Advanced Stats Review: GT vs. Louisville

Final Score: Georgia Tech 46-27

Projected EPA Margin of Victory: GT by 16.34

GT Win Probability (Based on Success Rate, Yards Per Play, and EPA): 83%

That was fun. After a frustrating start that led to a 21-7 deficit midway through the second quarter, GT displayed an impressive patience and resilience to create a steady and ultimately decisive comeback win. After falling behind early, GT outscored Louisville 39-6 the rest of the game, using a steady efficiency edge and a surprising explosiveness edge to put Louisville away in the fourth quarter. My preseason outlook on this matchup turned out to be much more accurate than my preview from this week. My prediction that Louisville would win this game because of explosive plays turned out to be wrong, and I’m happy for that. Our friends at College Football Data show GT with a 1.72-1.48 advantage in explosiveness (the amount of expected points added per play that was successful). That allowed GT to capitalize on the continuing run of efficient games from Jeff Sims and the offense. Let’s see what else went into this encouraging victory for the Jackets.

Advanced Stats Comparison

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.

Once again, GT wins the success rate battle, with an especially encouraging performance on first down plays. Notice how GT’s efficiency remained steady throughout the game, remaining between 46 and 50 percent in each quarter, while the GT defense shut down the Louisville offense to put the game away, putting up a 27% success rate in the fourth quarter. Though we are continuing to build quality depth, the number of players who were able to contribute on the defense line and in the secondary appeared to sea Louisville down and take over the game in the fourth quarter. So far, GT has posted offensive success rates in three of our four games that are better than any single game last year. This is hugely encouraging, especially given the predictive nature of success rate looking at future weeks.

As you can see, the running game centered on Jeff Sims and Jahymr Gibbs more than in previous weeks. This was a recipe for success, as Gibbs was able to get to the corner on quite a few power running plays, and Sims found creases in the middle of the Louisville defense on called draw plays. These guys are going to be something special for the next three years.

Sims targeted fewer pass catchers than in previous weeks, but the passing game was both efficient and explosive. These targets don’t include throwaways, so you will see that every single receiver targeted has a success rate >=50%. That is spectacular. Sims’s average depth of target was shorter this week (likely due to the weather conditions), but the yards after catch were responsible for creating 7 explosive plays in the passing game.

We’re introducing two new receiving stats this week. RACR compares a receiver’s total receiving yardage to the air yards on targets to them, and anything above 1 is good. Brown, Carter, and Leonard all put up good games in that metric, while Sanders and Gibbs are off the charts because of the ways they were able to create big plays after the catch. WOPR tells us how heavily the receiver is being leaned on in the passing game by combining the proportion of targets with the proportion of air yards. You can see that Malachi Carter was the go to guy for Sims last night on throws further down the field, and he did a great job capitalizing on that.

Let’s turn now to the full advanced box score to find out more of what happened.

After turnovers and big plays negated GT’s efficient performance against UCF and Syracuse, the Jackets were able to combine another efficient performance with a big play advantage that resulted in a resounding total EPA margin in this game.

On offense, GT continued to show clear improvement over last year. While committing no turnovers, Jeff Sims produced 11.21 yards per called passing play (this includes passing plays that resulted in sacks and scrambles).. This is a tremendous performance, especially combined with his 80% success rate on called runs. Two other positives to not are Louisville’s ridiculously low 8% havoc rate to go along with an opportunity rate (the % of carries that gain at least 4 yards) of 51%. Jeff Sims and the offensive line both took meaningful steps forward Friday night.

When Louisville had the ball, GT was able to be disruptive enough to keep Louisville off schedule, stuffing 23% of Louisville running plays and adding 15% havoc plays. GT did give up 5 passing plays of more than 20 yards, but the defense was quite effective in limiting the intermediate passing game. The GT secondary was able to hold Malik Cunningham to 6.57 yards on called passing plays The defense put up an average performance against an excellent offense. That’s a good thing!

