clock menu more-arrow no yes

Filed under:

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

New, 5 comments

GT stood up to one of the big boys of college football on Saturday and had a chance to win a game that no one thought would be close.

Georgia Tech v Clemson Photo by Jacob Kupferman/Getty Images

Final Score: Clemson 14-8

Model Prediction: Clemson by 25

Projected EPA (Offense and Defense) Margin of Victory: Clemson by 8

GT Win Probability (Based on Success Rate, Yards Per Play, and EPA): 22%

In a game that Vegas gave GT about a 2% chance to win, Georgia Tech played well enough to win the game almost a quarter of the time. I haven’t talked to any GT fan who seriously expected to win going into Saturday, and coming within one score with a chance to tie the game from the one yard line in the final minute exceeded my wildest expectations.

The offensive coaching staff, facing what is surely a top five defense nationally, had a great plan to manage the game amidst severe limitations. The defensive staff played from a look that they had not shown before, and it worked masterfully to accomplish one thing we were looking for going in: limit explosive Clemson passing plays. It was a hopeful day for the Georgia Tech football program, but of course, there are many areas where we can learn and grow. Let’s dig in.

Success Rate Comparisons

GT vs. Clemson Success Rates

Success Rate O GT Offense D Opp Offense National Avg
Success Rate O GT Offense D Opp Offense National Avg
Down 1 29.00% 1 32.10%
2 52.00% 2 54.50%
3 20.00% 3 66.70%
4 66.70% 4 0.00%
Qtr 1 10.00% 1 36.80%
2 47.80% 2 31.30%
3 38.50% 3 54.50%
4 35.70% 4 65.00%
Pass P 36.20% P 34.60% 41%
Rush R 34.60% R 53.80% 42%
Overall 36.50% 47.00% 42%
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.

Georgia Tech started slowly, succeeding on only one of ten first quarter offensive plays. The offense found a great rhythm in the second quarter but could not sustain enough efficiency in the second half to cross the goal line. Clemson, on the other hand, almost entirely abandoned the passing game in the second half and managed to crawl to a win on the legs of Will Shipley and DJ Uiagalelei.

A 10% success rate deficit is not the goal for GT, but the past two meetings saw Clemson put up a 17% and then a 38% margin over GT. Georgia Tech stayed on the football field with Clemson this year, and that’s cause for some celebration.

Let’s turn now to the full advanced box score to see some highlights and lowlights for GT.

Advanced Stats Comparison and Positional Breakdowns

GT vs. Clemson Advanced Box Score

Adv Box Score GT Offense Opp Offense National Avg
Adv Box Score GT Offense Opp Offense National Avg
Snap Count 75 66 71.5
# Pass Plays Called 47 26 31
Avg Starting FP 83.56 71.78 70.5
YPP 3.73 4.3 5.7
YPA (incl. sacks, scrambles) 4.68 5.46 7.39
% of Passes on 1st Down 66% 32% 40.17%
% of runs on 2nd and long 24% 36% 39.80%
Avg EPA/play -0.1 0.02 -0.01
Avg EPA/pass -0.06 -0.04 0
Avg EPA/rush -0.12 0.06 -0.01
Total EPA -7.23 1.47 -0.96
Avg Air Yards / Completion 6.11 4.18 6.14
Air Yards / Attempt 8.09 6.92 8.89
CP 59.38% 70.83% 62.54%
CPOE -3.21% 6.16% 1.90%
Line Yards per Carry 2.85 3.08 2.55
Opportunity Rate 45.83% 57.89% 42.42%
Power Success Rate 100.00% 68.60%
Stuff Rate (Offense) 19.23% 10.26% 19.17%
Havoc Rate 9% 9% 21.00%
Pressure Rate 12% 30% 27.00%

Quarterback

GT QB vs. Clemson

Player Success Rate EPA/play EPA/pass CP CPOE CPOE10+ CPOE20+ EPAwPressure EPAnoPressure
Player Success Rate EPA/play EPA/pass CP CPOE CPOE10+ CPOE20+ EPAwPressure EPAnoPressure
Yates 0.36 -0.08 -0.06 0.59 -0.03 -0.22 0.15 -0.64 0.19

Jordan Yates won the starting role with his play last week, got the start in Death Valley, and acquitted himself with aplomb. Last week, we focused on the tremendous Completion Percentage above Expectation that Yates put up. This week, that metric came in just below average, at -0.03%, but context requires noting that GT’s CPOE number was -16% against Clemson last year.

This week, I want to focus your attention on the splits between how Yates performed under pressure, or not. Predictably, Yates faced intense pressure from the Clemson front. He struggled to the tune of -0.64 EPA/pass in those situations. Significantly though, he posted a very strong 0.19 EPA/pass play when not under pressure. Football Outsiders has done research demonstrating that a QB’s EPA numbers are much stickier (in terms of year to year consistency) when facing no pressure. That is, the impressive number Yates put up with a clean pocket should tell us more about what to expect about his future performance. That kind of EPA average against a destructive defense bodes well for what we can expect from Yates for the rest of the season. He earned another start and more with his showing Saturday afternoon.

