Stats/takes from UM, OSU, MSU, and PSU's Week Two games

I took a deep dive into the box scores from Pro Football Focus and Game on Paper for the Michigan, OSU, MSU, and PSU Week 2 games. (Last week, I covered every Big Ten team, and I’m going to try to do that as frequently as possible, but this week Life made that an impossibility.) I tried to generate a couple blurbs based on those box scores from the statistics that I found compelling from each game. It could be compelling for the game itself, or maybe for what it portends for the team or player that season. It also could be a season-long statistic that I find compelling. Really, there’s no formula. This is just a very paper-based, statistical approach to evaluating teams. And, also, sometimes I watch a little bit of film if I don’t believe the statistics and want to know what the heck is happening (see: Drew Allar).

Note: Games are listed in order of how much I think a Michigan fan will care about the (a) teams playing, and (b) outcome of the game.

MICHIGAN defeats UNLV (35-7)

It was boring, but not THAT boring. Game on Paper has this nifty stat called “Game Excitement Index.” The GEI score is determined based on the fluctuations in win probability throughout a game. The higher the score, the more exciting the game. A thriller might score somewhere around five or six points. A blowout will be less than two, and often less than one. The GEI in the Michigan-ECU game last week was 1.43 (which seems kind of high since the win probability never dipped below 95%), but against UNLV, the GEI was just 0.17 (the win probability never dipped below 97.5%). That said, there were two games on last year’s schedule that were somehow even more boring than Michigan’s systematic demolition of UNLV: against Hawaii and UConn last year, the GEI was (hold your butts) 0.03 and 0.05 respectively.

Sit. Stay. Good boy. What Michigan’s defense did to UNLV’s “go-go” offense was sadistically punitive. The Aggies offense averaged -0.32 EPA/play and just 4.0 yards per play. Through the first three quarters, they only gained 20% of the available yards. Their offensive line averaged a measly 0.8 OL yards/carry, and Michigan’s defense created 10 “havoc” plays (5 sacks and 5 TFLs). Having three monsters in the middle (Kris Jenkins, Mason Graham, and Kenneth Grant) makes it immensely difficult for any team to (a) run the ball, and (b) feel comfortable in a collapsing pocket.

Somehow, it’s going to get even better. Through the first two games, three of the five starting defensive backs have barely seen the field: Makari Paige (10 snaps), Will Johnson (8 snaps), and Rod Moore (zero snaps). While Quinten Johnson, Kechaun Harris, and Keon Sabb have done a serviceable job against overmatched opponents (the last of which looks to be a promising multi-year starter), there’s no doubt that Paige, W. Johnson, and Moore will be an upgrade. Q. Johnson, Harris, and Sabb have grades of 59.9, 57.0, and 68.3 respectively on PFF. Throughout the entire season last year, Paige, W. Johnson, and Moore received grades of 79.1, 88.2, and 83.8. If anything happens to get past the front seven, it’ll be very nice to have them clean it up.

I am not throwing cold water. J.J. McCarthy’s ascendance to bona fide Heisman candidate has been the most exciting thing about the Michigan football season so far. Watching him create extra time for himself, manipulate coverages with his eyes, and deliver dimes downfield has been delightful. If you’re not watching him do these things live, though, you might think that he hasn’t improved all that much. Why? Through three non-con games last year (so, really, two because he barely played in the opener), McCarthy’s stats looked like this: 30/34 (88.2%), 477 yards (14.0 Y/A), 3 TDs, 0 INTs, 8.3 aDOT. Through the two non-con games so far this year, his stats look like this: 48/55 (87.3%), 558 yards (10.1 Y/A), 5 TDs, 0 INTs, 8.5 aDOT. There’s really not much variance in excellence there. Again, I’m not throwing cold water on this. Great stats against cupcakes are often puntable, and it’s much more important to watch the games and see how those great stats were achieved. It’s just something interesting.

