How the spread offense conquered college football, from Hal Mumme to Joe Burrow

It had to be LSU that landed the final blow.

You knew the battle was almost over when Nick Saban's Alabama changed its stripes, opened up its offense and kept winning. When the Los Angeles Rams nearly won the Super Bowl with a quarterback from an Air Raid offense, then the Kansas City Chiefs did win one with an even better Air Raider, the ref had to think seriously about stopping the fight. But when LSU not only adopted a spread identity in 2019, but then proceeded to put together maybe the best offensive season in the history of college football, the fight was done.

The spread offense revolution is over. The spread won.

The Tigers had come to personify Big Burly Manball more than anyone. They were the school of workhorse backs and impossibly physical defenses. They beat Alabama in 2011 while scoring six points in regulation, after all. But after a couple of aborted modernization attempts, head coach Ed Orgeron put together the perfect mix of personnel to operate a devastating, innovative spread offense. Despite playing half the teams in the 2019 SP+ year-end top 10, LSU threw for more than 6,000 yards and rushed for more than 2,500, went 15-0 and took home both the national title and, through quarterback Joe Burrow, the Heisman Trophy. Seven Tigers offensive players were selected in the 2020 NFL draft, and more will be next year.

Despite ties to national titles won by Oklahoma (2000), Florida (2006, 2008) and Auburn (2010), the spread was, for much of the 2010s, still regarded by many as an alternative, as something you attempted if you had the perfect quarterback for it or if you didn't have the recruits to run a regular offense.

Everything else is the alternative now. The spread offense is the default college football offense, and considering which quarterbacks are getting selected the highest in the draft each year (Baker Mayfield in 2018, Kyler Murray in 2019, Burrow in 2020), it is the pro-style offense of the day too.

What the college football offense has become

Part of the draw of college football comes in the variety. There are lots of different ways to try to win a game; in a landscape that often includes dramatic talent differences among teams, a bold attempt to do something different can pay off handsomely, at least for a while. After all, the spread revolution itself was defined by a series of bold gestures: Kentucky hiring Valdosta State head coach and Air Raid offensive mastermind Hal Mumme in 1997, for instance, or new Tulane head coach Tommy Bowden hiring an unknown offensive coordinator named Rich Rodriguez from Glenville State, also in 1997. Bob Stoops took on a risk of his own when in 1999, as new head man at option haven Oklahoma, he hired Mumme's protégé, Mike Leach. There is a place for bold and different in this sport.

Still, coaches follow what works, and as Miami head coach Manny Diaz once put it to me, "There's such a thing as a 'college football offense': 90% of America runs 60% of the same plays."

So how does that 60% of 90% look at the end of the spread revolution? Teams aren't throwing more, but they're throwing more efficiently. The spread revolution has been more or less defined by four innovations: the Air Raid, the zone read, tempo and the run-pass option.

In 1989, the year Mumme got his first collegiate head-coaching job at Iowa Wesleyan, the average top-50 college football quarterback (per passer rating) averaged 25.8 passes thrown per game. The average had jumped to 29.1 by 1999 -- a jump that probably had more than anything to do with the phasing out of the triple option as a mainstream attack -- but in 2019, the average remained just 29.9. This isn't the NFL, where analytics breakthroughs have led to higher pass rates. For all the air-it-out attacks, there are still plenty of grind-it-outs.

Mumme and Leach, however, still made a significant impact in both the literal spread of college football offenses, horizontally and vertically, and the development of the passing game itself.

"For so long in college football, it was an I-tight formation with a fullback and a tight end," Memphis offensive coordinator Kevin Johns said. "You allowed the defense to, what we would say, play in a box, in a nice, tight, little area. Offenses eventually said, 'We don't want to do this; we're gonna make you defend the entire field.' You were going to spread the defense out, get the linebackers in space and try to create some mismatches with some fast slot receivers."

Those mismatches, combined with better route combinations and a larger variety of receiving targets, have resulted in dramatically more efficient passing:

  • Top 50 quarterbacks in 1989: 132.6 passer rating, 57% completion rate, 13.7 yards per completion, 4.0% interception rate

  • Top 50 quarterbacks in 1999: 136.8 passer rating, 59% completion rate, 13.3 yards per completion, 3.2% interception rate

  • Top 50 quarterbacks in 2009: 147.9 passer rating, 64% completion rate, 12.6 yards per completion, 2.6% interception rate

  • Top 50 quarterbacks in 2019: 154.1 passer rating, 65% completion rate, 13.0 yards per completion, 2.0% interception rate

The difference is even more stark among the best of the best. Seventeen of the 20 highest single-season passer ratings ever, and each of the top eight, were recorded in the 2010s. Mayfield set the bar at 196.4 in 2016, then topped himself (198.9) in 2017. Tua Tagovailoa (199.4) and Murray (199.2) both cleared that in 2018, then Burrow (202.0) topped them all last fall.

