clock menu more-arrow no yes

Filed under:

Option Advanced Stats Report, Week 7: Stuffs Happened

New, 2 comments

Lots and lots of them

NCAA Football: Duke at Georgia Tech Brett Davis-USA TODAY Sports

Author’s note: Nishant here—I’m subbing in to handle the write-up this week. As always, though, it will draw upon Andrew’s play charting and the observations that Kieffer and Andrew have made during film review. Also, I realized right at the very end that this isn’t in power rankings format, but whatever.


This week’s report opens with a groundbreaking analytics-fueled discovery: Tech did not run the ball very well against Duke.

Shocking, right?

Matt broke down several aspects of Tech’s struggles in the Option Strategy Report, including Duke’s sound discipline in the front seven and Tech’s assortment of blocking miscues. It’s all reflected in the advanced stats breakdown at both the playcall level and the player level, and it is perhaps most evident in this week’s stuff rates.

First, a brief refresher on the terminology:

  • Success Rate: the percentage of plays that gain 50% of the required yards on first down, 80% of the required yards on second down, or 100% of the required yards on third/fourth down
  • Stuff Rate: the percentage of run plays where the ballcarrier is stopped for no gain or a loss
  • Opportunity Rate: the percentage of run plays that go for more than four yards
  • Highlight Yards: on “opportunity” plays, a weighted tally of the extra yards gained where yards 5-10 count as half a yard each and everything beyond that counts as a full yard; for example, a 3-yard run = zero highlight yards, a 7-yard run = 1.5 highlight yards, a 10-yard run = 3 highlight yards, and a 16-yard run = 9 highlight yards

With that out of the way, let’s dive in.

Data by Play Type

Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yds/Opp
Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yds/Opp
Triple 15 5.2 40.0% 13.3% 46.7% 3.6
Zone Dive 5 5.4 60.0% 20.0% 60.0% 2.5
Midline 1 1.0 0.0% 0.0% 0.0% 0
Counter Option 1 -1.0 0.0% 100.0% 0.0% 0
Counter Speed 1 -1.0 0.0% 100.0% 0.0% 0
Trap 1 4.0 0.0% 0.0% 0.0% 0
Belly 4 2.3 50.0% 0.0% 0.0% 0
Belly Option 7 5.4 42.9% 28.6% 57.1% 4.4
Toss 5 4.2 80.0% 20.0% 40.0% 1.5
QB Follow 4 2.5 25.0% 25.0% 25.0% 0.5
QB Draw 2 7.0 100.0% 0.0% 100.0% 1.5
Speed 9 4.0 66.7% 22.2% 44.4% 1.8

Tech’s offense came back down to Earth in a brutal way. The super-high success rates and heaps of highlight yards from the Louisville game seem like forever ago. Against Duke, the only plays that had anything resembling consistent success were rocket toss and speed option, and even those were far from perfect—the Jackets ran a toss to Clinton Lynch on third and 1 on their second possession, and Duke stopped it for a loss to force a punt.

The ugliest stats are the stuff rates. Belly (handoff, not option) was the only play that Tech ran multiple times that wasn’t stuffed at least once. All in all, 11 of Tech’s 55 carries went for no gain or a loss. In other words, every fifth carry was stuffed. For an offense that relies on getting at least three-ish yards on every play to move the ball, that’s a death sentence. (There were also five more carries that went for just one yard, which isn’t much better... plus three sacks and a botched center-QB exchange that were left out of the stuff rate calculations.)

One bizarre development was the complete disappearance of counter and trap plays a week after they were wildly successful. Tech ran six counter options and six traps against Louisville and averaged nearly nine yards per carry on those plays; against Duke, they ran just one of each, plus one more counter speed option. It’s possible that Paul Johnson simply looked at the poor blocking execution on those plays and kicked them to the curb, but it is still weird; trap was called to start Tech’s second drive and went for a four-yard gain, and then it was never seen again.

Data by Play Type: Left-Side Runs

Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yds/Opp
Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yds/Opp
Triple 4 3.8 25.0% 25.0% 25.0% 2
Zone Dive 1 7.0 100.0% 0.0% 100.0% 1.5
Midline 0 0.0 0.0% 0.0% 0.0% 0
Counter Option 1 -1.0 0.0% 100.0% 0.0% 0
Counter Speed 1 -1.0 0.0% 100.0% 0.0% 0
Trap 0 0.0 0.0% 0.0% 0.0% 0
Belly 1 4.0 0.0% 0.0% 0.0% 0
Belly Option 4 8.8 50.0% 25.0% 75.0% 5.7
Toss 2 3.5 50.0% 50.0% 50.0% 2
QB Follow 2 2.5 0.0% 0.0% 0.0% 0
QB Draw 0 0.0 0.0% 0.0% 0.0% 0
Speed 5 5.6 80.0% 0.0% 60.0% 1.7

Data by Play Type: Right-Side Runs

Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yds/Opp
Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yds/Opp
Triple 11 5.7 45.5% 9.1% 54.5% 3.9
Zone Dive 4 5.0 50.0% 25.0% 50.0% 3
Midline 1 1.0 0.0% 0.0% 0.0% 0
Counter Option 0 0.0 0.0% 0.0% 0.0% 0
Counter Speed 0 0.0 0.0% 0.0% 0.0% 0
Trap 1 4.0 0.0% 0.0% 0.0% 0
Belly 3 1.7 66.7% 0.0% 0.0% 0
Belly Option 3 1.0 33.3% 33.3% 33.3% 0.5
Toss 3 4.7 100.0% 0.0% 33.3% 1
QB Follow 2 2.5 50.0% 50.0% 50.0% 0.5
QB Draw 2 7.0 100.0% 0.0% 100.0% 1.5
Speed 4 2.0 50.0% 50.0% 25.0% 2

Tech ran triple/veer option to the left much more frequently and had more success running it to that side. It makes sense—going left means running behind left tackle Zach Quinney and left guard Parker Braun, who have been the team’s most effective linemen this season. That said, the very uneven split seems to be a new development; past games have seen a more balanced mix of left vs. right for triple option runs.