PFF’s blocking grades may not be wholly worthless. There’s something not right with the run game. Last year’s slashing and mauling has given way to this year’s clenching and squirming. Why is that? The OL line yards per carry in last year’s non-con games were 1.89, 1.76, and 1.23, and this year they’ve been 1.71 and 1.75, so there isn’t much change there. Last year’s highlight yards per rush through the non-con (3.23, 7.26, and 3.11) also don’t see much change (5.50 and 4.15). They are being stopped for two yards or fewer more frequently (28%, 23%, and 44% vs. 58% and 44%), but that’s more of a “what” than a “why” stat. Some of it may be ECU and UNLV flinging guys into the line of scrimmage, and some of it may be stubborn play-calling. But it’s also been mentioned a handful of times on the blog that the receivers were much better at blocking last year than this year. PFF’s numbers very much support this claim. Last year, Ronnie Bell received a 67.6 RBLK grade on 391 snaps, Cornelius Johnson received a 53.6 RBLK grade in 309 snaps, Andrel Anthony received a 68.9 RBLK grade in 200 snaps, and Roman Wilson received a 52.4 RBLK grade in 98 snaps. Obviously, Bell and Anthony are gone, and Johnson and Wilson’s numbers are actually down to 41.3 and 44.0 respectively this year. Those are easily the two lowest RBLK grades amongst the WRs and TEs. If Johnson and Wilson can do better at getting a hat on their guys, the run game will improve.

OHIO STATE defeats YOUNGSTOWN STATE (35-7)

There was a battle this off season? At QB? I suppose I can understand why Ryan Day considered Devin Brown as a viable option to start at quarterback: he uses his legs, which brings an element to the offense that Kyle McCord doesn’t. However, I wouldn’t exactly call Brown a gamebreaker: he kept three reads and scrambled once for a total of 16 yards on the ground. He also fumbled once (but recovered the ball). Through the air, though, McCord was statistically superior: he went 14/20 (70%) for 258 yards (12.9 Y/A) with 3 TDs, whereas Brown went 7/13 (54%) for 101 yards (7.8 Y/A) and zero TDs. The eye test backs up that McCord simply slings it better.

Revisiting my new favorite stat for critiquing quarterbacks. But are McCord’s yards really because McCord is a marksman, or can we attribute them to something else? In order to attempt to statistically answer this, let’s look at something I’m going to call “YacPy,” which is calculated as follows: (total yards after the catch) / (total passing yards). It’s not 100% accurate to say that the closer this number is to one, the more work the WRs are doing, and the closer this number is to zero, the more work the QBs are doing, but I think that’s very roughly true of the metric. Since I don’t have nearly enough time to go through every pass thrown in that game and determine how many of the yards after the catch were generated on McCord throws and how many yards after the catch were generated on Brown throws, the only YacPy number I can offer today is the team YacPy: 0.53. (I will add, though, that Marvin Harrison Jr. had 47 YAC on his first TD catch, and Emeka Egbuka had 29 YAC on his TD catch; both of those passes were from McCord.) Based on the limited YacPy numbers I’ve calculated this season, 0.53 is pretty high. (If we remember from last week, Noah Kim’s YacPy was 0.54, and J.J. McCarthy’s was 0.29. This week, Michigan’s YacPy went up to 0.39, and Noah Kim’s went down to 0.26, albeit against Richmond.) All this is to say that McCord’s statistical success could be more a product of busted coverages or excellent route running/post-catch evasiveness, rather than genuine marksmanship.

Why do it in three when you have four? This year against substandard competition (Indiana’s defense is 66th in F+, and Youngstown State is an okay FCS team), the Buckeyes’ offense has totally stalled on third downs. Against the Hoosiers, they converted 2/12 opportunities, and against the Penguins that was 5/12. Combined, that’s 7/24, which is just over 29% (they were 16/24 against Arkansas State and Toledo a season ago.) Last year, a third down conversion rate of 29% would’ve ranked 124th in the FBS, somewhere in between Miami (OH) and FIU. The Buckeyes have managed to convert 4/6 fourth downs this year, bringing their total conversion rate up to 46%, which is one percentage point higher than their third down conversion rate last year. OSU still has Notre Dame (9th in DF+), Penn State (4th), Wisconsin (13th), Minnesota (8th), and Michigan (3rd) on their schedule. This could bode poorly.

