Option Football
Option Football
Nebraska’s season hinges on the offense finding ways to generate dependable, easy yardage on early downs. Are the spring game's numerous RPOs the answer?
Though the unit was very poor as a whole, Nebraska’s offense had no real issues at generating explosivity last season. With a decimated roster, the staff was able to cobble together a good deal of chunk gains, generating them from different places in different parts of the season: under-center misdirection lead runs early; long Heinrich Haarberg keepers off the option and even longer play-action passes off those option looks in early conference play; and downfield bombs to Jaylen Lloyd late in the year. If Nebraska scored on a drive in 2023, it’s probably because it got one or multiple of these big plays.
Where the offense mainly had issues1 was in efficiency. Nebraska simply did not have the ability to execute well down-to-down to maintain drives. If you asked NU’s players to put on a long possession to put points on the board or run off some clock, the drive was probably going to be ruined by a lineman blowing a block, or a quarterback misreading a play or running themselves into a pressure, or a receiver messing up the spacing on a concept. If not multiple of those things at once. Or someone was just going to turn the ball over.
Below is a chart visualizing every FBS offense last season by both its efficiency and explosivity. The horizontal axis left to right measures success rate, a statistic that essentially illustrates how consistently effective you were down-to-down. The vertical axis measures teams’ explosive play rate. So a team in the top right would be both very efficient and very explosive (USC, Kansas, LSU, Oregon), and a team in the bottom left would be both inefficient and inexplosive (Iowa):
From a success rate perspective, things were pretty bleak for NU, only finishing ahead of Iowa, Mississippi State, Iowa State, BYU, Vanderbilt, Michigan State, Stanford, Arizona State, and Pittsburgh among Power 5 teams last season in how they performed down-to-down. But in terms of explosiveness, Nebraska was … not bad! NU actually turned out big plays more often than teams like Michigan, Georgia, Oregon, Washington, Tennessee, Ohio State, Arizona, and Kansas State, who all were among the best overall offenses in the nation last year.2
If NU is baseline competent at generating explosive plays, the goal for the offense this year is to get the efficiency up to the level. If the offense can continue to produce some big plays and add better ability to methodically move the ball, this will be a vastly improved unit.
NU can expect some better operation in that area with just the general development of players — what was a pretty young offense at the end of last year has another offseason of practice and reps under its belt — but part of the lack of efficiency was also because Nebraska didn’t know who it was at any point of the season on offense.
Most good offenses have a focal concept they have repped so much in practice that they can execute it with a blindfold on. A play an offense knows it can run for 3 or 4 yards even in the most adverse of circumstances, when nothing else is working. Michigan last year always was able to run Duo. If Texas was in a tight spot, you knew you were facing a Slide RPO. Georgia’s back-to-back title run was built off throwing play-action passes to tight ends Brock Bowers or Darnell Washington in the flat. Lincoln Riley teams have GT Counter. Even when Kirk Ferentz’s offenses at Iowa have been bad, they’re usually still pretty good at operating the wide-zone-and-boot game most years.
Ask yourself, what was that play or concept for Nebraska last season?
In the first two game, with Jeff Sims under center, NU had some success with the more gimmicky quarterback runs off draws and lead blocks, but those plays aren’t really base concepts you can run 10 times a game without a defense adjusting to take them away. During Haarberg’s entrance in the lineup, it seemed like the load option concept or the I-formation power stuff could be something, but then better Big Ten defenses essentially laughed those off the field. Late in the season, some of the under-center and pistol jet-motion read-and-run plays with arc and slice blockers were pretty consistently hitting, but Wisconsin and Iowa both were able to take them away. NU was a pretty consistent interior running team, with a notable 47% success rate on Duo concepts last year, but other than that the things that moved the ball for Nebraska last season were things that only worked for a few weeks at a time.
Part of the lack of identity was that injuries forced a lot of inexperience and role players into the starting lineup, and the NU staff was just trying to triage together some way to get points on the board each week. For the circumstances, I don’t think they did a bad job. I also think part of the issue is also that coach Matt Rhule is sort of an offensive identity mercenary, not particularly married to any offensive scheme as long as it contains a certain sense of physicality and works. So he wasn’t walking in to Lincoln with a manifesto on how he wanted his offense to work, as some other coaches have.