Rushing

GT Rushing vs. Clemson

Runner Carries Rushing Success Rate Highlight Yards
Runner Carries Rushing Success Rate Highlight Yards
Gibbs 11 0.36 -2
Smith 2 0.5 4.5
Mason 9 0.22 -1.5

This author thought that Georgia Tech’s only hope for offensive success on Saturday would come on the ground. This author was wrong. In success rate, yards per play, and EPA/play, Georgia Tech was better through the air on Saturday. The success rates for Gibbs and Mason are just dismal, and they also weren’t able to create outside of what the offensive line gave them. We have to stop for a minute and ask why Dontae Smith got only two carries all game. He had one run that popped, in addition to an enormous 19 yard screen play, on just 3 touches. If he had enough touches, Smith would rank in the top five nationally in PFF’s running back grades for the season. It’s time for him to take over the #2 back role.

Receiving

GT Receiving vs. Clemson

Receivers Receiving Success Rate Avg Target Air Yards Targets % of Team Air Yards YACatch
Receivers Receiving Success Rate Avg Target Air Yards Targets % of Team Air Yards YACatch
McGowan 0.5 12 6 16.94% 36
Smith 1 -2 1 -0.47% 21
Carter 0.8 10.2 5 12.00% 11
Sanders 0.67 4.33 3 3.06% 6
Gibbs 0.67 1.33 3 0.94% 11
Norris 0 35 1 8.24% 0
Devaney 0.4 4.6 5 5.41% 2
McCollum 0 1 1 0.24% 1

Kyric McGowan made the big plays, Malachi Carter got open consistently in the intermediate areas, and Dontae Smith busted one big screen. Between the three of them, they were successful on 8 of the 12 targets they received. Yates clearly has a chemistry with McGowan, and it’s going to be exciting to see them keep growing together. Dylan Devaney was finally able to return from injury and picked up two huge third down conversions. The one deficiency that showed up was the inability for anyone in the receiver room to get space for the kind of chunk play that GT needed in order to get into the end zone. The receivers were solid, but no one was able to take the top off of the defense.

Offensive Line

GT OL vs. Clemson

Player # of FLOPS
Player # of FLOPS
Cooper 1
Minihan 1
Johnson 4
Lay 1

It was always going to be a tough day for GT’s offensive line against what is a top 3 defensive line in college football (perhaps behind only the one yet to be faced on the Saturday after Thanksgiving). Overall, the line performed better what I thought they were capable of. In our preview, we said that if GT chose to drop back 40 times, they might get sacked 10 times. In reality, they dropped back 47 times but were only sacked 4 times. Now, they did give up a 30% pressure rate and 19% run stuff rate, but Clemson managed a havoc play only 9% of the time. Yates had to do plenty of eluding pressure, but the line managed to put up far more resistance than last year, when Clemson posted a 29% havoc rate and 45% run stuff rate.

Looking at the individual play, our FLOP numbers (failures of lineman or penalty) probably undersell how many plays were destroyed up front. I had to watch the game on a TV that was not exactly equipped with the latest technology, so I wasn’t able to see some details and numbers like I would want. But the point remains that there was trouble across the line but nowhere more so than the RG position. This one just confuses me, because the coaches seem absolutely unwilling to try someone else there. It’s time for that to change, even if just for a couple of series to see if there’s another combination that can hold up more effectively.

Defensive Disruption

GT Disruption vs. Clemson

Player Defending Havoc Plays # of Pressures # of Run Stuffs
Player Defending Havoc Plays # of Pressures # of Run Stuffs
Thomas 1 0 1
Jackson 1 0 0
Oliver 0 1 0
Harris 1 0 0
King 1 0 1
Ivey 1 1 1
Thomas 1 0 0
Kennard 0 1 0
Biggers 0 0 1

Georgia Tech just wasn’t able to generate the disruption up front that Clemson’s offensive line weakness seemed to allow. All of the top line numbers were well below average: 10% run stuff rate, 9% havoc rate, and 12% pressure rate. Last week’s star, Jordan Domineck, was largely held in check. Jared Ivey continues to show up and looks like a future All-ACC guy to me. Zeek Biggers is making a case to be in the top 3 of the DT rotation.

But the star this week- even though these numbers don’t quite show it - was Charlie Thomas. As you’ll see below, he was targeted three times in pass coverage but allowed 10% completion under expected and only 2 yards after the catch. He had 12 tackles, including one for a 7 yard loss, and many more just beyond the line of scrimmage. To give you an idea of his level of play, PFF has Thomas graded as the #2 linebacker in the country after 3 weeks. Georgia Tech hasn’t seen that kind of linebacker play since the days of Phillip Wheeler in 2006 or Darryl Smith in 2000.