In 1989, Iowa State's Bret Oberg ranked seventh in the country with a 143.7 passer rating. That would have ranked 44th in 2019.

It's still all about options (but the pitch man is 12 yards downfield)

Rodriguez allegedly stumbled into the zone read concept when Glenville State quarterback Jed Drenning bobbled a snap while attempting to run an inside zone handoff. He kept the ball instead, which fooled the defensive end, and took it upfield for a nice gain. Looking for any advantage he could find, Rodriguez put it in the playbook. It followed him to Tulane, Clemson and eventually West Virginia, where as head coach he won or shared parts of four Big East titles, booked three top-10 finishes and nearly made the 2007 national title game.

Needless to say, other coaches adopted it too.

"It's always said that there's no patent on scheme," Rodriguez said. "There's no secret sauce. If someone's having success, whether college, pro or even high school, coaches are gonna watch the film. I remember, I think it was the year we won the Sugar Bowl at West Virginia [in 2005], we might have had 50 other colleges visit us. And we were sharing everything! We might learn from you too!"

The zone read forced the defensive end to stop and read the situation instead of making an all-out charge at a shotgun quarterback. It evened the numbers out, forcing defenses to account for an extra ball threat. This set into motion a fun game of cat and mouse. Defenses deployed safeties or outside linebackers as overhangs, protecting the end and positioning someone to tackle the QB on the keeper. Offenses responded by throwing bubble screens and quick passes to the perimeter, while coaches such as Oregon's Chip Kelly had a lot of fun reading not only the end but basically every player near the line of scrimmage on the zone read. And as defenses began to further reposition themselves horizontally, offenses began to look for ways to capitalize vertically. Enter the run-pass option.

Joe Moorhead, now the offensive coordinator at Oregon after stints as Penn State's OC and Mississippi State's head coach, has a pretty clear memory of where the concept came about for him: "We were playing at Miami of Ohio in 2005." He was Akron's OC at the time. "I can see it -- we were going left to right on the screen, and the ball was on the left hash," Moorhead explained. "It was a two-tight end set. We called it 15 Lock Stick, and the [strongside] linebacker dove in to stop the inside zone, so we threw the ball to a tight end in the flat. It was a run coupled with a throw off of a second-level defender," or what we now know as the RPO.

Moorhead and other coaches began figuring out the guys in conflict -- guys with both run and pass responsibilities who had to react quickly to what was unfolding -- and reading them like they did the defensive end. Whatever he chose to defend, the offense did the opposite. "It was the point guard era of the spread offense," Diaz said. "You had to take your guy out of run-pass conflict and play some version of man-to-man to defend those things. And if you can't play man" -- and only so many can -- "there are issues."

"Everything now is conflict," Iowa State defensive coordinator Jon Heacock said. "It's always been about conflict, but in the past, it was always some form of same-play conflict: run or pass." The triple option, for instance, put edge defenders in conflict when it came to the quarterback keeping the ball, giving it to the fullback on the dive or pitching it to the pitch man out wide. But if it was a run, it was a run.

"I'm old enough to remember when the wishbone came about," Auburn defensive coordinator Kevin Steele said. "When Alabama sprang out the wishbone against Southern Cal [in 1971], that's not me reading it in a book -- I'm old enough to remember it! Coaches would talk with other coaches about how to defend the dive-quarterback pitch: 'Get somebody inside the loaded [box], get somebody outside the loaded.' The spread offense is at the peak of the new norm, and when you add the RPO, it's really dive-quarterback pitch! The pitch man's just running a route down the field."

Rodriguez has found plenty of ways to implement the RPO game into his playbook. It's hard to see why you wouldn't. "Anything you run in the quick passing game, you can basically tag onto a run play," he said. "In college, where the [blockers] can get three yards downfield, you can run just about every run play, specifically your zone, with your quick game and have the best of both worlds." Even predominantly run-based attacks such as those of Minnesota and Louisiana found extreme benefit to incorporating the RPO.

Tempo is mostly situational now

In 2009, teams averaged 26.6 seconds of clock time between snaps. By 2013, that had sunk to an even 25. That season, 22 teams snapped the ball at least 1,000 times. Led by examples such as Art Briles' Baylor and Kelly's Oregon, some spreads had grown devastating in their ability to maximize personnel advantages and keep defenses frazzled and scrambling with no-huddle attacks.

