Part of getting involved with sports analytics is learning to rage against the small-sample narrative. It’s a thing in basically every sport: baseball (Batter A is 1-for-5 lifetime against Pitcher B — that’s a .200 average! Play somebody else!), basketball (Shooter C has made three of his last four 3-pointers — he’s got a hot hand!), etc.
It’s an issue that is custom made for football, however. Baseball features 162 games at the major league level and 50-plus in college. Basketball features 82 games in the NBA and 30-plus in college.
Football: 16 games in the NFL, 12 in FBS. Depending on the topic at hand, it could take years for you to develop what is truly a statistically significant sample. Therefore, I tend to reflexively push back on certain types of narratives. Your coach is 0-3 against top-five teams, therefore he can’t win the big one! Your quarterback was bad these two times in the fourth quarter, therefore he chokes in crunch time! And so on.
Help me out a little bit, Kevin Sumlin. For a couple of years, I casually glossed over the “fast start, poor finish” narrative that has begun to define your tenure as head coach at Texas A&M, but every time it happens, it gets harder to resist.
- In 2013, the Aggies had defending Heisman winner Johnny Manziel, began the season seventh in the country, and remained ninth in late November before losing back-to-back games to LSU and Missouri and narrowly getting by Duke in the bowl game.
- In 2014, they began the season 5-0 and rose to sixth in the country, then lost five of seven and finished unranked despite winning the Liberty Bowl.
- In 2015, they began 5-0 again and reached ninth. They then lost three of four, rallied briefly against Western Carolina and Vanderbilt, then lost to LSU and Louisville to finish 8-5.
- In 2016, they began 6-0 to change things up, reached sixth in the country, then lost five of seven once again.
I could point out that we’re really only talking about three years here (losing at Mizzou and LSU in 2013 was forgivable) and that most of the teams A&M tends to lose to late in the year are pretty good. They are in the SEC West, after all. In their back-to-back-to-back 8-5 seasons, they’ve suffered six losses to Alabama and LSU and five losses to other teams that won at least nine games that year. There’s not just a ton of shame in that.
A&M has, in terms of full-season averages, established a pretty solid level of success, ranking 26th or better in S&P+ in four of Sumlin’s five seasons; Mike Sherman and Dennis Franchione did that only three of nine times, and A&M hadn’t pulled such a four-in-five run since 1992-95 under R.C. Slocum. Just about anything you do for the first time in 20 years is noteworthy.
But the way things have unfolded has limited the potency of Sumlin’s accomplishments. Surprising successes, like what Sumlin experienced in 2012, are what contract extensions are made of. Teases are not, and because of how it shapes our memories, it’s better to be a slow starter than a slow finisher.
Last year was particularly frustrating. After dealing with roster turnover and constantly young depth charts, the Aggies had their first truly experienced team since 2012. That should have gone a long way toward proving staying power.
Instead, A&M not only lost to Alabama and LSU after the fast start but also fell on the road to a shaky Mississippi State, blew a late lead in a home loss to an Ole Miss giving a true freshman QB his first start, and fell to former conference rival Kansas State in the Texas Bowl. This was still the best Aggie team since Manziel left, but that only means so much when the record doesn’t reflect it.
Now, as tends to be the case following a year with an experienced team, there are some key pieces to replace, namely starting quarterback Trevor Knight (who provided a one-year cushion after A&M lost two blue-chippers to transfer), six of the top seven wide receivers, an all-conference left tackle (Avery Gennesy), and about half the two-deep in the front seven.
This isn’t a wholesale rebuild. The depth chart will still feature plenty of upperclassmen, and there are still stars like running back Trayveon Williams, receiver Christian Kirk, defensive tackle Zaycoven Henderson, and linebacker Otaro Alaka. But once you’ve lost the benefit of the doubt, it’s hard to get it back. One glance at the schedule shows that another 5-0 start and 3-4 finish are very much in play. S&P+ projections suggest seven wins are most likely; win your bowl, and Eight-Win Sumlin further becomes a thing.
From a booster perspective, not improving is the same as getting worse. Eventually, Sumlin either overcomes this plateau or gets overrun.
2016 in review
- First 6 games (6-0): Avg. percentile performance: 80% (73% offense, 61% defense) | Avg. yards per play: A&M 6.8, Opp 5.3 (plus-1.5) | Avg. performance vs. S&P+ projection: plus-11.4 PPG
- Last 7 games (2-5): Avg. percentile performance: 61% (63% offense, 57% defense | Avg. yards per play: A&M 5.9, Opp 5.8 (plus-0.1) | Avg. performance vs. S&P+ projection: minus-7.3 PPG
The shifts weren’t dramatic, but they were enough. The Aggies handled Auburn and Arkansas with relative ease in September, kept South Carolina at bay, and knocked Tennessee from the ranks of the unbeaten.
