clock menu more-arrow no yes

Filed under:

Paul Johnson’s Georgia Tech: even less predictable than usual in 2017

New, comments

Buckle up, Yellow Jackets.

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

This preview originally published July 17 and has since been updated.

Johnson won the ACC less than a decade ago — and the Orange Bowl not even three seasons ago — while running the triple option. That’s a wonderful sentence to type. For all of us who love college football because of its countless paths toward winning, Tech has served as a go-to example.

Johnson lives in a recruiting hotbed but barely signs top-50 classes, and he’s won nine or more games four times in the last nine seasons regardless. And in an era with increasingly proficient passing quarterbacks, he asks his to spend most of their time reading whether to stick the ball in a fullback’s belly or run with it. That is delightfully off-kilter.

When Johnson left Navy in 2008, the Midshipmen promoted offensive coordinator Ken NIumatalolo, and Niu has responded with some of the steadiest play in college football. Over nine seasons, Navy has won either eight or nine games six times, winning 10-plus twice and dipping to 5-7 just one time.

Johnson has had similar success, but with none of the stability. He has averaged a healthy 7.9 wins per year, but the ride has turned into a roller coaster:

  • 2012: 7-7 and 28th in S&P+
  • 2013: 7-6 and 41st
  • 2014: 11-3 and 16th
  • 2015: 3-9 and 60th
  • 2016: 9-4 and 45th

Tech finished eighth in the AP poll in 2014, its best finish since splitting the national title with Colorado in 1990. The Yellow Jackets began 2015 ranked 16th but succumbed to massive youth and injury issues. In 2016 — with no expectations, a painfully young skill corps, and a banged-up line — they surged back to nine wins.

Like a good option quarterback, Tech has kept expectations flat-footed. But here’s what we know:

  • The skill corps finally has some experience again. Fullback Dedrick Mills is back [Update: and he’s now dismissed, via a violation of team rules, per GT] after leading Tech in rushing as a freshman, and the top three slotbacks and both starting receivers are back. There was major turnover in the play-maker department following 2014’s Orange Bowl, and the depth chart appears to have stabilized.
  • The offensive line has all sorts of experienced pieces after a disastrous 2016. Nine players started at least one game, and no one started all 13. The result: three “starters” are gone, and six “starters” return, including monstrous, 380-pound guard Shamire Devine.
  • This is Johnson’s most experienced defense in a while. The entire lineup could consist of juniors and seniors, and the entire secondary returns, but we’ll see what that means for a unit that has been neither strength nor weakness. Tech has ranked between 45th and 66th in Def. S&P+ in each of the past five seasons.
  • Thomas is gone. The longtime QB threw for 4,750 yards and rushed for 2,400 during his career, and his presumed successor had to undergo foot surgery midway through spring practice.

Everything falls into place ... just in time for uncertainty at quarterback. That won’t make the Yellow Jackets any more projectable than they have been over the last three or four years.

This is a big year for Johnson. Defensive end KeShun Freeman talked at ACC Media Days about the need to “break this cycle” of ups and downs. Plus, new athletic director Todd Stansbury has been in place for less than a year, and Tech fans have long been frustrated with recruiting.

It was easy to assume a “beginning of the end” narrative after 2015.

Now what? We can see whatever we want to see from Johnson and Tech heading into 2016. If you want to say that this is the beginning of the end for Johnson, that the talent departed from the 2014 squad was of a higher caliber than its 2015 replacements, then you haven't been proven wrong yet. If you choose to believe the defense is never going to improve enough to offset whatever talent drain has taken place on the other side of the ball, you might be right.

Despite flaws, the Jackets responded by winning close games again — they were 3-1 in one-possession finishes in 2016 after going 5-15 over the previous four seasons — and finishing on a four-game winning streak.

What that means for 2017, I have no idea. But my opinion shouldn’t carry much weight; I’ve been a step slow in my Tech expectations for a while now.


2016 in review

2016 Georgia Tech statistical profile.

What happens when you’ve got a veteran quarterback surrounded by inexperience? High upside and drastic inconsistency. Over the first seven games of 2016, Georgia Tech averaged the following yards per play: 4, 7.2, 8.1, 2.4, 5, 6.8, and 8. The Jackets pummeled a decent Vanderbilt, then lost three in a row. They lost by 28 to North Carolina, then beat Virginia Tech and Georgia. They went 4-4 in the ACC Coastal and 3-0 in the SEC East.

The offense did stabilize ... just in time for the defense to lose its way.

  • First 6 games (3-3): Avg. percentile performance: 55% (55% offense, 52% defense) | Avg. yards per play: Opp 5.7, GT 5.6 (minus-0.1)
  • Next 6 games (5-1): Avg. percentile performance: 63% (76% offense, 40% defense) | Avg. yards per play: GT 7.6, Opp 6.0 (plus-1.6)

The offense improved more than the defense regressed, and that worked well — aside from a 48-20 loss to UNC in which the Jackets allowed 9.1 yards per play, anyway.