Defenses have adapted, shrinking and simplifying their calls in the same way that offenses did.

"Part of the offense's advantage was just trying to go as fast as you can," said Jeff Thorne, head coach of Division III national champion North Central. "We could limit the defense's calls and force them to not be able to communicate well. But I think that has totally flipped. A defense feels like they can communicate with one word, just like an offense can. One word can mean a coverage or a blitz or a front, whatever it might be. Their communication has caught up, and I don't think that just going fast all the time is necessarily the answer anymore."

Indeed, by 2019, FBS offenses were back to averaging 26 seconds between snaps, the lowest since 2010. Only eight had 1,000-plus snaps. Once offenses realized there wasn't as much of an advantage to tempo, it quickly became a situational thing. "When this all started," Steele said, "the spread was tied to the no-huddle. You didn't hear any conversation about college football where someone was not talking about either how great the no-huddle was or a defensive coach was talking about how unfair it was. Now, no one even talks about it. It's just the way we play."

In one way, offenses have maybe ceded too much ground.

Per Sports Info Solutions, about 24% of snaps in 2019 -- not including the first play of a given drive -- took place within 30 seconds of real time from the end of the last play. That included 28% of first- and second-down snaps, but only 14% of third-down snaps. After using a little bit of tempo, offenses reset before a big third-down call.

That means the defense gets to reset too.

"I think that's a mistake," Rodriguez said. "I understand you want to get it right, but sometimes we think too much. And if you do that, you're playing back into the defense's hands. On third down, you're gonna see a whole new crew come in, they're gonna have third-down personnel. If you're an offensive guy, you have the ability to not let them sub. We make a point of that: When we play teams that sub a lot, we're gonna go faster on third down."

Good defense is all about multiplicity

Today, offense comes down to reads and conflict. How much can your quarterback process both before the snap and directly after? How well can your structure isolate specific defenders and make them guess wrong? That makes it tricky for a defense to scout opponents in the traditional way.

"People always try to take away whatever a team does best," Heacock said. "Well, the hardest part now is when they're in these offenses, what are they running the most of? That's hard to decipher. They're so based on what you're doing, and what they do best in one game might not be what they do best in another. I think it used to be, a team lines up, and, 'Hey, they're a power team, a tight end run team, an inside zone team.'" And now they're just designed to do whatever you're not set up to stop.

Heacock set up his Iowa State defense, then, to show as little as possible. He crafted a unique version of the 3-3-5 defense, with a tight front three and eight players who swarm to the ball. In a way, they do what offenses have long sought out to do -- create space for their runners. Their effect is to prevent big plays and force offenses to tolerate going five yards at a time. "I'd never heard of doing this, to be honest," Heacock said. "We just tried to do what we could do in the conference we were playing in with the guys that we had. But when you look out there on offense, everything looks the same. That's where you're trying to get an advantage."

Other defenses have gone in a different direction. If the spread offense is about getting defenses to declare themselves, declare the thing offenses are least interested in doing. "Someone told me a long time ago," Diaz said, "one way to take away the triple option is to take away the option. You tell them what you want them to do and then force them to do it."

You can do that in part with the alignment of your players. You also can do it by convincing the quarterback he sees something he doesn't.

"Give 'em as many false reads as you can give 'em," Steele said. "We went to seven defensive backs against LSU [in 2019]. Now we had the luxury of having players who could do what we asked 'em to do -- it comes back to talent and having the players -- but it helped us some. Now the key to that is being able to hold up in the run game. So you've gotta have the right kind of DB." Most don't. But Auburn held LSU to 5.8 yards per play and 23 points -- by far LSU's season lows -- in a narrow loss.

"Defenses do a great job of showing quarterbacks something pre-snap," Slippery Rock offensive coordinator Adam Neugebauer said, "but you better not just assume that's the coverage and that's the picture. We tell our quarterbacks, 'Find the proof.' They're going to show you what you want to see, but on the snap, the leverage is going to change."

The more familiar defensive coaches get with the spread, the more they can adapt. "Because so many offenses are running these things now," Moorhead said, "defenses aren't just seeing it three or four times a season, they're seeing it every day in spring ball and fall camp. That makes a huge difference."

"When I was coaching on other staffs before I met Coach [Matt] Campbell," Heacock said, "I had coached against it, but I had never coached with it. What's helped me understand it is practicing against it every single day on the practice field. Instead of having to be an expert three times a year, you have to be an expert 365 days a year. It helps you understand their mindset."

