I don't want to be one of those loonies you encounter in Southern football, who calls for the coach's job after one bad game, but, after the Kansas game, it's time to fire Paul Johnson at Georgia Tech. In this post, I want to break down why I'm done being emotionally invested in The Yellow Jackets for the time being. I'm sure I could be accused of fair weather fandom, but I'd like to point out that in our run-up to (and whoopin' in) the Orange Bowl, last year, I'd been making a lot of the same complaints. The problem is, against Kansas, these problems came home to roost.
While I think Tech also has some problems with discipline and focus, I want to discuss some technical problems that I see with the Yellow Jackets. Specifically, beyond lacking the ability to consistently execute the option, I think we have some serious problems with play-calling and scheme design. I will use Chris Brown's constraint theory of offense, which states that a team should build an offense around a small collection of base plays to run well against regular defenses and then add in more exotic stuff to run when the defense starts to react, to show that Paul Johnsons' scheme for the Tech offense is problematic in that it places undue stress on our ability to execute well when things go poorly.
How the Option Works
As you probably know, Johnson's offense is based around a series of play concepts called, collectively, "the option." An option offense is a running based attack that gives the quarterback various options (hence the name) as to who to give the ball to. Here's a diagram of the classic option play, called "veer," to show how it works:
This play, in the diagram above, is run out of the I formation, in which two running backs (FB and HB above) line up in a line behind the quarterback, a tight end (labelled TE above) lines up on one end of the offensive line (the five players "T G C G T" above), with a flanker to the right (W) and a split end to the left (W). This formation is more specifically associated with the option as it was run during the 80s and 90s at Nebraska by Tom Osborne.
In veer, and other option plays, the offensive line and tight end all block various defensive players but leave one defender unblocked, in this case one of the defensive ends (DE, highlighted above). At the point labelled "Read 1," with a red line pointing towards the DE being read, the quarterback places the football in the arms of fullback (FB) who is running towards the DE being read. If the DE retreats or hesitates as a result of the hand off, the quarterback gives the ball to the fullback and the fullback probably runs for 50 yards (not always). If, however, the DE drops down to tackle the fullback (which is called "crashing" in option football), the quarterback takes the ball back out of the fullback's arms and runs down the line towards the tight end. The fullback, then, blocks the crashing DE. That is the first of two reads in veer.
Now, running towards the edge of the line, the quarterback begins to cut up field somewhere between the tight end's right shoulder and the sideline. While the quarterback is doing all of this, the halfback (HB) has been mirroring the quarterback's movements but somewhere 5-7 yards behind him. When the quarterback cuts, the halfback will be a yard or two behind the quarterback and 5-7 yards closer to the sideline. This is called being in a "pitch relationship". When the quarterback cuts upfield, he makes his second read, in this case on the strong safety (SS and highlighted). If the safety breaks to hit the quarterback, the quarterback will pitch the ball to the halfback who will continue to run upfield. If the safety instead moves to cover the halfback (also, at this point, called the "pitch man"), the quarterback keeps the ball himself and runs upfield.
That is the second of the two reads in veer. As you can see, the quarterback has three options: the fullback dive, the halfback pitch, and the quarterback keep. This is why, sometimes, an option offense will be referred to as "the triple option." Here's a video of Nesbitt keeping the ball in a veer play against Wake Forest. Anthony Allen (18) is the pitch man and Jonathan Dwyer (21) is the fullback:
The name "triple option" is a misnomer, however, as there are a few other option plays in an option offense (some with only one read), for instance: the midline option, the speed option, and the counter option. So, really, the option is a whole set of plays.
Tech's Option
While we saw above how a veer option play works, we saw it run out of an I formation, a la Nebraska's old option offense. At Tech, we run something called the "flexbone option". This is because, originally, the option was run, mostly, out of a formation called the "wishbone", which featured a seven man line and three running backs in the backfield with the quarterback. This was the offense that Texas and Alabama won a lot of national championships with, but, by the 1980s, it was becoming apparent that an all running offense wasn't going to work as well, anymore. That led Fischer DeBerry, the option guru at Air Force for a number of years, to declare the need for "more flexibility from their wishbone," which led to the birth of the option system Tech runs today.
Take a look at this diagram of how Tech runs veer:
In the flexbone, the offense usually lines up in formation called "double slot," in which there are four wide receivers, two on the ends of the line (called split ends, again) and two off the line (the two off the line are the two players labelled "W" above who are closer to the tackles, labelled "T") called slot backs. In the flexbone, the pitch man, instead of lining up behind the quarterback, as we saw above, is a slot back who goes in motion behind the fullback to get into a position similar to where the half back would have lined up above. The ball is then snapped to the quarterback when the pitch man is in position. This gives the pitch man a running start (which can be an advantage) and nominally forces the defense to spread out more (instead of covering the middle of the field), but we'll talk more about that in a second.
