Defense in the EagleBank Bowl

Bumped. GO BRUINS. -N

Happy New Year everyone. After a 2nd viewing of the EagleBank bowl I wrote this up. Like everyone else I was curious as to the "adjustments" made by the coaches at halftime vs Temple. So I imported the UCLA defensive snaps off of the play-by-play data into my scouting spreadsheet as well as some info from the tape.

From watching the game live, I noticed that we tended to stay in either "quarters" or "sky" coverage throughout most of the game (hard to tell with the camera angles). Generally that means we had a safety rolling down looking to play run support for at least 8 in the box. Sky is meant to get 8 in the box with a safety rotating down into the box and 3-deep behind it (although there are many ways you can do it). Quarters is a flexible coverage that can get you a 9-man front with the ability to also cover 4 vertical routes; the safeties are run-first players and only bail to deep coverage if their key goes vertical.

One change I saw in the second half is that we ditched quarters (or what looked like it - TV angles were tough to tell). It's my favorite coverage, but it is very practice-intensive as players need to "pattern read". For a clinic on that watch Alabama in the BCS Title game as Nick Saban is a guru.

I think that the deep pass down the middle on the first drive really set the tone defensively for us. We gave up the deep pass early on the first drive on quarters, a pretty aggressive call. You're not sending extra guys across the line but you are still playing with fire, looking to stop the run with 9 in the box which is a sensible call vs Temple. Safeties make it work as they have to read #2 (2nd man from the outside) The safeties both were playing run, both stepped up due to the playaction, as both were reading the TEs and Temple kept both their TEs in to block. They released the back on a wheel to the left which Rahim Moore picked up, while Anthony Dye, the other safety, dropped down to cover a crossing route, leaving the middle wide open. Quarters became an invert-2, with a post right down the middle - a bad situation if you have two deep safeties, even worse if you only have two deep corners.

I have no way of knowing what each guy was supposed to do, but if I'm coaching the D and we run quarters I would like Moore at free safety to look at #1 on the post and pick off that pass rather than chasing the back on the wheel - that should be the Sam backer's (Ayers) responsibility to carry #3 out of the backfield and wall him off until help gets there. Dye does what he should've done and picked up the crossing route; once that receiver declares that he is not going vertical then the CB should be looking at deep half, which he does, and he (Price) eventually makes the tackle.

Other than this breakdown our first drive was not all that bad. The TD pass was against a pretty generic cover 2 with a stunt up front, they just threw a corner route to the TE in the space between the corner and the safety, ran a basic smash concept which is perfect against that look. I think that giving up that big gain early might have made us play a little more tentatively.

Charts and more after the jump.

Here is our defensive performance for the first half vs the second half. I messed up and reversed them...second half is on the left and first half is on the right. Temple ran way more offensive plays in the first half.

The biggest difference is that those "under 4" and "over 4" (plays between 0-10 yards) were a lot more common in the first half - Temple was not able to move the ball consistently. We also cut down on the big plays of 10 yards or more, just three long runs in the 2nd half. In the first half we gave up four plays of 20+ and five between 10 and 20, which will kill a defense.

People like to talk about blitzing, so here are the blitzing numbers for the 1st and 2nd half. I defined blitzing as anytime we rushed a non-down lineman (hard to tell vs. the run but I just eyeballed it).

This time first half is left and second half is right. A lot more defensive snaps in the first half, and a lot more plays where we did not send anyone. The key is the number of guys we sent - between 3 to 6. In the 2nd half it was almost even - we were fairly split between sending 4 and rushing an extra guy, usually a linebacker (although vs the run it was kind of hard to tell - possible our LBs were just really quick to react to the run).

Above is the chart of the average gain vs no blitz, a stunt, and a blitz. As you can see....about 3 YPP when we rushed an extra man, vs about 4.5 YPP when we did not. There were a few big gains vs DL stunts, including the first Temple TD. I defined stunt as a 4-man rush where we swapped gap assignments by DL.

So far it looks like blitzing is a good choice, but this chart shows the average "yards to go" after a play: the difference result of the play and the yards they needed for a first. The lower this number, the better. For example, if it is 3rd and 10 and they pick up 5 yards then this value will be -5; if they pick up 7 yards then the value will be 2. I think this provides a better picture than just the raw yards gained per play since a 6 yard gain on 3rd-and-10 is probably not as successful as a 3 yard gain on 3rd-and-1. Defensively, I am more concerned with this number rather than the total yardage given up. The rest of the charts use this "yards to go" rather than the raw yardage result - various criteria of defensive calls crossed with the average yards they had to go after the play. Also keep in mind that this is only data from about 65 snaps so it can be easily skewed by a few big plays.

