The defense against Oregon was a travesty. The team underperformed in every facet, looking confused, out of position, and vanilla the whole time. Ryan Day took drastic action and altered coaching responsibilities on the defensive side, including moving Kerry Coombs to the booth, moving Matt Barnes to the field, giving Matt Barnes the play calling, and giving Larry Johnson more responsibility in the game plan. While some have said he is panicking, I think, given the sheer level of incompetence the Buckeye defense showed against Oregon, he had no choice. Guys weren’t filling gaps correctly, covering ineligible receivers, and otherwise failing on a very basic level. So, did they get better against Tulsa?

Before we get to the Play-by-Play, just a real quick reference to last week’s article here. Kyle Jones did a schematic breakdown of the things that changed against Tulsa here, and I would recommend reading it before reading this if you haven’t already, as he is much better at explaining these defensive coverages and changes than I am. Additionally, this series is borrowing the format from Mgoblog, whose latest post is here, though this series is changing every time I do it to better match what I feel is relevant.

Here is the play-by-play chart, which, as always, you are welcome to skip:

After the Oregon game, most people pointed out clear issues with the defense. The ones I saw most were:

More varied formations and disguised looks
More pressure
Less Rotation

My specific takeaways were similar, but additionally, I wanted to see:

More playing time for potential 1-techs like Tyleik Williams and Ty Hamilton
More playing time for “wildcards” like Steele Chambers and K’Vaughn Pope
Better understanding of responsibilities during the play (fundamentals)

So, how did the coaches respond? Well, they basically did everything on that list, though the effectiveness of these changes varied.

The first issue was the simplicity of formations and the lack of disguising looks against Oregon. As I noted in that analysis, the Buckeyes only ran a 4-2-5 and 4-3 against Oregon, and the vast majority of the time they ran a 4-2-5. Additionally, they only ran with 1 high safety the entire game. When they did line up with 2 high safeties, it was just to “disguise” going back to 1 high safety, which was always Bryson Shaw, while Ronnie Hickman always moved downward.

In the Tulsa game, they experimented with multiple different formations, including the 3-3-5, the 3-1-7, the 3-2-6, and the 2-4-5, though the 2-4-5 was primarily a variation of the 4-2-5 with the two defensive ends standing up and spreading out like outside linebackers in a 3-4. Still, all of the new formations were welcome with the lack of formational diversity in the previous game.

Furthermore, they ran plays out of 2 high safeties more often this game, and when they did use it to disguise a 1 high, sometimes Hickman would be the high safety, while Shaw would move into the box. The predictability of the defense decreased dramatically. While I was grading the first game, I didn’t have to pay any attention to Ohio State’s actual alignment on any given play, as it was the same every time. I still did pay attention, but if I had just assumed it was a 4-2-5 every play, I wouldn’t have been wrong that often. Additionally, they would roll the safeties like they did in 2019, so if a man went across the formation, the cover safety would roll up to the deep safety spot, and the deep safety would roll down into the cover safety spot. They lined up differently in trips than they did against Oregon as well. Instead of having the boundary corner move over to the field side with the other corner, the deep safety would move to that side, and the boundary corner would stay on the boundary. The Bullet would then move to deep safety. This helped significantly with weakside runs.

Also, as Kyle Jones brought up in his column, Ohio State started sprinkling in some Cover 2 occasionally, which, while not a very good defense against the run, as some utility against an air raid team like Tulsa. They also added a couple of new pressures: a double A gap blitz with the two linebackers (usually Eichenberg and Simon in this game), and a five man pressure with a lineman dropping to the flat (Smith and Cowan in this game). This also addressed the pressure concern that people had, as Ohio State had their best game yet when it came to pressuring the quarterback, though there is still lots of work to be done there.