Still, whatever the reasons, NU needs to find a couple concepts to get really good, or else it’s going to find itself in the big-play purgatory it existed in last season.
And I think the spring game showed us a possibility: option football, just not the kind you’re used to.
RPOs Can Be The Answer
Did you think true freshman quarterback Dylan Raiola was the star of the spring game?
Wrong. It was this concept:
https://twitter.com/b_peters12/status/1811500442674716690
This is called an “arrow” run-pass option. Nebraska ran this specific RPO concept eight times in the spring game, for an average of an even 8.0 yards per play. Seems like a pretty good early down option to me?
A “run-pass option” (oftentimes called an “RPO”) is a play in which the offensive line and a ball-carrier execute a traditional running play, with real run blocking, and other skill players execute a traditional passing play, with real pass routes — and then the quarterback chooses whether the run or pass portion of the play will get the ball based on the alignment or post-snap behavior of the defense. This is different than a “play-action pass,” in which everyone on the offense is intending to execute a passing play, but certain players may take actions that look like runs for the purposes of deception.3 There are many different variations to RPOs, including bubble screens that read overhang defenders (often called “tags”), downfield post or slant routes that read whether a linebacker steps up to defend the run (often called “glances”), and cross or orbit screens that make defenders try to get to the edge while running through traffic (often called “shallow” or “traffic” plays.)
RPOs are, essentially, a way to incorporate optionality into modern football. They’re ways to have multiple answers to a defense inside one play, without having to line up in the flexbone or get your quarterback hit 25 times a game. If the defense stacks a box, blitzes against a run, or comes out with its corners playing deep, you have the ability to throw it wide and pick up easy yardage, without having to audible or call a new play. The opposite is true if the D comes out with a light box or has its safeties deep: You get an easy carry up the middle with the blockers-to-defenders numbers in your favor. They’re also useful ways to slow down pass rushers or make defenders think instead of reacting or moving full speed. It may not be the “option football” NU fans are used to, but it’s a series of dives, QB keepers, and pitches that react to what a defense is doing. Sounds like the ethos of Husker offensive football to me.
Their ability to pick up easy yardage at low risk has led to a boom in RPO popularity in both college and pro football, and RPOs can be this early down efficiency answer for Nebraska, too: If operated well, they’re constantly sending the ball to where the defense is weak. They’re a tool to gain more consistent yardage. If a team’s corners are playing 10 yards off the ball to set up a deep zone shell to protect a heavy interior box, throwing a quick hitch to an outside receiver that makes it second-and-5 is better offense than running into a stacked box or throwing an incomplete pass. An offense that operates in that was is going to end up in fewer bad spots.
RPOs are just a part of a play-calling menu; you’re still going to want a traditional run game, a traditional dropback passing game, and a traditional play-action passing game. And I don’t think it’s in Rhule’s ethos to start running a super spread or anything. But NU really didn’t run many RPOs last season, and I think turning the dial up a bit is probably the easiest ways this team could generate better down-to-down efficiency. And getting really good at a couple of these concepts could just help the Husker offense pick up a lot of yardage it’s leaving on the field.
Based on RPO rates in the spring game, and some quotes from Rhule, the Nebraska staff might agree.
How Much Usage Is More Usage?
Nebraska last year was pretty allergic to RPOs. NU tried them at a decent clip when Sims was at quarterback, but Haarberg struggled with the mechanics, ball handling, and accuracy required to operate them well, so they largely went to the back burner after the first couple of games of the season. My charting last year had downfield RPO usage at about 6.9%, with 54.0% true run plays and 38.6% true pass plays (and a few trick plays smattered in). I don’t have charting numbers for any other teams, but I have to imagine that rate is toward the bottom nationally among FBS teams. It’s also a lot less than Nebraska has run in the past, or what coordinator Marcus Satterfield has used in his previous stops:
But the spring game was different. I charted 34 RPOs used during the game, which would represent a massive jump in deployment. They also delivered some of the game’s biggest and best plays. I’ve made supercuts of every RPO type used in the spring game:
I don’t really believe in making any huge takeaway from a spring game. What you run in a glorified May practice is not always tied to what you use in the regular season. Nor do I think the in-season usage will be as high as it was in the spring game. But I do think the heavy usage in the spring game shows us that Nebraska spent a lot of time practicing RPOs in its spring practices, which does mean something about the identity of the offense.
