EPA/pass results. In both cases, we see a convincing case that throwing the ball further down the field regularly made Georgia Tech a more effective and efficient passing team in 2021. Is this a fluke, or is there something here?

Thanks to Paul Sabin of the ESPN Stats and Info Group, I was able to get my hands on these numbers for all of college football. Though the correlation is not quite as strong as it appears for just GT, the relationship certainly holds as we look across the national landscape. Teams that throw the ball further down the field have more efficient passing games.

You can see Georgia Tech appears almost squarely on the trend line, just below and to the left of center. Fortunately, the new man at the helm of the GT offense could help here. When we did a statistical deep dive on Long’s last three stops, we found that the most consistent effect that Chip Long has on his offenses is creating more explosive plays. Unfortunately, I don’t have access to historical CPOE and ADOT data for Long’s teams, but his clear ability to create explosives pairs well with this finding that increasing depth of target correlates with more effective passing.

In 2022, college football teams must be able to throw to win, and throwing deeper usually means throwing more effectively. The final piece of implementing that strategy for GT next season will be winning on the outside of the field. As we gleaned last year from Eric Eager, Wide Receiver is an “Or” position. One guy can make a lot of difference. Ricky Jeune played a version of this role from 2015-2017, but GT simply hasn’t had that guy since Smelter and Waller in 2014. The passing game can’t fully takeoff until we have him.

From all accounts out of preseason camp, the coaches were hopeful that true freshman Leo Blackburn would have the chops, but he blew out his knee just weeks into camp. The other returners in the receiving corps have had enough time to emerge but not done so. Georgia Tech’s coaching staff should continue combing the portal for another potential difference maker on the outside in order to effectively implement a deeper passing attack.

Looking to 2022

The plan for 2021 rested on these four pillars: “eliminate offensive line weak links, find a go to receiver, improve quarterback play, and call more first down passes.” I’m here to report that we went 0/4! Nothing got significantly worse, but it didn’t get better: the line had plenty of weak links, there was no consistent outside threat , we did not get better quarterback play, and CDP did not provide more advantageous play calling.

Heading into 2022, GT has a new offensive coordinator, the quarterback position is up for grabs, the offensive line has turned over once again, and last year’s most relied upon playmakers (McGowan and Gibbs) are gone. But, all is not lost. The transfers could bring stability to the line. A Chip Long-Zach Gibson offense could throw the ball down the field more consistently and produce the kind of passing efficiency that leads to winning football. Leo Blackburn or an unnamed portal arrival could give GT the outside threat necessary for the offense to hum.

Georgia Tech’s offensive growth stagnated in 2021. There are no guarantees for improvement this year, but there is enough turnover in coaching and personnel that 2022’s results could be different. We have a good idea of what it will take, but we are flying blind about whether next fall’s team will be able to execute the winning formula.