There was also a call for much less of a rotation, which they sort of addressed. They still have a rotation, but I think the new rotational approach makes more sense. So for Minnesota and Oregon, the defensive coaches would sub out the linebackers in shifts, so that, for example, one series Simon and Eichenberg would play, but the next series Chambers and Pope would play, and the series after that would give playing time to Mitchel and Simon, with Gant and Gaoteote coming in on particular packages. This type of rotation doesn’t make a whole lot of sense, because none of the linebackers really got enough experience to learn from their mistakes quickly, so one shift of linebackers would make the same mistake as the previous shift in linebackers. This got messy quickly, as no one got any better, and guys who would normally would not be on the two-deep were getting minutes in big moments. I addressed this in greater depth in my last defensive analysis, but suffice to say, the rotation hurt the linebackers.

Against Tulsa, while there was still a semblance of a rotation, they were all situational rotations. For example, in the base set, Eichenberg and Simon never came off the field. I imagine when Mitchell comes back, this will be Mitchell and Simon. However, in passing downs, Pope would switch for Eichenberg, which makes sense since Pope, previous to this game, had performed better in coverage than most of the other linebackers. Then, in dime packages, Steel Chambers and Pope would be in the game, with Chambers being the guy if they were only using one linebacker.

The reason why situational substitution makes sense for the linebacker rotation, is that each linebacker can focus on a much more narrow role. They don’t have to learn all of the linebacker responsibilities in the world, just the ones that they will be playing with. This might end up being an advantage in a season where you need rapid improvement quickly. By limiting what each linebacker is asked to do, you can hopefully get a bunch of specialists, who are really good at particular things, even if they don’t know how to do much else. While this isn’t ideal for an every down linebacker, in a season that started with none of the linebackers understanding which gaps they needed to fill or which zones they needed to cover, you might be able to salvage some decent play out of the unit by shortening the list of what each linebacker is asked to learn over the course of the rest of the season.

Also, the rotations seem to better fit the skillsets of the players. Cody Simon is capable of being an every down linebacker, so he rarely leaves the field. That makes sense. Eichenberg struggles in pass protection, but has a big body and can hit hard? Put him in running situations. Pope seems to miss his gaps, but is always in the right place in the passing game? Put him in passing situations. Chambers is supremely athletic and goes hard every play, but is new to the position? Put him as a QB spy in the dime package with occasional zones. In one week, the coaches seem to have done a much better job with utilizing each player’s strengths, which is not something I would have said at all last week.

As for my requests, the biggest one I had was trying to find a good 1-tech at the defensive tackle spot, as, although Haskell Garrett is dynamic at the 3-tech spot, he got eaten up by double teams at the 1-tech due to his smaller size. I suggested Tyleik Williams, and to a lesser extent, Ty Hamilton, as I thought both guys flashed in the Oregon game in the 3 or 4 snaps they played each.

It looks like the coaches saw exactly what I did, because both of those guys saw a big increase in playing time against Tulsa. Tyleik Williams was the big surprise on defense, and genuinely dominated the Tulsa offensive line. I am hoping that these guys see more playing time down the road, as they both have good size. Williams in particular looks like the prototypical 1-tech, and based on the way he played against Tulsa, I think he has earned more playing time.

I also liked the increased time for guys like K’Vaughn Pope and Steele Chambers in their new roles. I don’t think either one of these guys is completely ready, but they create chaos on the field, and it can be effective at times. I think their insertion into the game against Oregon in the late third quarter is one of the things that helped the defense turn things around, because for the only time in the game, Anthony Brown didn’t know what to expect. As I said then, Pope and Chambers may not have been doing what they were supposed to, but it was usually more effective than the guys who did know what they were supposed to do.

Finally, I just wanted to see Ohio State get better at the basics. If you run Cover 1 and Cover 3 to be good against the run because every gap is accounted for, you better be getting to your gaps and accounting for your gaps. The fact that players seemingly never knew which gaps to hit was probably the biggest indictments on the coaching staff all of this year.

Frankly, they looked way better. I know Tulsa isn’t Oregon, but the players for Ohio State flew to their gaps quickly and attacked downhill whenever there was a run. They looked confident and well-coached in the run game, even though it was only a week’s worth of changes. The pass defense didn’t look great, but even there I saw improvements among the linebackers in coverage. It looked to me like players still make mistakes in Cover-2 and don’t always have a great feel for Cover-3, but I don’t think it’s too alarming. They were better in Cover-3 at getting to their zones than against Oregon, and they have only been running Cover-2 for presumably a week. They can still get better with these. They were far more committed during RPOs and forced the quarterback to make tough throws to well-covered (usually) receivers. This was a significant improvement over last week. Barnes seems to have them on the right track.

