Sonny Dykes’ SMU offense has more questions than his defense? What?

Bill C’s annual preview series of every FBS team in college football continues. Catch up here!

It’s comforting to have constants in life. College football offers plenty of them, from North Dakota State winning FCS, to Indiana suffering gut-wrenching losses against good teams.

Sonny Dykes is another. Whereas certain Sons of Air Raid have evolved a bit — Lincoln Riley runs the ball a lot! Dana Holgorsen and Riley don’t really use tempo! — Dykes still leans into old-school air raid principles. He wants to throw the ball, he wants to create space and solo tackles, he wants to move quickly.

He’s also probably going to have an offense a few steps ahead of his defense. That’s been the case in five of his seven years as a head coach (three at Louisiana Tech, three at Cal, one at SMU). It is a sign of, and in some cases a result of, his own personal preferences. You can field a good defense with an up-tempo, pass-happy, air raid system, but it takes some finesse.

One other thing we know about Dykes by now: his team is going to be streaky as hell. At Tech, he went 1-4, then 4-2, then 1-5, then 16-2, then 0-2. At Cal, 1-11, then 4-1, then 1-6, then 5-0, then 1-5, then 6-3, then 0-4. In his first season at SMU, his Mustangs went 0-3, then streaked to the precipice of bowl eligibility by winning five of seven. They then lost to Memphis (forgivable) and Tulsa (less so) to come up just short.

Indeed, his first SMU season appeared in many ways like a continuation of the rest of his head coaching career. There was one primary exception: his offense stunk.

Defensive coordinator Kevin Kane — a rising coaching star, if ever one existed — actually engineered the second-best defense Dykes has ever had. The Mustangs ranked 79th in Def. S&P+, inferior to nothing he produced at Cal and second only to the 2011 Louisiana Tech defense, which randomly soared to 44th. And they pulled this off with almost no seniors.

Kane improved NIU’s defense by 65 spots, from 91st to 26th, in 2017, then improved SMU by 34 last year. Theoretically, there’s nothing stopping him from engineering something similar this fall. But for basically the first time ever, a Dykes offense has to figure out how to catch up to a Dykes defense.

The national average for yards per play is around 5.8. SMU topped that mark just four times all season, and two were against a really bad FCS team (Houston Baptist) and maybe the worst FBS defense of the 21st century (UConn). They hit 6.0 while getting torched by North Texas, and they hit 6.1 in a great overall performance against Houston. They averaged just 5.4 for the season and scored more than 31 points just three times.

Dykes inherited from predecessor Chad Morris a set of personnel that understood tempo and general spread concepts. It’s what made the Dykes hire seem so seamless to begin with. The receiving corps needed rebuilding, and the run game was surprisingly horrendous.

Granted, the offensive line was a pile of wreckage. Nine players started at least two games, and only two started all 12. Freshmen and sophomores accounted for 29 of 60 starts. There were two particularly experienced pieces up front (left tackle Chad Pursley and left guard Nick Natour), and they missed nine games between them.

So maybe the offensive struggles were just a product of circumstance — quarterbacks learning a new system, a new receiving corps, a banged-up and sieve-like offensive line. Two of those three things could rectify themselves, and while quarterback Ben Hicks left to re-join Morris at Arkansas, Dykes brought in Texas veteran Shane Buechele in case sophomore William Brown wasn’t ready.

No matter what, though, the offense has far more questions to answer than the defense. This is rarefied air, very much a non-constant, for a Dykes team.

In my 2018 SMU preview, I figured Dykes and new offensive coordinator Rhett Lashlee, a Gus Malzahn disciple and former OC at Arkansas State, Auburn, and UConn, would have a constant they did not end up having: a run game.

In 2017, SMU ranked 26th in Rushing S&P+, 84th in opportunity rate (percentage of carries gaining at least four yards), and 46th in stuff rate (run stops at or behind the line). They mixed solid big-play ability (21st in run explosiveness) with steadiness and minimal negative plays.

In 2018, the Mustangs ranked 117th in Rushing S&P+, 125th in opportunity rate and 129th in stuff rate. They rose to 10th in run explosiveness, but my explosiveness ratings look at the magnitude of your successful plays. If you don’t have hardly any successful plays, explosiveness doesn’t help you all that much.

Obviously my preview didn’t take the revolving door of linemen into account. In-season continuity matters more than perhaps anything else for a line, and SMU had none. Maybe that will change this fall, but line coach A.J. Ricker’s got some patching to do this offseason, though.

Pursley and Natour are gone, as are starting right guard Larry Hughes and two other experienced pieces. In their place are four sophomores and juniors with starting experience, anchored by 12-game starting center Hayden Howerton and sophomore and former star recruit Alan Ali. Columbia grad transfer Charlie Flores could carve out a role sooner than later, as could JUCO transfer Cobe (with a C) Bryant.

Until the line, and, consequently, the run game, are fixed, the passing game almost doesn’t matter, and that’s a really strange thing to say about a Dykes offense.