The spring is when you work on your basics. You’re not putting in designer gameplan stuff; you’re installing your base, bread-and-butter plays that you anticipate to use often during the season, plays you want to be very good at and have everyone on your team be able to run. That base identity, if you will. And it seems like a big part of that base identity NU was practicing in the spring was RPOs.
This would be a departure from some of the other offensive talking points mentioned in the offseason. At one point, Rhule said he wanted his offense to resemble the San Francisco 49ers.4 The Niners, and other teams in the “Shanahan” system, are primarily under-center, two-back teams that use diverse, designer run games and throw crossing routes downfield off play-action and boots. Those teams don’t really use RPOs or operate with much in-play optionality. The Niners only ran RPOs on 3.9% of their plays last year, and similar NFL schemes like the Vikings (2.5%) and Rams (3.7%) didn’t use them much either.
NU definitely did some of that “Shanahan” stuff in the spring, with several reps of condensed 12 personnel runs and shot plays. But we also saw the RPO rate rise. So, assuming that RPO increase in the spring isn’t a complete misnomer when the actual season starts, we can probably expect this to be a more hybridized attack than Rhule indicated through that “49ers” quote, a sort of offense that combines the Shanahan system of heavy-personnel runs and play-action with the optionality of a substantial RPO or read package. Mashing systems together doesn’t always deliver great results, but it’s certainly a clearer identity than what NU was running by the end of the season last year.
What could this RPO package look like, then?
The RPO Menu
During the portion of the spring game TV broadcast when Rhule was in the booth with the announcers, Kenny Bell5 noticed the increased RPO usage and its effectiveness and asked Rhule about what he was seeing. Rhule said this:
“Danny [Kaelin] and Dylan [Raiola] aren’t traditional zone read runners. They can do it, but that’s what Heinrich is special at. I think having Dylan and Danny read it and get the ball out of their hands is the way we’re going to try to attack people in the read game when they’re in the game.”
“Read game” is the operative phrase here. This, to me, is Rhule confirming that this needs to be an offense that embraces a substantial “read”-ing of the defense, as RPOs do. He’s also indicating the nature of those reads needs to change with his two new freshman quarterbacks. When Nebraska used a “read” on its run plays last year, the pull/second option on those plays were largely quarterback runs with Sims and Haarberg keeping the ball, running into the defense, and taking a hit from a linebacker or safety. Not only are Raiola — named the starter Wednesday — and Kaelin not the runners last year’s two primary quarterbacks were, but getting your QBs hit a bunch is a great way to get them hurt, fumble the ball more, and generally get them to think, “Hmm, that bigger program probably wouldn’t have me doing this … .” So the option-based reads of last year’s offense needed to change. Rhule wants to get the efficiency of those read plays, but without the quarterbacks’ legs being the primary outlet.
Here are some of the plays run in the spring game a bunch that can accomplish that:
Arrow RPOs
Arrow RPOs were the most used RPO concept in the spring with the starters in, and they’re the one I think we’ll see the most this season. Here’s a cutup of all the Arrow RPOs that NU ran in the spring game:
The play, essentially, is the Millennial Triple Option:
In the example above (the first play in the cutup), the five offensive lineman block down in a traditional run play on three of the defensive linemen and climb, leaving the strongside edge player (purple circle) unblocked to be read by the quarterback. The tight end (blue arrow), aligned in a wing, gives a rocker step inside to catch the defense’s eyes, then sprints out into the flat.
The quarterback’s initial read is that edge player in the purple circle: If the edge player stays wide or follows the tight end into the flat, the QB will hand to the back running into a five-player box (orange path) with numbers. If that edge player collapses down the line to play, the run, the quarterback can pull the ball and look to the second read.