Except, Tulsa still had over 500 yards of offense, and they still scored 20 points. How can this be better than Oregon, when Oregon was a significantly worse team? Let’s look at the individual scores.

The first number that will pop out is the extremely positive opinion I have about the defensive line. As I have noted, I don’t think the pressure was all that great, but they played very well against the run. Tyreke Smith is the only defensive lineman who can generate a consistent pass rush. I just have a feeling that he will break out and have a multiple sack game soon, as often as he is beating the offensive tackles. With 0:24 left in the second quarter, he blew past the tackle immediately and had to be held in order to avoid a sack. That killed the drive that Tulsa was on at the time. Later in the game, he and Antwuan Jackson combined to force an intentional grounding with 10:14 left in the third, which I have always believed should count as a sack. Every other defensive lineman is a non-factor in pass-rushing scenarios. This is why it is nice to see some blitzes installed, as they genuinely improved their pressure when they ran blitzes, as expected. The defensive-ends are pretty good with defending the run though, with Jean-Baptiste and Tuimoloau having success defending against the run. Tuimoloau had a great play in the second quarter, where with 12:50 left to go, he shed his block and stuffed the run immediately when it looked like there was going to be a hole.

Tyleik Williams was unbelievable and was maybe the best player on the field. He looked dominant at times. He is excellent against the run and the pass, standing up to double teams and getting pressure in passing situations. His sack late in the game sealed the victory against Tulsa. With 11:59 in the third, he and Jerron Cage (who also played pretty well) got a tackle for loss with both of them making their way into the backfield. With 2:36 left in the third, he again got into the backfield, forcing a throw away from the quarterback after getting immediate pressure. I am not saying he needs to start yet, but he needs to be in the regular rotation getting at least this much playing time every week. He deserves the time with the way he has played the last couple of weeks.

As for the linebackers, they looked surprisingly playable. Everyone knew their gaps for the most part, and everyone knew their zones for the most part. I talked about this earlier, but linebackers are now mostly rotated based on the situation, rather than being rotated in arbitrarily. The guy who improved the most week-to-week was Tommy Eichenberg, who went from a -7 last week to a -1.5 this week. That is still not what I would call “good,” yet he did show improvement. In zone, he got back to his spot and recognized routes developing in his area, even having a nice pass breakup on the first play of the second quarter. Against the run, he filled gaps correctly for most of the game, though during one of the double A gap blitzes with 8:39 left in the second quarter, he followed Simon into his A gap instead of attacking the adjacent A gap, exposing a huge hole. He still played far from perfect, but he did enough good things where it seems plausible that he had improved from the debacle that was last week. He looked much more comfortable and had a much better idea of where to go in the passing game. His improvement was one of the more encouraging things that happened in the game.

Steele Chambers was the highest graded player I had, but most of this was hustle plays and good coverage in the dime package, which is really a great package for his skillset right now. I said it last week and I will say it this week: Steele Chambers has really impressed me with his motor combined with his athleticism. This guy makes plays, and he does so because he genuinely never takes a play off, and he always commits to his decisions on the field. It is so refreshing to see. I hope to see him out there more in the next few games, because he is the one other linebacker on the team besides Cody Simon that I think has real potential to be a playmaker in the mold of a Darron Lee. He has that type of athletic ability and motor. 

Gaoteote, despite having only a couple of plays last week and this week, looked better as well. While I have seen some calling for Gaoteote to play more over Eichenberg, I don’t know if I totally agree. Besides the fact that they are playing two different spots in the defense right now, I think Gaoteote was, per play, the worst player I charted last week. This week, he is one of the best due to his one TFL in the red zone. I can’t say whether or not he should be ahead of Eichenberg right now, but what I can say is that both of them looked like totally different players this week, which is incredibly encouraging.