Little Flags Everywhere. After watching the game and checking the box scores, this was some unholy combination of undisciplined football and a ref show. Ohio State was penalized 6 times for 65 yards, and Youngstown State was penalized 9 times for 76 yards. Last season, 117 FBS teams were penalized fewer than 65 yards per game, and no team was penalized 76 yards or more per game. Some of these penalties were wholly deserved (OSU LT Josh Simmons took off a guy’s helmet with his hands while trying to block him) and some were almost certainly not (just a few plays after the aforementioned, YSU got hit with a very questionable unnecessary roughness which saved OSU’s drive late in the second quarter).

If he can do that, what can’t our guy do? The YSU player who stood out most to me was RB Tyshon King. The Northern Michigan transfer(!) carried the ball 12 times for 66 yards. He also had a 26 yard rush taken off the board because of a light hold by a receiver––without that hold, that rush would’ve gone for at least 10 yards, too. To be honest, King reminded me a lot of Blake Corum: a little bit on the smaller and thicker side, but laterally quick, with a lot of patience and vision. It beats the hell out of me why YSU didn’t feed this guy more, and instead chose to hand it off to Dra Rushton 9 times (he gained just 22 yards). Frankly, the fact that the Buckeyes allowed an FCS back (nay, a back who transferred to the FCS from a D-II program!) to average over 5 yards per carry should be extremely concerning to OSU fans.

Pathetic and classless. Many will not have watched this game to the end, but the final snap from scrimmage came with 25 seconds left on the game clock. Ohio State had the ball and was up 35-7. You take a knee, right? WRONG! You hand the ball off to Miyan Williams because it looks like you’re running down the clock anyway, but there’s a small chance he breaks it for a touchdown and your score looks a little better. Jerks.

MICHIGAN STATE defeats RICHMOND (45-14)

The worst gets better in the worst way possible. Even in the wake of the 2021 season when saying, “Mel Tucker is a good coach,” was a semi-viable statement, his 10-year, $95 million contract extension seemed like a little (way) too much. When the Spartans fell off a cliff nearly immediately after the extension by going 5-7 in 2022, I very much looked forward to the next 10 years of distressed Spartan football, as well as the culminating day down the road when I could say, “Mel Tucker was paid X million dollars per win at MSU,” and smirk about it. I suppose today is that day. I’m not exactly certain how much money MSU paid Tucker since he signed the extension, and at this point it's unclear how much he'll be paid when the dust settles, but if we assume this was his contract for the past 22ish months, he's "earned" roughly $17 million. Since then, Tucker has won 9 games, meaning each MSU win under that contract was worth roughly $1.9 million. And that’s with the benefit of starting this season against CMU and Richmond! What's more, if Craig Ross' buddy is to be believed, MSU could end up paying him upwards of $20 million more, meaning each win was worth $4+ million! I couldn’t even begin to imagine the self-loathing, disreputable bittersweetness I might be feeling if I were a diehard Spartan. What a horrible schmuck of a human being. “Who’s got it worse than the Spartans?” “Nobody!”

Let’s keep counting. One of my closest friends sent me an image posted by MSU Football Twitter that pictured QB Noah Kim and RB Nathan Carter. On the image, it stated that those two were leading the Big Ten in passing yards and rushing yards respectively at the moment. His caption: “STOP THE COUNT!” (Yes, he's a Spartan. Sigh.) The truthiness of these facts is a truthiness I can’t deny: Kim has 571 passing yards, which is 13 more than J.J. McCarthy, and Nathan Carter has 224 rushing yards, which is 8 more than Maryland’s Roman Hemby. Let’s rain on this parade a little bit: McCarthy’s completion percentage is 21.8% better than Kim’s; his passer rating is 9 points higher; and his PFF passing grade is 9.5 points higher. McCarthy also does not have a turnover-worthy play (Kim has two), and has thrown for 5 more first downs. Meanwhile, Hemby’s Y/C is 0.9 higher than Carter’s, he averages 1.71 more yards after contact per carry, he’s broken 4 more tackles, has 34 more receiving yards, and has an elusiveness score on PFF that’s 200% better (149.6 to 49.9). So good job, MSU: you lead the Big Ten in the most rudimentary ways of evaluating a quarterback and a running back. Keep it up, I guess.