As was the case in 2015, however, a big loss to Alabama turned the season in the wrong direction. (They got blown out in 2014, too, but that year the Aggies had already begun to fade before the trip to Tuscaloosa.)
Injuries didn’t help; Knight stunk against Mississippi State and missed the Ole Miss and UTSA games. (To his credit, backup QB Jake Hubenak was strong against both Mississippi schools.) Still, aside from a little bit of shuffling on the offensive line, that was about it. Everybody deals with some injury issues, but A&M’s offense faded all the same.
Noel Mazzone’s first season as A&M offensive coordinator was fine. The Aggies got back about half of what they’d lost in 2015 — after three years in the Off. S&P+ top 15, they had plummeted to 55th that year, then they rebounded to 30th last fall.
They were extremely reliant on big plays, however. Unless you are Penn State, that’s a recipe for volatility — for, say, scoring 39 points on LSU, 29 on Auburn ... and 23 on UTSA. For all of his relative strengths, Knight was never prone to a high completion rate (he never hit 60 percent at Oklahoma), and that didn’t change in College Station. Meanwhile, short-yardage issues held back an otherwise sterling running game.
The big plays were huge; A&M had 26 gains of 40-plus yards, ninth in FBS. But efficiency issues prevented the Aggies from again having an elite offense.
So now it’s Hubenak’s turn then? The senior-to-be completed 61 percent of his passes and produced a passer rating of 165.3 against MSU and Ole Miss. At this point, he’s at least a little bit of a known quantity, a high-floor option. But Sumlin’s always thinking about the ceiling — he chose redshirt freshman Manziel over the older Jameill Showers back in 2012, after all — and it appears redshirt freshman Nick Starkel might currently have the edge in fall camp. Four-star true freshman Kellen Mond is in the race, too.
The prospect of starting a relative newbie with an almost completely new receiving corps is rather scary. The winner of the QB job will still have junior Christian Kirk at his disposal, and that’s a decent start. The former blue-chipper caught 83 passes last year, albeit at only 11.2 yards per catch with a 40 percent success rate.
The next leading returning wideout: Damion Ratley. He caught two of eight passes last year. Yikes.
Almost by necessity, youngsters like four-star true freshmen Jhamon Ausbon and Hezekiah Jones and redshirt freshmen Quartney Davis and Clyde Chriss will be in the rotation.
Continuity in the passing game bears a strong correlation with improvement or regression in Off. S&P+; it’s no wonder, then, that the Aggies’ offensive ratings are projected to fall.
The new QB will at least have Trayveon Williams on his side. That’s something. Williams was a revelation as a freshman last season, rushing for 1,057 yards at 6.8 yards per carry; his efficiency was decent for a newbie (he gained at least five yards on 41 percent of his carries, right at the national average), and his explosiveness was off the charts. Among the 109 FBS backs with at least 150 carries, his average of 8 highlight yards per opportunity was sixth behind only LSU’s Derrius Guice, USF’s Marlon Mack, UTEP’s Aaron Jones, Oklahoma’s Joe Mixon, and Penn State’s Saquon Barkley. Among power conference freshmen in this group, Missouri’s Damarea Crockett was the only other player to top 6 highlight yards
Ford has a lovely complement in senior Keith Ford; the 215-pounder was more efficient if less explosive. And either sophomore Kendall Bussey or junior Kwame Etwi could provide a lovely third-string option.
They’ll be running behind a line that is simultaneously young and experienced; of the seven players who started at least two games on the line, five return, including full-time starters Colton Prater and Erik McCoy. But Prater and McCoy are both mere sophomores, and the three other returnees are juniors. This should be a decent line in 2017 and a potentially fantastic one in 2018.
Thus far, John Chavis’ tenure as A&M defensive coordinator has been fine. After two years in the wilderness — the Aggies ranked 87th in Def. S&P+ in 2013 and 63rd in 2014 — Chavis came to town and restored some semblance of stability. A&M ranked 27th in 2015 and 36th last fall. Not good enough considering the recruiting rankings, but it’s something.
A&M was really good at making what you might call key plays last year. The Aggies were eighth in power success rate, sixth in stuff rate, 21st in passing downs sack rate, and, perhaps most importantly, 12th in points per scoring opportunity allowed. They were capable of making the plays they absolutely needed to make, and it helped to keep non-Bama opponents within reach even when the offense wasn’t clicking.
(The Aggies got run over by LSU, but that’s sort of an annual tradition.)
The problem was that A&M wasn’t really big on making the mundane plays. Despite high rankings in all of those categories, they were just 57th in success rate allowed, and they ranked 53rd in opportunity rate, allowing at least five yards on 38 percent of carries. They also gave up 70 gains of 20-plus yards, 107th in the country.
