Before the season started, Ryan Day had clearly outlined his expectations for the Ohio State defense with one short sentence:

“We expect a top-ten defense.”

When asked about Day’s quote, Jim Knowles upped the ante:

“I wish he would have said top-five.”

Halfway through the 2022 college football season Knowles’ expectations seem to have been fulfilled; in the SP+ rankings, which combine success rates, drive production, and opponent adjustments have Ohio State ranked as the 5th best defense in all of college football.

This Michigan State game was a perfect showcase for the defense, as the starters exited only allowing 100 total yards, including negative rushing yardage.

Much like the offense, it is sometimes hard to take away much from a game against an opponent like that. In this one, there were a couple of notable moments. For one, I think most people are concerned about the corner play, but outside of a couple of moments from Cameron Brown, I think it was fine in this one. He has been mostly solid this year, so I am not too concerned with one game, especially defending Jayden Reed all game, who will be the best receiver on Ohio State’s schedule until Michigan, maybe until the playoffs.

Also, their adjustments were outstanding. While it wasn’t completely obvious until rewatching, after Reed scored a touchdown with 11:07 remaining in the second quarter, the starters held Michigan State to -10 yards until the end of the third quarter. That is a third of a football game without allowing positive yardage. I am not going into every game expecting that dominance, but sheesh.

That is also why, if you were expecting negative grades, you are going to be disappointed. Here is the defensive chart:

The defensive line had about their usual day, but that has been outstanding this year. Just a week after I had said that Mike Hall had cooled off after his huge first two games, he goes out and has a 2.5 sack day in only 7 snaps. It wasn’t quite as dominant as the stat sheet showed. In a couple of those sacks, the quarterback was first pressured by someone without an official sack this year, first Zach Harrison, then J.T. Tuimoloau. Harrison didn’t play as well as he usually does, but the fact that he hasn’t been credited with a sack this year is bizarre. I have charted a couple of plays where he has tackled the quarterback in the backfield during a pass. Yet neither were even counted as TFLs. Maybe the QB barely got back to the line? I have no idea. He has been so snake bit the past couple of years with good pressure but no stats to show for it. J.T. is starting to be in the same position as well. Anyone who watched the game on Saturday could tell you that he was abusing the Michigan State tackle, and yet has not recorded a sack on the year. Last year, I thought he ran into a couple of sacks by sheer luck or after someone like Harrison had forced the quarterback to take off. This year, he is the one getting the pressure only to push the quarterback into the waiting arms of a teammate.

Of course, it is worth noting that of all Ohio State’s 14 sacks this year – and I will need to double check this before next week’s write-up to let you know – were completed faster than the NFL median sack time of 4.17 s. This means that opponents are game planning to an extreme degree around not holding the ball at all against Ohio State and trying to make a living throwing quick passes. Michigan State, for example, ran only 1 play-action the entire day. Toledo had 2 play-actions. Rutgers had 3 play-actions, and two of them were screens. The third was a TE flat. So far, it hasn’t worked. Oklahoma State had a similar problem at times last year until they played a team that had longer developing plays in Oklahoma, where Caleb Williams was promptly sacked six times. I don’t know if Ohio State will play a team that would be equivalent to that, but Penn State and Maryland are the two most likely to attempt to do this.

The linebackers played their usual too, though it was far more lopsided than usual. Even though he only finished with 8 tackles, he graded out higher than he has at any point this season. He was everywhere in this game and is so disruptive against the run. I think PFF had him as the best defender in the country against the run, and that feels entirely accurate. Still, for his play this year, he is vastly underrated around the country, though I certainly don’t need to tell you that.

The safeties are still playing better than any group I can remember except for the 2016 group. McCallister caught on pretty quickly to the RPO Slot Screen, and started to bait it as the game went along. The first time MSU ran it, they got seven. Then they got four. Then they got three. He continued to make adjustments to that play-call as the game went on. Ransom had another outstanding day. While the interception was an outstanding play, I think I was more impressed with the pass breakup along the sideline. While it was hard to see on first viewing, he and the rest of the Ohio State defense were not ready when this ball was snapped. For him to recognize his coverage responsibility, get to the sideline and break up a good pass was just exceptional. There is a real case to be made that he has been the best safety on the team this year, which, based on this play last year, I would have never expected.

Speaking of the defense’s performance for the year, here are their aggregate scores so far in 2022:

While I think it is somewhat intuitive how this group has performed thus far, it is probably easier when given a comparison. So, I also have last year’s midseason defensive chart here as well:

So, at least by comparison, the defensive line and secondary are playing about as well as they did last year, with the linebackers playing much better, though I kind of disagree with my own rankings from last year. First, while it seemed like the defensive linemen were winning their one-on-ones, they were likely making some mistakes that I was instead putting on the linebackers, so while their score seems pretty good, some of the negatives that the linebackers took should have absolutely been re-routed to the defensive line in hindsight. Furthermore, I was not as methodical in measuring pressure last season, so some of their pressure grades may be inflated. Given these considerations, I would take the play of the 2022 defensive linemen and linebackers over their 2021 counterparts.

