Baylor vs. ULM Preview Part I: The 3-3-5

When I sat down Tuesday afternoon to begin putting together tomorrow's preview/prediction thread, I realized that there was an aspect of the game Friday night that might necessitate splitting up that post: ULM's defense. Because the Warhawks run a scheme that is relatively unfamiliar to fans of the Big XII (those outside of TCU are probably not too familiar in the 3-3-5), I decided to go ahead and split the post for the first time. This is now Part I, focusing entirely on ULM's defense and our offense. Part II switching sides (our defense and their offense) and tying things together will go up this afternoon. This could be a little strange in that I don't intend to make any predictions here. I'll save that for later. I'm just talking about their defense, what it wants to do, and how we have to respond.

In the interest of full disclosure, most of what I've written here is based off the excellent post I linked earlier today on the subject from the SB Nation blog from ULM's last opponent, the Auburn Tigers, and the sources linked therein. I also found several extremely good references that I've identified in case you'd like more information. For last year's ULM stats, I pulled mostly from FootballOutsiders' rankings and FootballStudyHall's statistical profile for ULM before this year. Hit the jump.

What the 3-3-5 is:

For those unfamiliar with defensive nomenclature, the 3-3-5 is an alignment with 3 DLs, 3 LBs, and 5 DBs.* Pioneered by a coach named Joe Lee Dunn, who has a series of Youtube videos on the subject that I highly recommend you watch if you're looking for more in-depth analysis, the 3-3-5 was designed to allow teams with less talent on defense overcome their deficit through speed. In his book, The Essential Smart Football, Chris Brown talked about the defense at length in terms much easier to understand than those from Mr. Dunn himself. Thankfully, Grantland posted a large sample of Brown's work in which he discussed South Carolina's decision to go to the 3-3-5 in 2000 under then-DC Charlie Strong. That article formed much of the basis for the Auburn post I so enjoyed.

*For the more visually-inclined, Brown included a helpful diagram of a typical 3-3-5 play that shows how and where players line up relative to the offense. This is just one example, so look more to where the players are than what they are doing.

The first key to understanding the 3-3-5 is knowing that not all DBs are created equal. The 3-3-5 is not a Dime formation designed to stop the pass by sheer numbers in coverage, though it can do that if necessary. According to Brown:

[The 5 defensive backs] include three in the traditional mold and two hybrid strong safeties/outside linebackers that can patrol the flats, blitz, stop the run or even cover receivers or tight ends in man coverage. This is just one of several ways the 3-3-5 contrasted with more traditional defensive sets, like the 4-3 (four down linemen and three linebackers) or 3-4 (three down linemen and three linebackers).

Thus, the defense basically uses two hybrid defenders a la Ahmad Dixon to emphasize flexibility; they can hang back in a more traditional role in pass coverage or bring an extra man in run support or the pass rush. For ULM these two players are Cordero Smith and Isaiah Newsome, and their ability to fill multiple roles makes the defense extremely interesting. They allow ULM to play three linebackers and two hybrid DBs all capable of blitzing or covering on any given play. (For a more in-depth discussion of the coverage options available to a 3-3-5, look here).

Ahead of that, an offense sees three pairs of defenders that can shade whichever way necessary; they aren't anchored heads-up on the center and tackles like in a conventional 3-4. The reason for this is gap responsibility. Linemen in a 3-4 defense are typically responsible for the gaps on their left and right, while a 3-3-5 assigns each a single gap and brings at least one LB on a stunt or blitz to take the fourth.* You very rarely see only the three down linemen rushing unless the defense is absolutely sure the offense will pass and has fallen back. *NOTE: Just like in any other one-gap defense, the defender is not necessarily responsible for the gap directly in front of him.)

The idea behind giving each each player only one gap is to allow for maximum upfield penetration. This scheme lets players "shoot the gap" in an assigned direction and disrupt not only the blocking schemes designed to stop them but also the offensive play itself, be it a run or pass. Brown notes that this is often much simpler for defensive players that only have to identify their gap and then go. For this reason, it's a defense known primarily for speed and aggressiveness with the possibility of multiple players breaking into the backfield on any given play. Brown also notes, however, the main problem of this "guess and go" attack: you have to guess correctly. The defense has to literally play the percentages and identify offensive tendencies in each formation to try determine where an offense will go. If, for example, the defense overloads one side only to find the offense going the other way, you can have a serious problem. If things go well, however, you put a ton of pressure on the opposing offensive line and quarterback to diagnosis your intentions and adjust accordingly.

