STILLWATER, Okla. — Before its renovation in 2000, Oklahoma State's Gallagher-Iba Arena held a capacity of just 6,381 people. That was all it needed to be the loudest place I've ever seen a basketball game. It holds over 13,000 now, but it's not possible for it to be any more menacingly loud than it already was.

With the arena basically butting up to the east side of the football stadium, expansion options were limited, so they just lifted the roof and expanded straight up.

Be it by choice or necessity, Oklahoma State was long a basketball school. The Cowboys won the NCAA Tournament under Henry Iba in 1945 and 1946 and nearly won in 1949 and 1951 as well. They struggled for a couple of decades after Iba's retirement — for a while, you could legitimately call OSU a wrestling or golf school — but surged back under Eddie Sutton's guidance. They rode Bryant "Big Country" Reeves to the 1995 Final Four, then made another appearance there nine years later.

Perhaps it only makes sense, then, that for the OSU football program to rise to a similar stature, it would have to do so by paying homage to the roundball.

When Mike Gundy first became Oklahoma State’s offensive coordinator in 1994, his Cowboys operated out of the I-formation and offset-I. Sure, his Pokes would attempt a little bit of misdirection out of this, and they threw more than some other I-formation teams. But that formation was the standard of the day. It was especially the standard for Gundy’s mentor and head coach, Pat Jones.

The game began to change, however, and Gundy, out from under Jones’ wing, began to change with it. When he became his alma mater’s head coach in the mid-2000s, he began to find extreme value in changing before it.

“We’ve been on the cutting edge on no-huddle, on hurry-up offense, on nutrition, on hydration, on sleep, on cutting back on physicality in practice,” Gundy says. “We’re not scared to take chances.”

In his heart, he’s still a Pat Jones guy. Jones was a nose guard for Frank Broyles at Arkansas; running the ball is what football is supposed to be. “I still want to run the ball,” Gundy says. “I still want to be a tough football team, based on what was instilled in me as a player through Pat Jones.”

Even in the spread-heavy Big 12, Gundy started out running more than most. In 2008, the Cowboys broke through with nine wins and a No. 16 AP finish, and they did so by averaging 245 rushing yards per game, eighth in the country. But at the end of an injury-plagued 2009 campaign, they were shut out by Oklahoma and scored only seven points in the Cotton Bowl against Ole Miss. At one point, they were held scoreless for more than 90 minutes.

Gundy decided changes were necessary. It was time to innovate.

Enter Dana Holgorsen.

One of the first branches off of the air-raid tree — he played receiver for air raid father Hal Mumme and godfather Mike Leach at Iowa Wesleyan in the early-1990s — Holgorsen came to Stillwater from Houston, tasked with marrying air-raid principles to a Pat Jones core.

It worked well enough that Holgorsen was named West Virginia head coach a year later.

Photo by Tom Pennington/Getty Images

Over the seven years from 2010-16, OSU boasted four 1,000-yard rushers, five 1,000-yard receivers, and another 14 500-yard receivers. The Cowboys ranked in the Offensive S&P+ top 10 four times in that span and won at least 10 games five times. They damn near won the national title in 2011.

Holgorsen left, and Gundy replaced him with old friend and former OSU receivers coach Todd Monken. When success earned Monken the Southern Miss job in 2013, Gundy brought in little-known Shippensburg offensive coordinator Mike Yurcich. After weathering some offensive line and youth issues, the success continued.

The main step in the progression: accepting what football was becoming, and getting out in front of it.

“We stood back as a staff and said, okay, what direction is college football going?” Gundy says. “Well, the majority of schools are trying to use the width of the field and the pace of the game as an equalizer, almost like in basketball, when a team used to slow it down when they played good teams.

“They said, instead of there being X number of possessions, they’re better than us, so we want 25 percent less so they can’t be better than us more than they are.

“So essentially,” Gundy continued, “it’s basketball on grass.”

He obviously isn’t the first person to make this comparison, but while the concept of comparing floor spacing to the spread offense makes intuitive sense, Gundy is talking about something else, too -- adaptability.