In any case, after the ball is snapped, the play is run the same as it was out of the I formation above.
As you can see, the option is actually a set of plays and can be run out of any number of formations. Oregon and Florida run these plays out of shotgun formations. Air Force uses the shotgun, the flexbone, and the I to run its multi-faceted option attack. The key thing is that, in any option play, the offense leaves a player unblocked (both the DE and the SS in veer) and lets the quarterback read his reaction to take him out of the play and, therefore, gain a numbers advantage over the defense.
Having said that there are a variety of formations from which to run option, I don't want to suggest that formation choice doesn't matter in the option. In fact, it makes huge differences but not how you may think.
The Constraint Theory of Offense
As I mentioned above, Chris Brown's constraint theory of offense stresses building an offense around a group of solid, "run anytime" plays. In the case of Paul Johnson, these are the option plays we discussed above. What separates a good option offense from a great option offense is in the kind of plays the offense runs when the defense starts to load up to stop the base plays.
For instance, an effective way to counter the option might be to start moving the cornerbacks into the pitch lane (off the wide receivers in the flexbone diagram). In the veer play, the split ends don't actually do anything of value (except block the cornberbacks, of course). Moving the corners in, on the part of the defense, will give an advantage and will necessitate the offense running some kind of play to make sure the corners stay outside and cover the split ends. Such plays are called "constraint plays," in Chris Brown's terminology.
For instance, here is a play that Nebraska liked to use as a counter to its option, if the defense started to play the option too hard:
This play is called outside zone and, without getting into too much detail, it is a type of power run play in which the offensive line and tight end block together as a unit to counter the front of the defense. The fullback then runs in front of the half back (who receives the ball via a hand off from the quarterback) between the tight end and the sideline (this is called "off tackle" running). It's a basic football play for a lot of offenses, and, in fact, some NFL teams (like the Texans and the Colts) run almost nothing but outside and inside zone. Here's a video of USC running outside zone out of an offset I formation:
As an additional constraint, the ability to run outside zone and other I formation run plays opens up the ability to run simple playaction passes off of fake runs:
This play, run off of a fake outside zone play has the flanker running a go route up the sideline, to try and pull the cornerback away from the line and the full back running into the flat (the space vacated by the corner). If the corner doesn't follow the flanker, the quarterback throws the ball to him. If he does follow, the quarterback throws to the fullback. It's a simple 1-2 read and the QB is also running out of the pocket and could run up field if neither route is there.
In essence, these kinds of passes and runs provide simple extensions of what the offense already does well (each play basically has the offensive line run-blocking), with a few small changes.
Here's another example of a constraint play in an option based offense:
This play is from Rich Ellerson's offense at Army and was used to good effect a number of times in their game last weekend against Hawai'i. This play is called a fly sweep and was developed to run out of an offense called the Wing-T. Note that in Ellerson's offense (which is partly based on Paul Johnson's, as the two coached together at Hawai'i), there's a subtle difference from Johnson's flexbone option: namely, Ellerson lines two backs up in the backfield, instead of having four receivers. Despite being a small difference, this allows a lot of variation in the offense. For instance, Ellerson can motion the slot back down the line of scrimmage and still run some kind of option with the two backs (in other words the motion doesn't set up the option). Often, in fact, the slot back will motion to the other side of the formation and stop before the ball is even snapped.
Every few plays, when the defense has started ignoring the motioning slot back, the quarterback will, instead of running option, snap the ball while the slot back is still in motion and quickly hand the ball off, as the slot back runs down the line of scrimmage. Here's a video of the play being run out of a double wing formation (pay close attention, the play happens fast):
As with the two constraint plays off of the I formation, the fly sweep allows Army to take advantage of defensive negligence without having to worry about doing anything radically different (the offensive line only has to change who they are blocking). In fact, Tech has recently begun running a play called "rocket sweep" that attempts to do something very similar to this play (this video features University of Charleston running the play, but it's the same offense):
In both cases, we see option offenses that can simply adapt to changing defensive requirements and still shred their opponents for big yardage gains. Additionally, these constraint plays force defenses to respect all 11 players on the field and forces them back into base alignments that can be attacked from the option.
Tech's Constraints, or Where the Problems Are
The problem that Tech has, with running constraints, lies in the type of plays normally suggested by our formation. As we saw above, Ellerson's Wing-T formation and Osborne's I formation suggest certain plays. I could also talk, for instance, about how Florida's shotgun option sets up bubble screen and certain easy downfield passes as constraint plays; the point is, all option formations suggest natural constraint plays. With Tech's flexbone formation, outside of the rocket sweep, the only constraint plays available are radically different from our base plays.