In other encouraging news, Bryson Shaw played significantly better as well. While I don’t know if he is the long term solution at the deep safety position, I thought he played fine against Tulsa. He didn’t stand out at all, which, for a free safety, is probably the nicest thing I can say. Jordan Fuller’s whole career was magnificent, and he never stood out. That’s part of what made him great. Shaw still has plenty of limitations, but I thought he looked improved from last week. Seeing the players that struggled most last week start to improve is a strong indicator that the coaching has improved by shifting responsibilities.

In slightly less fun news, the DBs did not play as well as they did against Oregon. Burke still played great, but Brown struggled a bit in zone. In man he looked totally comfortable, and even in Cover-3 he usually looked comfortable. There were times in the game though where he was the deep man, and even though the receiver was 25 yards downfield, Brown was still way behind him, and couldn’t catch up when the pass was thrown by the quarterback. With 0:44 left in the second quarter, Brown gives up a 31 yard reception because he is playing a deep third in Cover-3, but goes way too deep. I think even if you are playing the deep zone, you have to come up and defend some passes, even with the clock as low as it was. Having said that, I think he played well enough in 2019 and well enough in the Oregon game where he has earned his status as a starter.

Even Cameron Martinez had an up and down game, though the positives outweigh the negatives. He made the pick six at the end of the game, he had excellent pass coverage to prevent a first down with 7:17 left in the second quarter, and he had great coverage in zone at the end of the first half when he tipped the interception to Ronnie Hickman. He looks comfortable for the most part in coverage. He did get beat a couple of times, particularly with the first pick route to score Tulsa’s first touchdown. He was also kind of involved when Lathan Ransom’s man scored on the pick route to score Tulsa’s second touchdown. Either way, it was good to see him make some big plays for a defense who desperately needed them.

So how did the coaches do? Pretty well honestly. The players who struggled last week (Eichenberg, Shaw, 1-tech) all played way better, and it wasn’t just because it was Tulsa. They all looked much more competent in terms of knowing where they needed to be, which was a clear reflection on the coaching staff’s efforts. Serious work was put in to making sure the players knew their assignments.

On top of ensuring that the players knew the basics, the coaches also installed new wrinkles to the defense, which already makes this the most complex defense the Buckeyes have run since 2018. The fact that they did this, and the players played more fundamentally, is a good sign that they made up a huge amount of ground this week.

The only real negative I had is a lack of in-game adjustments on the touchdowns from the Trips Bunch set that Tulsa was running. On the first touchdown, Tulsa exploited Ohio State’s man coverage by running a pick route from the Trips bunch set, which prevented Martinez from following his man for a touchdown. Once is fine. It happens. At this point though, an adjustment has to be made.

Then the very next series, Tulsa is once again in the red zone and lines up in a Trips Bunch formation. Ohio State looks like they are sufficiently backed off of the receivers. The ball is snapped, and the play is run, and it is another pick play. Before the snap though, Ohio State calls a time out, probably to address the pick play. The color analyst even mentioned, “There is no way that Tulsa will run that same play now that Ohio State has seen it. Look for something else.” As soon as the timeout was over, Tulsa lined up in the same set, Ohio State lined up in man-to-man again, and Tulsa ran another pick route for a touchdown. Getting beat on that once stings, but it’s understandable. Getting beat on that twice is maddening. What were you doing after the last time they ran that play? That was the only moment of the game where I was disappointed with the coaching, and the only time that I felt my optimism curbed a bit.

The reason that I was so optimistic was because I though Ohio State played fairly decently on defense. Before the final garbage time drive, Tulsa’s quarterback was averaging about 7.4 YPA, which was…the exact same number CJ Stroud was averaging. They also had similar efficiency metrics, including each having a 1:1 touchdown to interception ratio. If CJ Stroud was throwing against the Buckeyes on Saturday, but instead of averaging 8 YPC, they were averaging 2.1 YPA, would you still think of that offense as dominating the defense? I certainly wouldn’t. I don’t think many fans would say that Stroud was overly dominant. And this offers a little perspective, I think. The Ohio State defense played, below average, which, in my opinion, is a huge step forward from last week. The new coaching changes have me feeling cautiously optimistic, which is not something I would have thought I was going to say seven days ago.