In Frank McGuire’s seminal Offensive Basketball (1960), he lists five traits of team adaptability:

1. Slow down the offense when an opponent is too fast for us.
2. Crash the offensive board if the opponents cannot fast-break.
3. Play a possession game when the opponents adopt that style of play (meet fire with fire).
4. Use the stall offense when the opponents have an unusually tall defensive player who can block our close-to-the-basket shots (maneuver him out of position with a planned attack).
5. Utilize the fast break when the opponents do not observe defensive balance or are big and slow.

Basically: acknowledge reality and use tempo as a weapon, one way or the other. When you don’t have a clear advantage, slow things down. When you do, floor it. If they’re faster than you, go slow. If they’re slower than you, go fast.

McGuire also wrote: “In my opinion, team offense begins with the fast break. But I also believe that it should be a controlled break and one that lends itself to the best use of the player talent available. I like my players to be so skilled in the use of the fast break that they can pull out of it without a bad shot or loss of the ball when it is obvious that the advantage is lost.” Start fast, and then slow it down if you need to.

In 2009, OSU averaged 68 1/2 plays per game. By 2012, the Pokes were up to 78.0. They found plenty of advantages to exploit and minimal reason to slow down.

It’s not just how many plays you run, though. It’s also what kind of plays you’re running.

The pick-and-roll of college football

The zone read is to college football what the pick-and-roll became to professional basketball in the 1990s, Pat White and Steve Slaton its John Stockton and Karl Malone. It is one of the more direct, effective ways of getting from Point A to Point B, and it has become ubiquitous over the last decade and change.

As with the pick and roll, however, the zone read has had to evolve. Give defenses enough time, and they’ll find an answer for anything. It’s up to you to change the questions.

Oklahoma State has excelled for both of the past two years with “packaging concepts,” and in this case, putting different “coverage-beating” pass concepts to each side of the field. Doing this gives quarterback Brandon Weeden options on where he wants to go with the ball, depending on the pass coverage.

That’s Smart Football’s Chris B. Brown from 2012 talking about packaged plays, basically the predecessor to the run-pass option craze we have seen catch fire in college football in recent years. OSU was an early adopter of both. You have to stay ahead of the defenses.

And what have defenses done?

“Teams have gone to these quarters concepts and these two-safety looks because the spread and zone read have pushed them in that direction,” Brown tells me. Quarters is basically a Cover-4 defense, one that mixes man and zone responsibilities and puts a ton of pressure on the safety position to line up nearer the line than most safeties, quickly read run or pass, and adjust accordingly.

Led by head coach Mark Dantonio and defensive coordinator Pat Narduzzi, Michigan State’s defense ranked in the Defensive S&P+ top 10 for three straight years (2011-13) and ranked 26th or better six times from 2008-15. In an era of explosive spread offenses, State did as well as anybody in slowing them down, and as this is a game of imitation, it led to others experimenting with quarters coverage.

In turn, that led to coaches like Gundy, Monken, and Yurcich trying to figure out ways to make those all-important safeties’ lives hell.

“At some level,” Brown says, “this stuff is just remixing old things. It’s as old as football to have either a run or a pass, or then sometimes fake a run and throw a pass. And then the concept of reading defenders — well, in option football, you’re reading defenders. And if you just call a drop-back pass, you’re reading defenders. So it’s just sort of a remix of the whole thing, Girl Talk for football.

“The zone-read stuff started because you had inside zone, and if the defender didn’t crash, you called a bootleg. So then it’s just like, well, how about you just read the guy?”

The run-pass option, then, became a devastating extension of the zone-read remix. If the quarters safety or a defender in the flat comes flying up to play the run, you used to try to perfectly time a play-action pass. Now you just read those guys. You can basically call a play-action pass in real-time, and you can always be right.