As I mentioned above, the formation Tech runs out of is called the double slot. This formation became prominent as part of an offense that dominated college and pro record books in the 1990s: the run and shoot. Here's some video of run and shoot passing from 1990 (the Houston Cougars are running the offense in the video):
Note the formation: except for the running back being deeper, it's Tech's offense. This is by design. Paul Johnson has been forthright about the influence a visit to the Houston Gamblers (a USFL team that used the run and shoot offense) had on the development of his offense. The flexbone does allow for the introduction of run and shoot plays as constraints, but the transition isn't easy. Here's a digram of how Tech might run one of the base concepts of the run and shoot, a play called "choice":
In this play, the pitch man, runs a shoot route below the cornerback on the left of the formation. The other slot back runs an option route that attacks the middle linebacker and the strong safety. The left split end runs a go route that attacks strong safety and corner. You can read more about this play here, if you're interested.
In fact, here's a video of someone running choice out of our offense in one of the NCAA Football video games:
The quarterback runs what's called a half roll and based on how the defense reacts to the various routes being run, throws to the receiver with the best chance of being open.
Football teams like SMU and Hawai'i still run this kind of attack (although, changed slightly to incorporate shotgun and pistol formations and different reads on the option routes). In fact, the New England Patriots also run a lot of these concepts.
The problem for Tech, though, is that most of these types of plays end with Nesbitt throwing incomplete, Nesbitt getting sacked, or Nesbitt throwing out of bounds/missing an open receiver. Why? Because the run and shoot stresses football concepts that are entirely different from option football. As you can see in the diagram above, the offensive line is pass blocking, instead of run blocking, which is an entirely different set of skills to master (especially with the kind of specialized run blocking, called cut blocking, done by Tech's line). Additionally, the vertical nature of the run and shoot puts a lot of stress (in terms of long, quick throws) on a quarterback who spends most of his time being essentially a third running back. In other words, for Tech's constraint plays to work, the team has to become some entirely different team.
While this isn't necessarily a bad thing, in the UGA game last year, the bowl loss to Iowa, and Saturday against Kansas, we couldn't hit our run and shoot passes when those three opponents moved defensively to take away our option game. You can see the same thing happening to Navy (who runs the Paul Johnson option), who has had a number of low scoring games this season and seemed to struggle on offense. Teams are beginning to realize (thanks to the national TV attention that Tech has received) that our option is easy to shut down, because we can't hit the plays we need to in order to keep the offense honest.
Where other variation of the option feature simple, low stress plays as constraints for when the defense gets too focused on the option, Paul Johnson's insistence on running run and shoot passing plays takes our offense even further out of its comfort zone and demands that our players become good at things they don't practice regularly (and frankly, why would they?). Essentially, Johnson's scheme asks that our offense be two offenses: a run based option attack when things are going well and a drop-back passing based attack when they aren't. This is not a recipe for success, as it forces our players to get really, really good at two non-overlapping skill sets. Additionally, it's important to note that the run and shoot is incredibly specialized, focused, and difficult-to-master passing option; it's not at all like a standard passing offense (although a lot of NFL teams use run and shoot plays now). So basically Tech needs to be good at an incredibly specialized rushing offense and an incredibly specialized passing offense!
Conclusion
As I said, I don't want to sound like a fair weather fan. I've been concerned with our scheme since last season, at least since when, down to UGA, Paul Johnson decided to ignore the rocket sweep (which had been working all day) and focus on endless iterations of the counter option for loss of yards and incomplete passes by Nesbitt. By focusing on a difficult to implement offense (run and shoot guru June Jones says that it takes two years for him to install the offense when he arrives at a new program) as our constraint plays, we only increase our inability to execute when the defense takes away our base plays.
While I don't necessarily want to suggest that Johnson can't turn it around at Tech, the scheme we run now is broken and I question his desire to change it. Johnson's rhetoric in press conferences is often one in which in a win, the scheme succeeded and in a loss, the players failed to execute the scheme. In other words, it's never the scheme's fault when things go bad. This mindset is also what got us in trouble when Chan Gailey refused to see that Reggie Ball was not working out at quarterback. Lack of adaptability will get you in trouble very fast, in football.
I want to make clear that I am not at all opposed to the option; I was excited when Johnson was hired because the option is a great scheme for a school that has built-in recruiting difficulties like Tech. The option can and does work to great effect throughout college football, but at Tech, Johnson needs to rethink his constraint plays and install more stuff like the rocket sweep into the playbook or use more sprint-out play action passing in order to allow our offensive line to focus on run blocking and not have to suddenly convert into an entirely different unit when things start to go south. While Tech players are some of the smartest in the country, Johnson's use of the run and shoot as a constraint offense puts too much pressure on our players, even by Tech standards.