Why does ULM run it?

The answer is almost certainly recruiting. As a smaller school outside the BCS framework and virtually bereft of notable history, ULM's difficulties in recruiting are easy to understand. And as I've constantly maintained, the single hardest position group to recruit in college football is defensive lineman. Combine that with the fact that linebackers who are big and fast are hard to find while those who may be one or the other aren't, and you have a situation basically begging for a defense utilizing fewer down linemen and linebackers relying on speed and aggressiveness. Personnel-wise, that's exactly what the 3-3-5 does. Just like Mike Leach's spread offense at Texas Tech, it allows you to overcome a relative lack of talent through speed. That's why we're using some 3-3-5 looks this year. You can scheme aggressively. You can't scheme to make your players bigger or stronger than they are. The same goes in the secondary; as I said above, "hybrid" often means undersized, so if you have undersized quick linebackers it's easy to turn them into hybrid safeties. Or vice versa with bigger, slower safeties.

You see this on ULM's depth chart at the linebacker and safety positions. The aforementioned "hybrid" safeties Smith and Newsome are listed at 5-10, 190 and 5-11, 205, respectively. These are not linebackers, but they're apparently not true safeties, either. Following the trend, ULM is also relatively small at linebacker, with DaCorris Ford (6-1, 223), R.J. Young (5-11, 240), and Cameron Blakes (6-1, 223) starting across the board. These are not bruisers in the mold of Alabama or LSU. They're actually much more akin to what we have seen in recent years from guys like Elliot Coffey, Brody Trahan, and Eddie Lackey.

What does this mean for us?

Our offensive line has to step up their game in pass protection considerably. I've now watched both of ULM's prior games and seen everything said above born out in practice. On almost every play, ULM brings at least 4 rushers, often 5 or more, to create havoc in the backfield. They blitzed, blitzed, and then blitzed some more. We should expect them to do exactly the same thing again this week. Against Arkansas in the first half, it actually didn't work out so well because the offensive line protected Wilson well enough to get the ball away and Arkansas' receivers were able to get open in the spaces vacated. ULM's Vincent Eddie and Rob'Donovan Lewis, combined with deep safety Mitch Lane, were unable to fill the holes vacated by blitzers. To his credit, Wilson reacted quickly to find those holes and the receivers streaking through them. It's easy to forget that when Wilson left the game, his team was up 28-7. They had dominated on offense.

After he went down, however, his backup was not nearly so proficient and offense sputtered and stalled. That gave ULM the time it needed to get back into the game, but had Wilson never gone out, the result likely wouldn't have been close. The key for Baylor will be two-fold: Nick Florence has to make quick, accurate decisions (an area where he excels), and the offensive line has to protect him from a myriad of blitzing rushers. We can't let the same fate befall Nick Florence as did Tyler Wilson. Luckily, our coaches probably know all this far, far better than you or I and will likely give Florence the same opportunity Wilson had to dissect the aggressive Warhawks defense. There's no reason why, if we keep him from getting killed, he can't do it.

Moreover, in racking up 255 yards rushing last week, Auburn proved that you don't have to beat ULM through the air. A 3-3-5 defense, though it often has numerous additional defenders in the box, can be susceptible to a power running game. Auburn used that, as well as ULM's aggressiveness, to their advantage, by allowing ULM to rush their DEs out of the play on the outside while trapping the nose and blitzing linebacker inside. In doing so, Auburn averaged 6.1 yards per rush.

Something I'm left wondering about through all this is the effect that offensive pace and the fact that we can run so many different plays out of the same formation will have on ULM's defense. Like I said above, the 3-3-5 relies on the opposing defense correctly predicting the opponent's offensive intentions. If we can disguise our sufficiently while also not giving ULM the opportunity to huddle, we could make their lives a lot more difficult on defense. I also have to believe that our offensive line can take advantage of their tremendous edge in size to enforce their will on ULM's defensive line.

Of course, the best news of all (and I'm getting out of my original premise here) is that according to FootballOutsiders, we may not have to work all that hard against their defense to begin with. According to FEI, ULM has the 76th-ranked defense in the country this year in terms of efficiency. It is their offense, ranked 24th, that has carried their water. S&P+, which has yet to update from this past week's games and should be taken with a grain of salt for that reason, likes them even less in 2012, particularly against the rush. Last season, in a much larger sample size, it gave lower marks to their pass defense. They suffered in particular on third downs and in obvious passing situations. Those things are probably related. Overall, ULM ranked below average in almost every major category despite looking good superficially by yardage alone.