When he guest-hosted Podcast Ain’t Played Nobody back in May, Miami defensive coordinator Manny Diaz told me that the RPO “is a way to take the onus off of the play-caller to be in the correct play and” — ready for another basketball reference? — “turn the quarterback into a point guard where he has the different options. We can run pick and roll three straight times, and one time I get a lay-up, one time a 3-pointer, one time a dribble drive to the basket.”

It’s not hard to see why Gundy, a former basketball player, would be drawn to such an idea.

Narduzzi is now the Pitt head coach and is still a quarters guy at heart. And in two games against Oklahoma State in 2016 and 2017, his Panthers gave up 104 points and 1,355 yards. Perhaps it’s fitting that maybe the perfect example of a 2017 RPO came when OSU and Pitt met.

In one play, you can see why how the run-pass option was built to detonate a quarters approach.

The scene: Oklahoma State leads Pitt, 7-0, midway through the first quarter. The Cowboys have already scored on a 20-yard Justice Hill touchdown run. They line up on second-and-10 with three receivers to the right (trips) and one to the left.

Brown: “Pitt’s a quarters team, and the quarters adjustment for trips is usually to have a backside safety reading run on his side, but then he’s also responsible for the No. 3 receiver [the third from the sideline] vertically.”

Gundy: “Quarterbacks are now reading three people on the field at once on one play. It used to be one, then it got to a little bit of two. But now guys are reading three people — they’re catching a pre-snap of a guy, then they’re post-snapping a couple of other guys.”

Brown: “It only makes sense to have a true downfield RPO if you’re dealing with defenders who are in some sort of conflict. The basic examples of that being, in a quarters coverage, the safeties are the guys in conflict. Their job is to come up, and they flat-foot read. They either come forward against the run, or they drop back to pass either by matching the second receivers or doubling the No. 1 receiver [first from the sideline].”

At the snap, you see the roles taking shape. Dillon Stoner, the guy Brown called the No. 3 receiver, takes off on a vertical route up the seam, while the No. 2 sets up for a bubble screen. The linebacker lined up opposite Stoner crashes toward the line of scrimmage to defend against the run, and the backside safety (i.e. the guy guaranteed to be wrong) appears to react to the potential bubble screen.

At this point, Mason Rudolph has made his read. With the safety reacting to the bubble screen and the linebacker reacting to the run fake, he’s throwing to Stoner, and Stoner is all sorts of open.

This was exactly how Oklahoma State drew it up.

What the RPO isn’t

In 2016, OSU had four passes of 80-plus yards (first in FBS) and 18 of 40-plus (seventh). Though six games in 2017, it’s the only team with 15 passes of 40-plus yards and one of only five teams with at least five passes of 60-plus.

The machine has functioned as it was intended ... for the most part. But a couple of key injuries at very key spots knocked the Pokes from the ranks of the undefeated.

Facing TCU on Sept. 23 with first-time starters at right guard and right tackle, the Cowboys found one of the RPO’s biggest weaknesses far more exploitable than normal.

“The biggest concern with RPOs is the quarterback getting hit,” Yurcich says. After all, part of the deal here is that they are blocking as if it’s a run play; they are not protecting the passer. [ESPN’s] Jon Gruden calls it ‘Ridiculous Pass-Protection Offense,’ and I understand what he’s saying. So everything that we focus on from a run-pass option standpoint, we’ve got to get to where it’s sound so we feel the quarterback can make a read, and if he’s correct in what he sees, then he shouldn’t take a hit. That’s Goal 1.”

Goal 2, you could say, is actually having threats. Against TCU, the equation fell out of balance because OSU couldn’t run the football. Hill averaged just 4.1 yards per carry, and only seven of his 25 carries gained five or more yards.

An inefficient ground game meant an idle run threat and a lot of second- and third-and-longs for the Cowboys. Rudolph got sacked and lost a fumble on a second-and-10 in the second quarter, and he threw an interception on third-and-17 in the third. OSU got rolling late in the second half but lost, 44-31.

An RPO isn’t a cure-all. Nothing is.