Pass Coverage

GT Coverage vs. Clemson

Player in Primary Coverage Targets CPOE Allowed YAC Allowed
Player in Primary Coverage Targets CPOE Allowed YAC Allowed
Thomas 2 -24.00% 7
Carpenter 6 17.73% 19
Swilling 2 -0.80% 0
Oliver 1 27.67% 1
Walker 4 15.27% 0
Sims 1 -45.20% 0
Allen 1 33.00% 12
Thomas 3 -10.67% 2

Entering the game, the primary defensive key I identified was to hold Clemson to 5 or less pass plays that created 1.5 EPA or more. On its first drive, Clemson had two such plays. And they had one the rest of the game. Coach Thacker designed a defensive formation and personnel grouping that did what it needed to do to control the Clemson passing game and give GT a chance to win. It wasn’t perfect, but it was so much better than what GT’s back end showed in the first two games. Tobias Oliver and Charlie Thomas jumped out in coverage, and Juanyeh Thomas really rebounded from a disappointing outing last week. This was by far the best game of the season for GT defending the pass, and we will need much more like it against Sam Howell and Josh Downs next Saturday in the Benz.

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 an 8 point projected win for Clemson. Keep in mind, Clemson had a 40 point EPA margin at halftime against GT last year. We will say it again; this performance was night and day from the previous two efforts that CGC’s squad rolled out against the Tigers. As always, we’ll take a look at the most helpful and hurtful plays for GT. You’ll note that the magnitude of these plays is much lower than in most weeks.

Most Helpful Plays

  1. Quez Jackson’s forced fumble against DJU on 3rd and 1 from the GT 20 in the 4th quarter. This was the play that kept GT alive. 3.40 EPA.
  2. Jordan Yates’s 34 yard pass completion to Kyric McGowan on GT’s opening possession in the third quarter. Not getting points on this drive really hurt. 3.05 EPA.
  3. Georgia Tech forces a turnover on downs from Clemson on 4th and 2 from the GT 19 in the second quarter. What an effort from the GT defense. 2.89 EPA.
  4. Jordan Yates’s 9 yard completion to Malachi Carter on 4th and 1 from the Clemson 34 to set up GT’s first field goal of the game. 2.36 EPA.
  5. Jordan Yates’s 7 yard completion to Dylan Devaney on 4th and 7 from the Clemson 35 late in the 4th quarter. An absolutely incredible play by Yates after a bad snap. 2.30 EPA.

3 passing plays and two massive defensive plays gave GT a chance in this one.

Most Hurtful Plays

  1. GT’s failed 4th and goal play from the Clemson 2 inside of the final minute. It was an unexpected play call, but the key block was missed, and Devaney never had a chance. -3.35 EPA.
  2. Clemson’s 17 yard completion on 3rd and 16 early in the 2nd quarter. #18 was in the wrong place and took a bad angle to allow the receiver to get 12 yards after the catch. -2.32 EPA.
  3. Clemson’s 12 yard completion on 3rd and 6 on its opening drive to move to the GT 45. -2.04 EPA.
  4. Will Shipley’s 4 yard run to convert 3rd and 2 inside the GT 10 and set up Clemson’s first TD. -2.01 EPA.

The 4th down play for GT was a dagger. Otherwise, the lack of big numbers on this list was hugely encouraging for GT. No turnovers for the offense. No killer explosives allowed by the defense. Again, this was the kind of stuff that gave GT a chance. Hats off for the way the team showed up to play.

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.

GT Season Goals - Clemson

Metric Season Goal This Week Season Long
Metric Season Goal This Week Season Long
GT CPOE >= 2% -3% 1%
Pressure Rate Allowed <=26% 30% 26%
Pass Rate on 1st Down >=50% 66% 51%
Defensive Passing EPA/play <= -0.06 -0.04 0.22
Defensive Havoc Rate >=21% 9% 8%
Defensive Pressure Rate >= 27% 12% 19%

Coach Patenaude put the keys in Jordan Yates’s hands, and he came through time and again. The CPOE for the week and the season are right below our target, but Yates showed he has rightfully earned the trust of the coaches and should only improve from here. The pass defense was far and away the best its been all season, but the disruption numbers still aren’t anywhere near where we want to see them.

Takeaways

  1. Same as last week: Jordan Yates is the starting quarterback until further notice. He showed out against Kennesaw State. But he faced the heat against Clemson and kept his head. He had GT in position to force overtime with less than a minute to go. He knows what he can do and what he can’t. Give Yates a clean pocket, and he will go to work.
  2. Can he get some help? The line allowed a huge pressure rate and provided very little room for the RBs to work. Dontae Smith has to get the ball more, and Brent Key needs to shuffle things around at RG.
  3. The defense is beginning to shore up the weak links. As of this writing, PFF grades every GT defensive tackle above average in their player grading, the two starting DE’s have outstanding grades, Charlie Thomas is #2 in the country for linebackers, and Tobias Oliver splashed onto the scene the last two weeks. This side of the ball is starting to show the fruit of talent improvement but also more effective schematic deployment. The Sam Howell offense will certainly test that.

There’s plenty of improvement still to see, but Georgia Tech played above its head and stayed close to an elite opponent for the first time under Geoff Collins. Sure, DJU is nowhere near Trevor Lawrence, but the talent discrepancies all across the field didn’t overwhelm GT like they had in previous years. This was impressive, encouraging, and fun! Take a second today to smile about the fact that we were one yard away from scoring the potentially game-tying touchdown in Death Valley with 1 minute to play. Georgia Tech showed up and showed out and outdid all of my expectations.