Michigan State is well-built to upset Ohio State

Ohio State is in a must-win position against Michigan State as the two teams are essentially playing a Big Ten East championship game.

And in a lot of ways, despite the huge Vegas line favoring the Buckeyes, the Spartans are well-built to get the upset.

So how good are the Spartans? The rank just ahead of Ohio State in the College Football Playoff rankings, but still significantly behind in the S&P+. From Bill’s post on the Big Ten race:

MSU is good, but did need some turnovers luck to get by Michigan and Penn State. Based on national averages for fumble recovery rates and the ratio of interceptions to passes broken up, the Spartans’ expected turnover margin in those two huge wins was plus-zero. In real life, it was plus-seven. The average turnover is worth about five points’ worth of field position lost (by the offense) and gained (by the defense), so this discrepancy was worth around 35 points to the Spartans, who won these games by a combined seven.

So turnovers have played a significant role in the Spartans wins — but that shouldn’t be too comforting for Ohio State fans given their specific strengths and weaknesses.

Vegas favors Ohio State by a 15.5 point margin, which seems absurd on the face of it. But the raw statistics really love Ohio State because the Buckeyes’ best is the best in the country. The problem is that their floor, which we (hopefully) saw last Saturday in Kinnick Stadium, is just top-30, not top-4.

A few other notes:

• Ohio State’s average turnover margin per game has been steadily declining. Ohio State ranks 61st at just +.11/game on the year after ranking 3rd at +1.15/game last year.
• Kickoffs could be an issue, again. Michigan State ranks fifth in the country in kickoff return success rate
• Sometimes the more turnover, the better the end result. Michigan State had an insanely low percentage of returning production (116th and 107th), but appears to be better off for it.

When Ohio State has the ball

Before getting in to the specific matchup, it’s worth noting that Ohio State’s offense was not the problem in the loss to Iowa, but there were certainly some important issues that Iowa’s top-20 defense revealed: responding to defenses that take away intermediate throws, force Barrett to keep in the zone-read run game, and what happens with penalties or after offensive line injuries.

Michigan State’s defense is much-improved from last year, when they ranked 41st in the defensive S&P+ — they’re up to 9th now, and particularly excel in stopping the run. Where Iowa was ranked poorly in run defense heading in to last week’s game, and stronger against the pass, the Spartans’ strengths are flipped, ranking 6th in rushing S&P+, but 38th in passing S&P+. Like Iowa, they prevent big plays (7th in overall IsoPPP+) but they’re much worse at finishing drives, allowing an average of 4.22 points per opponent scoring opportunity (51st).

Interestingly, while they are one of the best defenses in the country at preventing runs of 5+ yards (6th, or 31.1% of runs), their run defense isn’t built so much on making plays in the backfield, ranking just 43rd in stuff rate. In fact, the defense’s success as a whole isn’t built on making havoc plays, ranking 49th in overall havoc and 71st in defensive line havoc.

So that gives us two big questions:

1. Can Ohio State run on Michigan State? This is the best run defense the Buckeyes will have seen this season. They held Saquon Barkley to 14 carries for 63 yards last week. And if the Spartans win the battle up front, how will the offense respond if forced to rely on the pass? Meyer and Wilson also mentioned ways to increase the running back’s success rate on read-based runs. That suggests we might see more dart runs or RPO variations, as well as potentially some non-read running back runs as well.
2. Can Ohio State exploit the Spartans’ relative weakness against the pass? At 38th in passing S&P+, the Spartans are still a good pass defense, but they allow a relatively high percentage of successful shorter completions. They allowed 368 yards on 50 attempts (7.4 yards per attempt) to Northwestern, then 401 yards (8.4 yards per attempt) last week to Trace McSorley. Opposing offenses have keyed on the Spartans’ pass defense, throwing on 49.9% of standard downs — which is the 5th-most in the country. All week we’ve talked about J.K. Dobbins getting the ball more — and Urban mentioned that it would be a point of emphasis this week after practice. But we should be prepared for a pass-heavy attack nevertheless, because that’s what is likely to be effective against the Michigan State defense.