Conversely, Tech had more success on right-side speed option and belly option runs, particularly the latter. Belly option requires the playside guard to pull, so right guard Connor Hansen deserves credit for his role in those plays.

Data by Option Read: Dive

Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yards Per Opp
Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yards Per Opp
Triple 11 3.8 36.4% 9.1% 45.5% 0.9
Midline 1 1.0 0.0% 0.0% 0.0% 0

Data by Option Read: Keep

Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yards Per Opp
Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yards Per Opp
Triple 3 10.7 66.7% 33.3% 66.7% 10.5
Counter Speed 1 -1.0 0.0% 100.0% 0.0% 0
Belly Option 4 0.8 0.0% 50.0% 25.0% 1
Speed 7 2.7 57.1% 28.6% 28.6% 1.3

Data by Option Read: Pitch

Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yards Per Opp
Play Count YPC Success Rate Stuff Rate Opportunity Rate Highlight Yards Per Opp
Triple 1 4.0 0.0% 0.0% 0.0% 0
Counter Option 1 -1.0 0.0% 100.0% 0.0% 0
Belly Option 2 8.0 100.0% 0.0% 100.0% 2.3
Speed 2 8.5 100.0% 0.0% 100.0% 2.3

Breaking things down by read points to some things that Duke did extremely well. First and foremost, they were able to contain the dive. B-back Jerry Howard had some success overall—he averaged over five yards per carry for the game—but on option plays that resulted in a handoff, he and Jordan Mason struggled to gain much traction. They combined for just 4.5 highlight yards for the game on option dives, which is not exactly a path to success.

The biggest play of the day on the ground was a 26-yard run by quarterback TaQuon Marshall early in the third quarter on a triple option keeper. Aside from that run, option keepers were mostly a train wreck. On paper, four of the seven speed option keepers were successful; in practice, three of the four came only after Duke had put the game away by opening up a 28-7 lead.

Tech was reasonably successful on option pitches, particularly on belly and speed options; however, as the small number of carries would suggest, there weren’t many pitches. Duke did an excellent job of taking away the pitch man and forcing Marshall to keep, then quickly containing him for a loss or short gain.

Data by Player

Play Carries YPC Success Rate Stuff Rate Opportunity Rate Highlight Yds/Opp Highlight Yards (Game) Highlight Yards (Season vs. FBS)
Play Carries YPC Success Rate Stuff Rate Opportunity Rate Highlight Yds/Opp Highlight Yards (Game) Highlight Yards (Season vs. FBS)
Marshall 15 3.3 26.7% 46.7% 40.0% 4.1 24.5 248.5
Oliver 8 4.6 75.0% 12.5% 50.0% 1.6 6.5 220.0
Mason 12 3.3 25.0% 16.7% 25.0% 1.8 5.5 173.5
Searcy 3 5.0 66.7% 33.3% 66.7% 2.3 4.5 104.5
Howard 12 5.1 66.7% 0.0% 58.3% 1.6 11.0 78.0
Lynch 5 6.4 80.0% 20.0% 40.0% 6.5 13.0 33.0
Cottrell 2 6.0 50.0% 0.0% 50.0% 2.0 2.0 33.0
Graham 1 4.0 0.0% 0.0% 0.0% 0.0 0.0 0.0

Marshall had a team-high 24.5 highlight yards and has slightly increased team lead in that stat for the season. It’s not exactly much to celebrate, though. For reference, last week Marshall had 81.5 highlight yards and Tobias Oliver had 65.5, and A-back Omahri Jarrett picked up 22 on his only carry of the game.

A-back Clinton Lynch had a few decent runs on the perimeter and picked up 13 highlight yards, bringing him up to a tie with Nate Cottrell for the season. Both of them are far behind the rest of the major ballcarriers, though, which simply speaks to the trouble that Tech has had in springing the A-backs for big runs against not-terrible defenses.

True freshman quarterback James Graham took his first snap in a Tech uniform, gaining four yards on a simple QB follow. Under the new NCAA redshirt rule, Graham can play in up to three more games and still maintain his redshirt, but it’s unlikely that he’ll play again this season (barring injuries or a combination of events like the one that got him on the field in this game). Not much can be gleaned from one carry that didn’t involve a read, but nevertheless, the former four-star recruit will be an intriguing prospect in next year’s QB competition.

Conclusion

The Louisville game showed what Tech can do against an overmatched ACC defense that plays without confidence or discipline. Duke brought both of those to the table in spades, and the across-the-board improvements that Tech had shown over the past two games went out the window.

Missed assignments were common to a disturbing extent. Rather than struggling with pitch reads, Marshall often didn’t even have a read to make, because Duke was frequently able to stop plays in the backfield before they could go anywhere useful. Big plays on the ground were more or less nonexistent as the team struggled to string together a few successful plays in a row. Oliver did what he could to rally the team in the fourth quarter after Marshall’s injury, but overcoming a 21-point deficit was a bit much to ask of the redshirt freshman.

To become bowl-eligible, Tech needs to win three games over a final stretch that includes division leader Virginia Tech, talent-rich Miami, resurgent Virginia, and powerhouse UGA. None of those are easy wins... and honestly, if the Jackets can’t fix the fundamental issues that sprung up left and right in this contest, then even beating hapless UNC may not be a given.