Good offense is all about multiplicity

In 2000, Northwestern offensive coordinator Kevin Wilson visited Rodriguez at Clemson to learn how to create a tactical advantage while maintaining the power principles he and head coach Randy Walker preferred. The move reaped immense dividends: The Wildcats flipped from 110th to 10th in scoring offense and won a share of the Big Ten title, all while maintaining their power principles.

Things began to get particularly interesting when Bob Stoops hired Wilson as Oklahoma's co-offensive coordinator in 2002.

"When we started years ago," Wilson said, "it wasn't by design, it wasn't to be smart; it was just kind of out of necessity. When we got to Northwestern, we just weren't quite as good. Randy wanted to run the ball and be in I-formation, but we didn't have fullbacks and tight ends. He didn't want to throw the ball, so we started looking at the gun spread.

"I got hired [at Oklahoma] because Bob wanted to run a little bit more, but we had that Leach-style passing that we kinda morphed into the spread run game. I think that's where it kind of took off. Fifteen, 20 years ago, you were either throwing the ball every play -- Air Raid -- or you were basically a quarterback option attack." Over time, you could do both.

By 2008, Wilson had come across the perfect spread recipe. The Sooners' attack, led by future No. 1 pick Sam Bradford, could go no-huddle or slow things down. They could line up players such as tight end Jermaine Gresham, running back Demarco Murray and H-back Brody Eldridge in a number of different spots to, as Wilson put it, "go from little to big without substituting." They used countless formations. A defense had no way to know what was coming next and, thanks to tempo, no time to prepare for it. The Sooners averaged 51 points per game and scored 60 or more in each of the last five games of the regular season.

OU stumbled at the finish line that year, blowing a couple of goal-line chances and losing in the BCS title game to Florida. But the Sooners showed how offenses could take things further as defenses adapted.

Johns, a former Wilson assistant at both Northwestern and Indiana, did with incredible Memphis running back Kenneth Gainwell what Wilson did with Murray years earlier, moving him around to create mismatches, helping him to gain 1,459 rushing yards and 610 receiving yards in 2019.

Gainwell types are a product of the time. "Down at the high school level," Johns said, "everyone lines up in spread. A lot of times those kids who end up in the slot are the kids that are maybe too small to be just running backs in college. Maybe they're not tall and long enough to be a true outside receiver, but man, they're really good athletes. If you can recruit the Kenny Gainwells, who are almost tweeners -- they were slot receivers in high school, but you can hand them the ball or have them run routes -- those are the kids that are coming out a lot more in high school."

And those are the kids who can prevent defenses from getting too comfortable.

Versatility was a key to North Central's 2019 Division III national title. "We use just about every formation you can be in," Thorne said. "You have to have the right personnel to do that, though. We had two really versatile tight ends, we were blessed with a couple of good running backs and really good receivers, so we could get into all sorts of different formations."

Slippery Rock had the best offense in Division II last year for a lot of the same reasons.

"We showed an unorthodox formation every single week," said Neugebauer, a disciple of Gary Goff (Valdosta State's head coach and a former player for Mumme and Leach). "Even with a great defensive coordinator, if they see something they haven't seen before, you make them adapt on the fly. That causes them to have to communicate more on the sidelines -- 'OK, this is what we're gonna do to set up' -- but now they're not talking about our base plays, they're talking about how to line up against us. And that's where we get an upper hand. If you allow the defense to get set and line up properly, they won."

"That's what made Central Florida so tough when [Scott] Frost was there," said Steele, whose Auburn team lost to UCF in the 2018 Peach Bowl. "It was like a new game every week: 'Have they changed coaching staffs this week?' It was different stuff."

In 2019, though, LSU proved you could still add concepts to the repertoire, not simply use a large quantity of what already exists. The adjustment to slowing down RPO attacks has generally been playing man coverage instead. It produces cleaner assignments and fewer in-conflict defenders. It also leaves you vulnerable to getting gashed if you don't have guys who can play man defense well.

"Offenses try to equate the numbers or get a numbers advantage by using the quarterback either as an extra guy in the run game or in the RPO part," Rodriguez said. "Defensively, the answer is to play press man, take away all those easy RPOs and throws. But now you're seeing teams like LSU that throw a bunch of rub routes and pick routes" -- route combinations designed to beat man coverage -- "because you're giving them so much single coverage.

"That's where the chess match goes."

We might be through with the giant innovations for a while. We might have to wait for another true offensive shock such as the zone read or RPO. But the chess match indeed continues.