The secondary is interesting, as though the safeties have drastically improved (look at Lathan Ransom), the corners have seen their production dip, resulting in the exact same score as last year. Honestly, not all of this was the fault of the corners. Last year, both Brown and Burke were usually just relied on to not get beat deep. They were willing to give up underneath routes, especially when they were in Cover 3. When that would happen, I would ding the play-caller, though I wasn’t as strict for this early on in the year. Still, one of the reasons why last year’s defense had a much worse play-calling score was the constant giving up of short-yardage passes. This year, those passes are nowhere to be found. A shift in defensive philosophy has put the corners on islands, and so far, they haven’t been up to the task. I am relatively bullish in Burke and Brown down the stretch; it is relatively easy to teach a corner who is already in consistently good position to be a little better with their ball skills. It is much harder to teach a guy who has the inverse problem and can’t stay within seven yards of a receiver. If they can just get slightly better, this score will increase quite a bit. Also, I used to chart the backups as well, so they would often pad the stats in garbage time.

Before finishing up, I thought it might be fun to see how this defense has been stacking up against the previous defenses since Urban Meyer took over in 2012. In particular, I wanted to see how they looked using the percentage of opponent averages allowed stat that the College Football Nerds YouTube channel have popularized. I like the stat because it controls for the level of competition a team plays. The general idea is that if you allow 80% of your opponent averages for YPC, then you will likely give up 4 YPC against a team that normally runs for 5 YPC. So, to clear it up, lower is better. I did remove the garbage time data for Wisconsin as it is was an outlier of the data. Considering how beat up Ohio State was, their second/third team against Wisconsin’s starters is not something I would expect to play much of a role in predicting future performance. I also through in Michigan’s defense this year, of which I have been tracking.

Here is how this defense compares to the rest:

Now it is early, and it is worth mentioning that last year’s defense was looking pretty good by midseason as well, but this defense is looking like one of Ohio State’s best of the decade. Are they the best? I wouldn’t say so yet. First, they are kind of in a three-way tie with 2017 and 2019 for being the best defense of the past 11 years. 2019 wasn’t quite as good against the run, but it was a good bit better against the pass.

The 2017 unit is almost the exact same defense as 2022’s version according to this metric and is a fascinating defense on its own. In fact, halfway through its own season, that defense had allowed 56.1% of opponents’ YPC per game and 70.8% of opponents’ YPA per game. For reference, that was better than 2021 Georgia. Then, they went to Iowa City. I don’t remember anything after that.

What should we worry about for the rest of the season?

Unknowns about this defense’s ability to scale. Last year, after Matt Barnes took over, the defense was essentially just a patchwork collection of schemes and fundamentals that were installed in a matter of weeks. Due to this, they were actually pretty capable against the middle part of their schedule against some offenses that really weren’t that bad. Michigan State, Purdue, Penn State, and Maryland all had decent offenses last year, and they couldn’t do much against Ohio State. However, as soon as Ohio State had to play top ten offenses like Utah and Michigan, they couldn’t hold up. They were like the defensive equivalent of the Ed Warriner 2016 offense, when they would put up 60 like nothing against some teams and then get score one offensive touchdown against Michigan in regulation before getting shut out by Clemson. They were in some ways a poor man’s version of Don Brown’s defenses at Michigan, when they would look like the best defense in the country before running into Ryan Day’s offenses.

Some defenses scale well, and others do not. If this defense does scale, then we probably have nothing to worry about. Until they play a top ten offense though, they won’t really know. And that won’t be until the Saturday after Thanksgiving.

What should we not worry about for the rest of the season?

The second half of the 2021 Michigan game. Jim Knowles came into this year as a great mid-game adjuster. While it might be a little too early to draw many conclusions, this trend seems to be continuing. Notre Dame was shut out in the second half. Toledo, Ark State, and Rutgers all scored less in the second half than the first. Even Michigan State and Wisconsin were held to less in the second half while the starters were in. And while it wouldn’t look like much of a trend without that caveat, I do think it is important to mention. The defense has gotten better throughout each game that they’ve played thus far. On top of that, the run defense has been as good as it has been in the past decade. I am not saying it can’t happen, but I will be shocked if Michigan runs like they did last year against Ohio State this year. JJ McCarthy will have to win the game for them through the air, for better or worse.

What should we get excited about for the rest of the season?

This has been a bit of a sup-plot throughout the beginning of the season, but Jim Knowles really hasn’t done a ton of different things regarding blitzing, which was his big hallmark at Oklahoma State. His biggest help has just been getting each level of the defense on the same page and running a variety of coverages while disguising looks. Furthermore, in most of these early games, he really has not been mixing it up much at all compared to what I was used to seeing last year with him. I absolutely think he has been working on different blitzes with the team, but they haven’t had to use it at all. I haven’t really seen a corner/safety blitz this year, which was relatively common for him in the past. I think this defense has another level it can get to against the premier opponents on the schedule, but I do not think they are going to get too exotic until they absolutely need it. Having that in their back pocket is going to be huge, and I think you might see it in a big spot against Michigan.

I know I have talked a lot about Michigan, but at the midpoint of this season, that game seems to be set for a collision course.