“I know a lot of high schools are like, ‘I’m gonna RPO everything — everything is an RPO!’,” Brown says. “At some point, there are diminishing returns. You could RPO everything, or you could just fake it and actually have seven-man [pass] protection and three guys running routes, and it’s actually a better play.”

According to Gundy, we’re approaching a quarterback’s limitations. “I don’t think it’s feasible for quarterbacks to do more than [read three guys], all during”—he snaps his fingers—“what could be three seconds.”

There’s one other problem with reading defenders in general: it only works if there’s somebody to read.

There are a lot of ways to defend the pick and roll in basketball, but one is perhaps far more common than the others: just go zone. It’s the opposite for RPOs — just go man.

“If you just play man on everybody,” Brown says, “there’s really not anybody to RPO, right? The guy who’s covering a slot receiver is his man — he’s not looking for a run play.” There’s no one in conflict. “And if you want to run an RPO [against that], then you’re better off checking to a typical three-step pass play so you get better protection.

“You can draw up a million cool RPOs, but it really only makes sense to RPO the guys who are in conflict.” If you have the athletes on defense to match the offense, then you simply play man coverage, tackle well on the short stuff, and take your chances.

And if you don’t have the athletes, well, you better find them.

What next?

So if QBs can’t read more than three guys, and if defenses are perhaps slowly beginning to figure out what it takes to address the RPO, where do things go from here? Gundy and OSU have figured out at least a few new wrinkles.

“The direction it’s gonna go is, multi-formation and multi-snap counts,” Gundy says. “That’s what we’ve done the last two years.

“People weren’t in very many different formations [at the start of the RPO evolution] because they said, if we can minimize the number of looks that a quarterback sees, then we can ask him to see more people. If there’s a whole bunch of looks, and there’s three people, then you [multiply] all that times all the people, and you’ve got big numbers.”

As Chip Kelly and Oregon were revolutionizing the spread offense in the late-2000s, one of the key concepts was being able to run the same play out of countless formations. Gundy and OSU are attempting that same concept for RPOs. “We’re multi-formation — all while trying to minimize the three people he’s looking at — but we’re also multi-snap counts. We’re changing it all the time, trying to keep [defenses] off balance, keep them from trying to move and disguise and stuff because they don’t know where we’re snapping the ball.

“Beyond that, I’ve got no clue where to go.” (That’s probably what he’d tell me if he had no clue where to go ... and what he’d tell me if he knew exactly where to go.)

Photo by J Pat Carter/Getty Images

Brown thinks offenses will move more in the direction of Penn State, continuing to chug down the RPO tracks but trying to get the quarterback hit less.

“If you watch Penn State,” he says, “almost everything for them is based on having five interior linemen and an H-Back. Oklahoma State is still in 10 personnel [one back, four receivers] because of those great receivers, but now it’s much more 11 personnel [one back, one tight end or H-Back] — and they’re usually blocking the six interior linemen and linebackers.

“They’re not very often leaving the defensive end unblocked so he can crash down on the quarterback as he’s making one of those downfield RPOs. They’ll do a bunch of variations of that.”

Just know this: We’re not going back to the offset-I any time soon.

Back to Diaz on PAPN: “Everybody talks about how things are cyclical,” he said, “but I don’t know that we’re ever going back because, why wouldn’t you have a run play, and then all of a sudden you’ve got a [defensive] box look that doesn’t look advantageous to the offense? People have been throwing bubble screens [as a counter to that] for a long time, but now they’ve just decided, ‘Let’s just throw the ball down the field.’ Why wouldn’t you?”

Says Gundy: “Offense is always based on your quarterback. Pee wee, junior high, high school, college, NFL, 90 percent of the teams are going to go as their quarterback goes. If you’re trying to accomplish something your quarterback’s not good at, you’re going to lose.

“I don’t care how good a throwing guru you are, you have to execute what the quarterback can. If he’s a very cerebral player, you can do a lot.”

In terms of reading defenders, quarterbacks are much more cerebral than they used to be.

They’re getting more like point guards every day.