That’s a lot of good and a lot of bad.
In theory, both the good and bad get tamped down in 2017.
A&M boasts a senior-heavy secondary that includes a steady cornerback (Priest Willis) who defensed nine passes last year, plus two safeties (Donovan Wilson and Armani Watts) who had at least 5.5 tackles for loss. Granted, this group hasn’t necessarily proved itself from the standpoint of big-play prevention, but this level of DB experience tends to correlate well with that.
Steadiness in the back might help to offset a downgrade in play-making up front. To be sure, few teams are more loaded at defensive tackle than the Aggies — Henderson and Kingsley Keke combined for 17 TFLs, seven sacks, and four pass breakups last year, former blue-chipper Daylon Mack returns, and blue-chip redshirt freshman Justin Madubuike has massive potential. But it’s really hard to lose players like ends Myles Garrett and Daeshon Hall and linebacker Claude George (combined: 39.5 TFLs, 16.5 sacks) without a drop-off in havoc levels.
Jarrett Johnson and Alaka should help, though. Johnson recorded 6.5 TFLs and 4.5 sacks among his 15 tackles as a backup end last year, and Alaka was maybe the best run-stuffer on the team, with seven non-sack TFLs.
Depth is perilous at both end and linebacker. After Johnson, the only two returnees who saw the field are Qualen Cunningham and Landis Durham (combined: 14.5 tackles, two TFLs). After Alaka and sophomore Tyrel Dodson at linebacker, you’ve got unproven juniors and true freshmen.
As always, you never know if or when that will backfire on a team. If the Aggies suffer a couple of injuries at tackle, they could survive. An injury to Johnson or Alaka, and they’re playing nothing but unproven entities.
Special teams have been a constant bright spot for A&M; I have Special Teams S&P+ data going back three years, and the Aggies have ranked 32nd or better all three seasons. Granted, it’s been a downward trend — first in 2014, 29th in 2015, then 32nd — but every aspect of this A&M last year was either decent (place-kicking, kick returns) or awesome (punting, kickoffs, punt returns).
That everybody but kick returner Justin Evans is back, then, is unquestionably good. Kicker Daniel LaCamera needs to get a little more consistent on his shorter field goals (you want to make 85 percent or more of your under-40 FGs, and he was at 77 percent), but the fact that he was 4-for-7 beyond 40 suggests it’s possible. Meanwhile, Braden Mann is a touchback machine in kickoffs, and Shane Tripucka averaged 42.9 yards per punt with minimal returns.
Oh yeah, and Christian Kirk might be the best punt returner in the country. The Aggies’ special teams unit is one giant field position weapon.
2017 Schedule & Projection Factors
|Date||Opponent||Proj. S&P+ Rk||Proj. Margin||Win Probability|
|18-Nov||at Ole Miss||26||0.1||50%|
|Projected S&P+ Rk||19|
|Proj. Off. / Def. Rk||33 / 24|
|Five-Year S&P+ Rk||14.1 (11)|
|2- and 5-Year Recruiting Rk||12 / 9|
|2016 TO Margin / Adj. TO Margin*||3 / 4.9|
|2016 TO Luck/Game||-0.7|
|Returning Production (Off. / Def.)||52% (39%, 64%)|
|2016 Second-order wins (difference)||8.8 (-0.8)|
Things could clearly be worse in College Station. Sumlin has won 44 games in five years — A&M had 46 wins in seven years before his arrival — and after three straight years of regression from 2013-15, the Aggies improved overall last fall.
Things could also be much, much better. Sumlin set the bar impossibly high in his 11-win, 2012 debut, and his Aggies have made a habit out of teasing with top-10 level play before slumping to 8-5 in each of the last three years. Their biggest problem under both Franchione and Sherman was that they were almost always solid and never great; after the initial surge, they’ve basically been exactly that under Sumlin, too. Granted, they’re slightly better in terms of both recruiting and performance, but the phenomenon is the same.
Sumlin needs a big year, and his Aggies will always have the talent to provide it. But they haven’t had staying power, and from an S&P+ perspective, that isn’t likely to change. S&P+ projects them 19th overall, but that’s only good enough for about a 7-5 record. The schedule features only a couple of likely losses (Alabama, at LSU) and four likely wins; if they find an edge in the relative tossups, they could threaten nine or 10 wins. But with a completely rebuilt passing game and pass rush, it’s hard to imagine that happening.
So A&M will be what A&M usually is: an explosive, athletic, high-ceiling team with depth issues and a relatively low floor. There are worse fates in the world. Sumlin’s problem is that there are also lots of better fates.