The second read is the next-widest second-level defender, in this case the field safety (green circle). If that safety stays tight to the formation, the QB can whip the ball out to tight end with a 3-on-2 numbers matchup for the offense. If that safety gets wide with the tight end, the quarterback can take the ball up the alley vacated by the safety (red path).
In this play, the edge player in purple crashes down, so Raiola pulls the ball away from the back. The safety is rotating down in the box to play a hook zone in Cover 3 and is outleveraged by the two receivers blocking, so Raiola immediately whips the ball out to the tight end in the flat for a 15-yard gain and first down.
They don’t have to be with just the tight end, either. You can also function them with a slot receiver, as NU did a couple times in the spring game. You can see one example at 0:40 in the cut-up above.
Here’s a quick clip from a clinic that explains how then-Buffalo offensive coordinator Andy Kotelnicki (who later went on to Kansas and is now at Penn State) runs the play:
This is a concept that Nebraska’s run in the past, with varying degrees of success. A more QB-run based variation was a big staple of Scott Frost’s RPO package and worked pretty well, especially with multiple tight ends on the field, and created a lot of alleys:
And NU tried to run some Arrow RPOs (and Slide RPOs, a similar concept, where the flat route comes from across the formation) last year, using the concept for about 13 snaps in 2023:
Thirteen reps is a pretty small sample size, but Arrow/Slide RPOs had a 53.8% success rate last season, significantly better than the rest of Nebraska’s offense as a whole. But there was a disparity in its usage and success by the quarterback:
Sims: Five reps; one keep, two gives, two pitches; 20% success rate
Haarberg: Six reps; one keep, one give, four pitches, 83% success rate
Purdy: two reps; zero keeps, two gives, zero pitches; 0% success rate
The staff utilized the concept with all three quarterbacks, though a little more frequently with Sims than anyone else. Haarberg actually had the most success operating them, and is presumably still capable of running them and being a real threat as a runner in that alley.
You’ll lose some of that rushing-up-the-alley capability with Raiola. He’s a decent athlete, but he’s probably not going to take that keeper run unless it’s very open by the defense way overplaying the arrow route, so expect to see a lot of handoffs or flat throws from him. But he will provide a boost as a passer in his ability to get the ball out quickly on this play. He brings a creativity and creation in delivering the ball that the other QBs don’t. Watch the Raiola reps in the cut-up video, and you’ll see plenty of quick sidearm deliveries and angles to whip the ball in there fast, like a baseball shortstop. Watch Haarberg’s last play, when he throws the ball behind the receiver and makes him slow down on his path.
Though the top two QBs will approach them in different ways, Arrow RPOs, then, are also a useful concept both Raiola and Haarberg can run in a game. That’s also useful for a gameplan.
Bubble (“Tag”) RPOs
Bubble RPOs work in a similar way to Arrow RPOs, horizontally reading the perimeter of the defense to get the ball to edge if given a juicy look. But there are some differences.
Instead of being a flat route, the outlet player runs a “bubble” route, in which they go backwards behind the line of scrimmage before getting wide. That makes these routes more likely to come from a slot receiver than a tight end, as the Arrow RPOs typically did.
The Bubble RPOs also predominantly function as box-count plays, and are not always true post-snap reads with the QB evaluating the movement of a defense. Instead, many instead rely on counting the players near the line and out wide to cover the receivers before the snap and making a decision. If a defense has six guys in the heart of the formation and two players over the receivers, it’s easy to whip the ball out for a short gain to the edge with numbers. Some analysts don’t classify these as true “RPOs” as there isn’t usually a downfield or second-level component, and instead classify them as “tags” that are added onto a traditional running play. I’m counting them as RPOs this year, as I think the basic premise is the same.
Here’s a cutup of all the Bubble RPOs from the spring game:
In the example below (the first play in the cut-up above), the quarterback’s pre-snap read is the first edge player outside of the box, circled in purple. With that edge player aligned wide, the rushing concept has a six-blockers-on-five-defenders advantage in the box, with three receivers and three defenders to the edge, meaning the correct read here is a handoff. If that edge player had been tightened down to play the run, the math changes, though: It would have been 6 on 6 in the box, and a 3 on 2 advantage to the outside, meaning a throw to the outside was the correct read.