If quarterback is stable in 2017, the offense could have enough punch to survive the defense’s inefficiency. But if Jordan is unready, the defense might have to force the issue.


Offense

Georgia Tech offensive radar

Full advanced stats glossary.

The basics of the flexbone option can be seen in the radar chart above. Tech was strong on standard downs, avoided losses in the run game, and created lots of big plays, in part because the play-action bomb can be quite effective in an option system.

That’s been the look of the Tech offense for a while, but the degrees of success change. Tech had this profile in 2014 and ranked fourth in Off. S&P+; the last two years, the Ramblin’ Wreck have ranked 56th and 44th.

The shifts have been subtle, but the changes from 2014 to 2016 were significant.

Georgia Tech offense trends and percentages

Category 2014 2015 2016
Category 2014 2015 2016
% of carries by QB 29% 34% 34%
% of carries by fullback 46% 42% 43%
% of carries by slotback 25% 24% 23%
Std. downs run rate 84% 82% 85%
Pass. Downs run rate 57% 49% 57%
Completion rate 52% 42% 52%
Yards per completion 17.8 18.0 20.2
Success Rate ranking 1 58 58
IsoPPP ranking 44 114 3

After drastic turnover in the skill corps, Thomas began to carry a bit more of the load. In 2014, he had the effective fullback duo of Zach Laskey and Synjyn Days, but in 2015 everybody got hurt, and in 2016, freshman Mills became the go-to up the gut.

For Tech, fullbacks drive everything that follows. Not including sacks, Thomas went from averaging 6.4 yards per carry in 2014 to 4.4 in 2015 and 5.6 in 2016; meanwhile, the A-backs (slotbacks) averaged 8 in 2014, 7.7 in 2015, and 7.4 in 2016.

In two years, Tech went from being ruthlessly efficient to reliant on big plays. Those big plays were damn big, and a rebound in the passing game — one that could continue with the return of receivers Ricky Jeune and Brad Stewart (combined: 44 catches, 809 yards) — made a difference, but the key for an option offense is efficiency. Tech lost the plot.

NCAA Football: Georgia Tech at North Carolina
Matthew Jordan (11) and Dedrick Mills (26)
James Guillory-USA TODAY Sports

Be it Jordan, fellow junior TaQuan Marshall, redshirt freshmen Jay Jones or Lucas Johnson, or true freshman Tobias Oliver, the goal is finding a steady signal caller. With Thomas injured, Jordan led a road upset of Virginia Tech; he was 2-for-7 passing with a sack but ground out 127 rushing yards, and the fullback was the key: backup Marcus Marshall carried 19 times for 143 yards, which meant neither the slot backs nor passing game were much of a necessity.

Mills managed at least five yards on 43 percent of his carries, easily the best among primary ball-carriers. With Mills and Marshall gone, the new guy — likely either of two sophomores, KirVonte Benson or Quaide Weimerskirch — is an unknown.

There are fewer concerns at A-back. Qua Searcy, Clinton Lynch, and J.J. Green are back after combining for 842 rushing yards, 713 receiving yards, and 11 touchdowns. Searcy got the most touches, but Lynch’s per-touch figures were astounding: 58 combined rushes and pass targets, 905 yards, eight scores. During Tech’s season-ending streak, he carried nine times for 117 yards and caught four passes for 101. You could make the case that he’s the best player on the 2017 offense.

NCAA Football: Georgia Tech at Georgia
Clinton Lynch
Dale Zanine-USA TODAY Sports

The line should be fine. Despite constant turnover, Tech ranked 12th in power success rate and 35th in stuff rate, and six players who combined for 38 of last year’s 65 total starts return. Third-team all-conference center Freddie Burden is gone, but Devine, junior Will Bryan, and sophomore Parker Braun should assure Tech’s interior is strong. And if the new fullback is healthy and productive, that should drive the rest of the offense.

Defense

Georgia Tech defensive radar

The Tech defense was the same as it’s been for a while: a passive, bend-don’t-break unit that prevents big plays and does just enough in terms of red zone defense and third-down defense to get off the field before allowing a touchdown.

Tech will give you a five- or 10-yard gain in the name of preventing a 20-yard gain. The Jackets gave up more frequently successful plays than any of their ACC brethren, but the successful plays were smaller than anyone else’s, too.

Georgia Tech defensive efficiency & explosiveness

This works if you’re in a conference with young offenses. If you wait for a mistake, it can pay off. But in a conference full of increasingly competent teams, that might not be enough. The ACC boasted four of last year’s top 10 offenses per Off. S&P+, and a lot of defenses are taking the fight to opponents and trying to force mistakes.

It’s not good to rank 123rd in havoc rate, in other words. In four years under coordinator Ted Roof, Tech has avoided bottoming out, like in 2010 (82nd in Def. S&P+ and 2011 (84th), but hasn’t been good.