Xavier Jones, a 1,000-yard rusher as a sophomore in 2017, barely touched the ball (69 rushes, 16 pass targets) and did little with it when he got the chance. His per-carry average fell from 5.9 to 4.5, 3.1 if you take out the UConn game outlier. He’s back, as is senior Ke’Mon Freeman, but you figure Lashlee and Dykes are hoping that mid-three-star youngsters like redshirt freshman TaMerik Williams or incoming frosh TJ McDaniel and Ulysses Bentley IV can carve out niches sooner than later.

But that’s enough run talk. This is still a Dykes offense. With Lashlee involved, SMU toyed with the idea of balance on standard downs (55 percent run rate, five percentage points below the national average), then went pass-heavy when (frequently) behind schedule. This did Hicks few favors. He completed 62 percent of his passes on first downs but only 52 percent on second and third.

Dykes gave Brown three starts midseason. He looked pretty good in garbage time against Michigan, then led SMU to wins against Navy and Houston Baptist. He struggled against UCF, then Dykes gave Hicks his job back.

Odds are good Brown will be the backup again this fall, as when Dykes lost Hicks, he brought in Buechele. One of many Next Great Texas QBs to go through Austin in recent years, Buechele is a safe, steady passer (62 percent career completion rate, with just 11.5 yards per completion) and a willing runner. He was supplanted by Sam Ehlinger, but he has shown toughness and potential leadership ability.

Either he or Brown will have a receiving corps far more proven than it was this time last year. Senior James Proche became a solid, if overused, No. 1 target; his 146 targets were seventh-most in FBS. He was enough of a center of gravity, though, that players like juniors-to-be Reggie Roberson Jr. (52 catches, 802 yards) and Tyler Page (12 catches, 234 yards) were able to occasionally find downfield mismatches.

Every member of the receiving corps is scheduled to return, and Dykes added both Rice transfer Kyle Granson, an exciting potential up-the-seam threat (he averaged 13.4 yards per catch despite woe around him), and three mid- to high-three-star freshmen in Keontae Burns, Calvin Wiggins, and Rashee Rice.

If the line is at least semi-stable, and there’s a clear winner between Buechele and Brown, one assumes this will look far more like a Dykes offense than what we saw last year.

I really can’t say enough good things about Kane. The former Kansas linebacker is likely to be leading a power-conference defense very soon.

SMU’s greatest defensive strength last year was its extreme well-roundedness. The Mustangs were top-65 in rushing efficiency and passing efficiency, in rushing explosiveness and passing explosiveness, and in sack rate and stuff rate. They were a bit too reliant on their pass rush on passing downs (21st in PD sack rate but only 89th in marginal efficiency), but they forced so many passing downs that that was only a marginal concern.

They did this despite a lack of seniors and quite a few injuries. The top two linemen both missed time, and a potentially key piece in end Tyeson Neals played only three games. And in the back, basically everybody but safety Rodney Clemons missed at least one game. Neals’ was the only long-term injury of note, but this unit was constantly getting shuffled around, and it still produced.

Theoretically, now comes the reward for the shuffling. Five linemen, five linebackers, and six DBs all return after making double-digit tackles a year ago, and that doesn’t even take into account a wave of sophomore and junior linemen who looked pretty good in limited opportunities.

SMU had a pass rush by committee, and most of the members of the committee are back. Linebacker Richard Moore (13.5 tackles for loss, five sacks, 25.5 run stuffs) returns, as does end Delontae Scott (10.5, 4.5, 14), but the power here is the depth and diversity. And the guy calling the plays.

The pass defense was decent enough, especially when you take the pass rush into account, but the coverage was a little on the soft side: SMU allowed a 63 percent completion rate (101st in FBS) and suffered the aforementioned passing downs issues.

A deep safety corps gets a bit of a cleanse — leaders Rodney Clemons and Patrick Nelson are back, but the next three on the list aren’t — and while three of the top four cornerbacks return (including Christian Davis, who led the secondary in passes defensed), you figure there’s room for competition from youngsters and newcomers like three-star JUCO Sam Westfall.

SMU had an all-or-nothing special teams unit in 2018.

This unit, which predictably ranked a middle-of-the-road 62nd in Special Teams S&P+, returns intact.

S&P+ was right and wrong about SMU last year. It projected six wins and came within a three-point loss to Tulsa of nailing that, but it also projected a lot of offensive competency and very little defensive play-making. Got it backwards.

Experience should be the Mustangs’ friend this year. SMU is 27th in returning production, 20th on a defense that so greatly exceeded expectations. The Mustangs are projected 85th, but that doesn’t take Buechele’s transfer into account — maybe he pushes them into the 70s.

Either way, odds favor bowl eligibility. SMU is a projected favorite in six games and a one-possession underdog in four others. As long as the Mustangs split their first two games (at Arkansas State, North Texas at home), postseason hopes should be realistic.

The future of this program is hard to figure out — Dykes will probably have his offense up and running by 2020 at the latest, but he might have to replace Kane by then, too. In a tough AAC, his own personal streakiness could manifest itself in good or bad ways. But for 2019, we’ll just set the bar at six or seven wins and worry about the rest later.

All 2019 preview data to date.