Some teams will read that edge defender after the play begins to watch their behavior to determine where the ball should go, and some even might incorporate a second-level read and quarterback run to make it a triple-option concept, but I don’t believe NU is doing that. Their QBs eyes were never on that defender after the snap, so I believe this is a true pre-snap determination only. That could change, though.
The staff was also sticking bubble routes on the backsides of other RPO concepts to give them multiple options on each play, something else to watch.
Glance RPOs
“Glance” RPOs are a little different than the previous two, as they are trying to punish a defense vertically instead of horizontally. Where the Arrow and Bubble RPOs were responding to how a defender widens or tightens to the formation on a play, Glance RPOs respond to a second-level players move closer or further away to the line of scrimmage.
Glance RPOs read second- or third-level players, typically an outside linebacker or apex-type player in single-high safety looks and the backside safety in two-high safety looks. If that read player gains depth at the snap or backpedals, the QB will hand the ball off to the back on the run as normal into a light box. If that player steps toward the line to play the run and makes it a disadvantageous box, the quarterback throws a short in-breaking route right into the space they vacated, called a “glance” route. The glance route is a 5-yard angle in cut, a little deeper than a slant route but not quite a post route.
Here’s a cut-up of the all the Glance RPOs NU ran in the spring game:
You can see the post routes from receivers turning and looking for the ball instead of blocking in most of these plays.
In the example below (which occurs at a little before the 0:30 mark in the video above, the read player is the strongside safety who is mugged up near the box near the apex, circled in purple:
If that player in the purple circle stays flat-footed or gains depth to drop into coverage, the quarterback would have a 6-on-6 blocker-to-defender box and should hand the ball off. If that read player were to come down at the snap to play the run, it would make it 6-on-7 blockers to defenders, and the the quarterback would pull the ball and hit that “glance” route behind the read player.
The read player blitzes, giving the defense the numbers advantage in the box, so the quarterback makes the correct read and pulls the ball to throw it.
You can see how this is a concept on early downs that can help NU just pick up some easier yardage: Instead of running into a disadvantageous box, a quick decision and a simple, low-risk pass gets Nebraska a first down instead.
Nebraska also ran a variation of this concept that didn’t use a glance route but instead did a double slants-type of look, but the general concept was the same.
Quick Game RPOs
These didn’t get as much run in the spring game, but the other variation used was a combination of quick hitch and slant routes attached to run plays:
These RPOs are more based on pre-snap alignments or leverage of defensive backs against receivers. If a corner is playing off coverage, the quarterback has the ability to throw it out fast and grab a handful of yards. The slants come into play if a DB lines up with outside leverage. The final rep in the above video, a corner lines up in press with no safety deep, and the receiver sight-adjusts to run a deep route in one-on-one coverage down the field that the quarterback takes. All just about getting advantageous matchups for pass routes.
In the example below (the second play of the video above) NU has a GF Counter run on with two quick hitches to the outside the quarterback can take if they get the corners playing off:
The quarterback hands off here, even though he gets off coverage and the DBs backpedalling at the snap:
It’s hard to know if it’s a misread without knowing Nebraska’s specific rules, but this is probably a play that should have been thrown out wide and picked up 5 or so yards.
Thanks, as always, for reading. If you’re just catching up before the season or new here, be sure to check out the archive, where you can read all of the newsletter’s offseason content. I’ve got one more piece coming before the season starts, then we’ve get the real action! GBR.
(beyond giving the ball away 31 times)
Before someone gets mad at me, I am not implying Nebraska was even playing the same sport last year as any of these offenses. Just in this one specific metric Nebraska kept pace or was marginally better.
This is a pretty common mistake for announcers, especially bad ones. About half of the plays they call RPOs are really just normal play-action passes. If you want to tell the difference, watch the whole line: on an RPO, the line is fully run blocking the whole time, moving forward. On a play-action pass, the line might have a puller, but some or the rest of the members will eventually get into a pass set, in which they’re moving backwards.
Honestly, a hilarious quote. I also want my offense to have five All-Pro skill players and the best playcaller in modern history. I, too, think that would be a good offense to run, Matt.
Who delivered some pretty insightful color commentary during the spring game from a scheme perspective? Go, Kenny!