When Michigan State has the ball

The Spartans don’t have the most elite offense the Buckeyes’ have seen this year, but they do have one well-positioned to take advantage of Ohio State’s weaknesses — similar to Iowa. The similarities are clear: the Spartans offense ranks 101st in rushing S&P+, but third in the country in passing S&P+ behind Brian Lewerke.

Lewerke’s stats don’t jump off the page immediately: 2,207 passing yards, 6.5 yards per attempt, five interceptions. But his passing success rate is 44.9%, which is 26th in the country, and they’ve faced a number of solid defenses. In the past two games against Northwestern and Penn State, Lewerke has thrown for 845 yards and over 7 yards per attempt. He’s had over 53 attempts three times this year.

Part of the reason for going so pass-heavy has to do with the poor run game. Not only are they 101st in rushing S&P+ overall, but they allow run stuffs on over a quarter of runs, ranking fourth-to-last in the country in stuff rate. Ohio State’s defense ranks 7th in the country in forcing run stuff. So you’d expect the Spartans offense to be completely one dimensional — but that’s also what should have happened last week, as well. But Ohio State’s defense, and linebackers specifically, failed to fill gaps behind the defensive line’s slants, and seemed unsure who had pursuit and coverage responsibilities. Michigan State undoubtedly seeks to take advantage of Ohio State’s defense in a similar way. If the defense allows the Spartans to have a steady run game in addition to a likely-efficient passing day, then it’s likely to be another long afternoon for Greg Schiano’s defense.

Given how much the numbers favor Ohio State’s run defense and don’t favor its pass defense, any schematic or execution issues for the run defense will likely be devastating.

Spartans tight end Matt Sokol is the fourth-most targeted player in the passing game, but he still only has 16 catches and 166 total receiving yards this year, while LJ Scott has 15 catches for 91 yards. Michigan State will likely try to use motions and shifts to create advantageous matchups with tight ends and running backs on Ohio State’s linebackers, as the linebacker corps hasn’t shown much improvement in pass coverage responsibilities since getting exposed against Oklahoma.

Assuming that Ohio State’s run defense corrects the most glaring issues from a week ago and realizes their statistical advantage in run defense, then even a steadily efficient passing game should be slowed in the red zone, where the Spartans average just 3.94 points per scoring opportunity, which is 105th in the country. Field goals should be the name of the game. Expect to see a fair number of third-down conversions, as they rank 3rd in passing downs S&P+ despite being fifth-worst overall in passing downs line yards per carry (which means that nearly all of that late-down success is due Lewerke and not the running backs).

Takeaways

1. The Spartans defense excels against the run, ranking 6th in rushing S&P+. They don’t make a ton of tackles for loss, but they also limit big gains.
2. Their relative defensive weakness is against the pass, ranking 38th, and 48th in passing success rate. The Spartans defense is very similar to Ohio State’s in that it intends to stop the run first, can allow efficient passers, but still rarely allows big pass plays.
3. Brian Lewerke is up there with Baker Mayfield and Trace McSorley in terms of passing threats. But while Lewerke may not be as dynamic as Mayfield or as deep ball-ready as McSorley, he has shown the ability to be efficient, potentially like Richard Lagow and Nate Stanley were. Lewerke leads the third-best passing offense overall and on passing downs.
4. Ohio State should have a huge advantage stopping the Spartans ground game (5th vs. 101st in rushing S&P+), but they should have been able to stop Iowa too, who has similar problems as the Spartans (127th in stuff rate). If this advantage isn’t realized, then the Buckeyes have big problems.
5. There are a number of less-quantifiable factors that could affect the game. Mark Dantonio has taken statistically inferior teams toe-to-tow with Meyer’s, forcing games decided by three points or less in four of the last six years. We’re also not sure where Ohio State’s mindset is now that their playoff hopes are almost assuredly gone.

Picks

• S&P+: Ohio State 33, Michigan State 19. 78.2% win probability
• F/+: Ohio State by 7.9. 67.6% win probability
• Adj. S&P+: Ohio State by 30.3 (!). 96% win probability
• My pick: Ohio State 31, Michigan State 28