If that were to change, it’d probably change in 2017. The Jackets have to replace four decent pieces in the front seven (end Rod Rook-Chungong, tackle Patrick Gamble, and linebackers P.J. Davis and Chase Alford), but the secondary returns intact, and the defense still returns seven players who made at least 13.5 tackles in 2016.

Gamble did have 7.5 sacks, but of the departures, he’s the only one who was particularly disruptive.

NCAA Football: TaxSlayer Bowl-Kentucky vs Georgia Tech
Antonio Simmons
Logan Bowles-USA TODAY Sports

Tech’s line was weird. A tackle was its only pass rusher, and two not-large ends (Freeman and Antonio Simmons, neither of whom is listed over 250 pounds) had almost no pass-rushing presence (1.5 sacks) but did well against the run (7.5 non-sack TFLs). I think there’s a lot of potential among some younger tackles — junior Kyle Cerge-Henderson and sophomores Brandon Adams and Brentavious Glanton — and junior end Anree Saint-Amour is now the team’s leading pass rusher after recording four among 15.5 tackles.

Maybe Adams’ girth (329 pounds) and Saint-Amour’s pass rushing ability mean that the line ends up playing like a normal line.

NCAA Football: Virginia at Georgia Tech
Lance Austin
Brett Davis-USA TODAY Sports

The return of middle linebacker Brant Mitchell and OLBs Terrell Lewis and Victor Alexander gives Tech some decent second-level experience, but the strength of the Tech defense should be in the back. The Jackets ranked 75th in Passing S&P+, thanks mostly to bend-don’t-break tendencies. They were 78th in allowing 115 passes of 10-plus yards, but only 12 of those gained 30-plus (fifth).

Safeties Corey Griffin, A.J. Gray, and Lawrence Austin combined for 10.5 TFLs and 10 passes defensed, and while shorter passes were usually open, returning senior corners Lance Austin and Step Durham still combined for 21 passes defensed. If Saint-Amour continues to develop as a pass rusher (and gets some help), Austin, Durham, and company could have an opportunity to make even more plays.

It doesn’t feel like this defense has a high ceiling. But the Jackets do prevent big plays, and they do have the experience to play the best version of this defense. If they can creep into the 30s in Def. S&P+, that could make a world of difference.


Special Teams

The offense only improved a little, and the defense didn’t really improve at all. So how did Tech go from three to nine wins and flip a bunch of close games? Turnovers luck (plus-2.7 points per game) helped, but so did special teams improvement. Harrison Butker was a big-time place-kicker, and three-quarters of his kickoffs resulted in touchbacks. Plus, Brad Stewart’s 11.5-yard punt return average was upper-tier.

Stewart’s back, and while J.J. Green was inefficient in kick returns, he counters that with some explosiveness. But the loss of both Butker and punter Ryan Rodwell hurts. It will be hard to avoid regression.


2017 outlook

2017 Schedule & Projection Factors

Date Opponent Proj. S&P+ Rk Proj. Margin Win Probability
4-Sep vs. Tennessee 24 -2.6 44%
9-Sep Jacksonville State NR 25.7 93%
16-Sep at Central Florida 78 8.7 69%
23-Sep Pittsburgh 33 3.3 58%
30-Sep North Carolina 38 4.6 60%
12-Oct at Miami 18 -7.4 33%
21-Oct Wake Forest 64 8.7 69%
28-Oct at Clemson 6 -16.2 18%
4-Nov at Virginia 70 6.5 65%
11-Nov Virginia Tech 25 0.1 50%
18-Nov at Duke 65 3.7 58%
25-Nov Georgia 20 -2.2 45%
Projected S&P+ Rk 31
Proj. Off. / Def. Rk 41 / 37
Projected wins 6.6
Five-Year S&P+ Rk 8.6 (33)
2- and 5-Year Recruiting Rk 54 / 51
2016 TO Margin / Adj. TO Margin* 4 / -3.1
2016 TO Luck/Game +2.7
Returning Production (Off. / Def.) 72% (62%, 82%)
2016 Second-order wins (difference) 7.4 (1.6)

Tech has been an unprojectable team. The Jackets have done poorly when projected well and vice versa.

They’re projected reasonably well this year. S&P+ says they should be about 31st, and if that’s the case, it sets up a fascinating schedule. S&P+ sees one sure win (Jacksonville State) and one nearly sure loss (at Clemson) and a whole lot of who-the-hell-knows in between. S&P+ says Tech has between a 44 and 60 percent chance of winning in six games and between 33 and 69 percent in 10. That is a remarkable schedule, one that could result in ... anywhere from three to nine wins.

That makes the narrative potential awesome. If Tech is top-50 or so, maybe Johnson gets fired. But with a quarterback, a little more disruption, and a couple of breaks, 10 wins — the most anti-social, Johnson-like result on the board — is also possible. Either result will be pretty entertaining, if I’m being honest.

Team preview stats

All power conference preview data to date.