Another stat an MSU guy might lead in. There’s no reason to expect that Charles Brantley will lose his grip on a starting cornerback job this year, and there’s also very little reason to expect that he won’t miss many more tackles throughout the course of the year. He’s already missed 4 tackles against limp competition (2 in each game), and missed more than 20% of his tackles in both 2021 and 2022. It’s very unlikely he leads the Big Ten in missed tackles, simply because of the nature of the cornerback position, but it’s not unreasonable to expect him to lead cornerbacks in missed tackles. He finished second in that measurement last season, and last season’s leader (Nebraska’s Quinton Newsome) is off to a good start: no missed tackles through two games.

Slow starters. In both games this season, the Spartans have stumbled out of the gate: against CMU, their first drive ended on downs and they punted on their second drive; against Richmond, both of their first two drives went three-and-out and they punted. They received a gift when the Richmond returner muffed their second punt, and the MSU offense had a short field with which to work (they converted for a touchdown).

Something funny. MSU received a delay of game penalty on the PAT attempt after their first TD. Lol.

PENN STATE defeats DELAWARE (63-7)

Varsity vs. junior varsity. This was a comprehensive shellacking by nearly every metric. Penn State had 541 yards to Delaware’s 140. Nearly half (66) of the Blue Hens’ yards came on a long touchdown run when backup LB Tyler Elsdon busted (he chose to run into the back of two of his defensive linemen rather than fill a gap made between the center and the guard). Penn State had 34 first downs to Delaware’s five. Penn State had the ball for 42 minutes and 22 seconds. Penn State ran 91 plays, and Delaware ran only 41. On those plays, Penn State’s success rate was 58% and Delaware’s was 24%. There’s putting a cupcake on your schedule to tune up, and then there’s putting a mini cupcake on your schedule just to step on.

The part where I remain skeptical of Drew Allar. On first blush, Allar’s statline against the Blue Hens looks almost McCarthy-esque: 22/26 (85%) for 204 yards (7.8 Y/A) with 1 TD and no picks. But here are the numbers that don’t show up in the unsophisticated narrative presented by most media outlets: 3.8 aDOT and 0.59 team YacPy (backup Beau Pribula threw 5 passes and completed 3 of them for 22 yards). Those numbers are alarming, especially when you consider that Delaware is ranked 156th on Sagarin (an SP+ like ranking system that also includes FCS teams). The vast majority of Allar’s throws were checkdowns, screens, or designed leak routes to TE Tyler Warren. (In my mind, a leak route is when a TE starts to block, but then leaks off the line and catches a very short pass, so more or less a dump off.) Is this Franklin keeping things simple against a way overmatched opponent, or is this Franklin revealing that he genuinely lacks faith in his gunslinger? You know where I stand.

Spot of real concern. Other than Elsdon’s boneheaded mistake on the long touchdown run, this was a pretty flawless performance for the Nittany Lions––except for RT Caedan Wallace. In the Enemy: Ranked preview, Alex Drain referred to him as “one notch above turnstile.” While PFF didn’t ding him for any pressures against West Virginia, he was the only offensive linemen to seemingly get beat against Delaware, allowing three pressures in the process. We’ll have to see this weekend how he fares against stiffer competition when PSU heads to Illinois.

It’s clear why they scheduled Delaware. Their jerseys make them look like a shitty, off-brand pinny of a team that whooped up on them last year. (Michigan.)