Let’s look more closely at the primary pass defenders, keeping in mind that negative numbers are good for CPOE on defense, telling us when primary defenders are giving up fewer completions than expected. Juanyeh Thomas continued his strong play, holding his assignments to 11% less CP than expected on three targets. His starting safety counterpart struggled again and ended up giving way to a breakout performance from Derrik Allen. The long awaited performance from the former Notre Dame transfer produced a CPOE of -37% on three targets. Look for 18 to see the field more in coming weeks. Tre Swilling struggled in his return from injury, and Kenan Johnson mightily struggled as his backup, while Wesley Walker had a very encouraging game.

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. Akshay Easwaran did a great job priming us for EPA this summer. Check out his column here for background.

The EPA totals for this game leave us with about a 16 point margin of victory for Gt. We’ll dig into the highlights and lowlights to help us see the particular plays that helped GT come back from an early deficit and run away from Louisville in the fourth quarter.

Five Most Helpful Plays

1. Jordan Domineck’s forced fumble when Louisville had 1st and 10 at their own 45, giving GT the ball at midfield to set up the go ahead score. 5.14 EPA
2. Bruce Jordan Swilling’s forced fumble on kickoff coverage to give GT the ball at the Louisville 27 in the third quarter. 5.09 EPA
3. Charlie Thomas’s sack and forced fumble on 3rd and 4 on the opening sequence of the game. 4.27 EPA
4. The 33 yard catch and run from Jeff Sims to Adonicas Sanders on 3rd and 10, moving the ball from the GT 34 all the way to the Louisville 18 after a targeting penalty. 3.95 EPA
5. Jeff Sims’s 22 yard touchdown run on 3rd and 5 to give GT its first lead. 3.60 EPA.

It sure is nice when the big plays and turnovers go your away. The defense created just enough havoc, and the offense created more chunk plays than earlier in the year. And the list below is going to include much less hurtful plays than in previous weeks.

Five Most Hurtful Plays

1. Malik Cunningham’s 36 yard pass to Fitzpatrick on 3rd and 10 to move from the GT 38 to the 2. 4.05 EPA
2. Malik Cunningham’s 11 yard touchdown run on third and goal to give Louisville the lead back in the third quarter. 3.26 EPA
3. Malik Cunningham’s 14 yard pass to Tutu Atwell on 4th and 6 from the GT 40. 2.85 EPA
4. Javian Hawkins 27 yard run on 3rd and 16 from the Louisville 9. 2.83 EPA.
5. Malik Cunningham’s 2 yard scramble to score on 4th and goal in the second quarter. 2.78 EPA
6. Right after the fourth down catch mentioned above, Tutu Atwell’s 26 yard touchdown catch from Malik Cunningham. 2.70 EPA

GT gave up explosive plays as we expected, but they were fewer in quantity and caused less damage than in previous weeks. Having no turnovers sure helps the EPA battle.

Tracking Season Goals

I set these goals for the 2020 season in some of my offseason preview work. We will be tracking them as we go this year.

Passing success continues to be massively improved and is probably the most encouraging development of this season so far. Interestingly, the offense had no power opportunities (3rd or 4th and 1 or 2). The defense made no stops on power opportunities and continues a season long struggle in those situations. The stuff rate (runs stoped for zero or negative yards) ticked up this week following defensive line improvement led by Antonneous Clayton. Finally, the havoc rate is still far below our goal, but the havoc plays were particularly impactful and timely in this game.

Takeaways

1. Jeff Sims ran the offense beautifully behind improved offensive line play. After giving up over 40 % pressure rate against Syracuse, the OL held Louisville to 25%, and Jeff Sims took full advantage in his best game to date as GT’s quarterback.
2. The defense held their own against an excellent offense that figures to finish top ten nationally. The personnel changes and improvements on the defensive line as the game progressed bode well for the next 7 games.
3. The mental mistakes that had ruined the previous few weeks were cleaned up. As we’ve said before, turnovers and penalties tend to even out over time, and Friday night we got to see what GT can do when not capsized by massively hurtful plays.

What a way to bounce back from two disappointing losses and a rough start to this game. The freshman under center was resilient and shined brightly as the game wore on. The freshman behind him in the backfield made two more brilliantly athletic plays that helped turn the game. Both lines stepped up significantly, and some new faces on the backend shored up struggles that the secondary had earlier in the game. This was a great performance from the players and staff alike. The present and future look brighter than they did two weeks ago. Let’s enjoy this one before we face the toughest test of the year on Saturday.