In this case above the "yards to go" is about the same whether we blitzed or not. So, on the average, when we did not blitz, they picked up 4.5, but they were still about 4.5 short. When we did, they picked up about 3, but they still had about 4 to go.

Above is the "yards to go" for pass/run and blitzing. As you can see; we did better when we blitzed against the pass, not so much against the run - blitzing the run did not work out that well. Stunting numbers are inflated by that TD pass, I believe.

Above is the same data, only coverage and pass/run by the yards to go. Quarters gave up that big pass play I wrote about earlier. Against the pass the 3-deep sky seemed to be preferred and seemed to work well. Against the run a basic zone (we ran a lot of cover 2) worked well while man did not.

This final chart is the number of guys we sent separated by run/pass, cross-tabulated with the yards to go average. As you can tell, sending 5 or 6 guys against the pass worked well, but against the run it was the opposite.

What Does This All Mean?

Yes, we did blitz a lot more in the second half and sent more guys across the line by (what looked like) design. It worked a lot better against the pass. I think that's about all you can really generalize. Coach Bullough gets a lot of flak for being conservative, but I wouldn't consider a lack of blitzing or sending extra guys as being conservative - there's other ways you can be aggressive, ie, running the 9-man front early on the game to stop the run. Sometimes it works and sometimes you get burnt down the middle. You can run press man, get everyone up within 10 yards of the LOS, which I would consider being pretty aggressive, and again we struggled in the run. I don't associate being aggressive on defense with only blitzing or running line stunts, you can also be aggressive in the secondary and I think we did a good job with that in terms of defensive calls throughout the entire game.

As far as blitzing goes we were good at timing them up especially in the second half for success. It did not really pay off in the first. I think one thing that may have thrown us off was Temple's playcalling out of the gate - we might have been caught off-guard by them, as we seemed to be playing run pretty hard on the first drive and seemed shocked to see Temple passing. After that things settled down. The big play on the second drive was also a pass where we had a blown coverage assignment by the nickleback...looked like he did not carry #2 down the middle. The third TD was set up by KP's pick and the facemask call. When we did bring extra guys across it seemed to really hurt us against the run - Temple liked that inside zone and cut back for good yardage when we tried to bring extra pressure. When we pressed the DBs into the box, or ran quarters, they did not make plays. "Putting 8 in the box to stop the run" is tossed around a lot as a catch-all solution but it makes no difference if your safeties are running themselves out of the play every time. One advantage to sitting back and reading vs. the run is that you are not engaged until you figure out where the ball is going - if you are up on the LOS it is very easy to run yourself out of the play.

I was in the game thread and we were talking about some of the poor safety play in the first half - as far as the magical adjustment the coaches made at halftime, I think it had less to do with anything schematic or playcalling-wise but just getting our safeties to settle down and make plays. You saw the images above of Rahim Moore vacating the deep middle to chase the wheel leading to a long gain on the opening drive; when we brought up the safeties throughout the first half we also had problems in run support - they were in position but did not make tackles, or over pursued. If you want to play aggressive you need great play from the safeties since they are your flex players who can be moved all over the field. If you are going to run the eagle front with the LBs shifted over then and playing aggressively downhill then your safety on the backside has to be disciplined and play cutback; if you want them playing run first then they have to make the correct reads in order to get back into coverage if they get a pass read.

I have no problem with our defensive playcalling - if that's the system he knows and understands then I see no reason to change it. The best scheme is the one you know and can teach the best. However my biggest criticisms of defense continue to be missed tackles, guys running themselves out of the play, guys taking bad angles, and missed coverage assignments. Just basic stuff. The schemes will work, players have the opportunity to be in position to make plays, but the breakdowns are happening when they do not make the plays.

As a DC, I have no issue with Bullouogh's calls - playcalling is probably like 15% of coaching. The real meat of the profession is whether you can 1) teach basic fundamentals, 2) teach your scheme, 3) get your players to execute your scheme, and those areas are where I think we have been lacking.