By the Numbers
There's no sugarcoating things: OSU is undoubtedly the best offense Northwestern will face this year. They're fourth overall in S&P+; in conventional stats, their 544 yards/game is second in the country, behind only Oklahoma's mind-boggling 584 yards/game, and their 6.7 yards/play is 13th in the country.
Stylistically, they're a pass heavy, high efficiency, moderate tempo team: on standard downs, their 55.9% run rate is 86th in the country, dropping to 26.8% run, 109th in the country, on passing downs. They have run more plays than all but two teams in the country this season, but they're only 42nd in Bill Connelly's adjusted pace, due to their pass-heavy play selection and fairly high time of possession. In top line S&P+ components, they're 8th in success rate but only 71st in IsoPPP.
The strength and primary weapon of the offense is the passing game. It averages 363 yards/game, behind only Mike Leach's Washington State offense, at a superb 9 yards/attempt. This efficiency, combined with very low sack and interception rates (starting QB Dwayne Haskins has thrown only 7 interceptions in 455 attempts, and they have allowed only 16 sacks on the season) has pushed the passing offense to sixth in passing S&P+. Stylistically, it leans heavily towards efficiency: fifth in marginal efficiency to only 43rd in marginal explosiveness. Targets are distributed into three main tiers: at the top are Parris Campbell and KJ Hill, with 89 and 86 targets respectively. The next tier has two (healthy, the injured Austin Mack recorded similar numbers of targets) receivers, Terry McLaurin and Johnny Dixon, with 49 and 43 targets; these two lean a bit more towards explosiveness than Campbell and Hill. The next tier, with 20-30 targets each, includes the main running backs and tight end Luke Farrell.
The run game has declined dramatically from Urban Meyer's normal standards (OSU has led the Big Ten in rushing yards per game and per carry in four of his six years in Columbus, including each of the last three), but it's still entirely respectable: 181 yards/game and 4.4 yards/carry are seventh and eighth in the Big Ten, and OSU is a reasonable 48th in rushing S&P+. The lack of explosiveness is very dramatic here, as OSU is a respectable 30th in rushing marginal efficiency but 113th in marginal explosiveness; by comparison, Northwestern's quite poor run game is 106th in marginal explosiveness. The obvious culprit here is the lack of contribution from the quarterback: while JT Barrett carried 165 times (11.8 per game) for 798 yards (4.8 per carry) and 12 TDs in 2017, Haskins has only recorded 63 carries (5.3 per game) for 127 yards (2.0 per carry) and 4 TDs. While the adaptation to playing a less dangerous runner has been successful for the offense as a whole, Meyer has struggled to figure out how to open up space for the backs in the absence of a quarterback whose feet demand attention from the defense.
To the extent they have a weakness, it's the red zone: OSU has scored a touchdown on only 58% of their red zone visits, 95th in the country. NU has had success against past Meyer teams in this area, holding the 2013 team to 3 TDs and 2 FGs on 6 visits and the 2016 team to 2 TDs and a FG on 4 visits; this team is worse in the red zone than either of those were, so hopefully the defense can at least hold OSU to field goals when they get close.
I'm going to focus on Ohio State's game against Michigan State. This is, in many ways, their worst game, so take it with a grain of salt (or better yet a whole shaker); however, I think it's a reasonable stylistic match (MSU is another 4-3 quarters based defense, and they even have eased off the aggressive corner play that long characterized their defense), and considering the concerning statistics above a look at how a defense held OSU in check is a welcome prospect. Statistically, what stands out in this game is OSU's greater than usual struggles with explosiveness: while they recorded roughly average success rates both on the ground and through the air (below their usual efficiency, but still fine), they were well below average in IsoPPP in both phases. Michigan State struggled to get OSU off schedule (47% success rate on standard downs), but once off schedule the offense struggled (20% success rate on passing downs).
Full chart is here, but I'll pull out the high points.
As is standard in modern spread offenses, OSU is primarily an 11 personnel team; while their most used single formation was a 2x2 four wide look (usually with four true receivers, as opposed to a team like Northwestern that prefers to move the tight end into the slot), various three receiver formations were an absolute majority of their plays before they started going heavy to lean on MSU late in the game, and even considering their late drives they ran 40 of 85 plays from three wide formations. The tight end in these formations is typically off the line and lined up with his body split by the outside leg of the tackle, though from time to time they put him on the line.
OSU showed a number of unbalanced formations, mostly three wide with a tight end or slot receiver covered but sometimes two wide with a tight end covered. I think they're likely to repeat this in the championship game, as it presents a dilemma for a quarters team about how to align the secondary.
OSU's runs leaned heavily on inside zone variants; I classified 25 of 43 runs as some sort of inside zone, though they vary the running back's path, the tight end's assignment, and how long they hold double teams before releasing to the second level. None of these variants were consistently successful, recording 3.0 yards per carry and just a 40% success rate. Whatever the variant, Michigan State's defensive line was hard to keep completely blocked, and their aggressive linebackers and safeties bottled up even the runs that crossed the line of scrimmage.
Ohio State's most successful run was the outside zone, which frequently caught overaggressive MSU linebackers shooting inside, like #28 below; six runs accounted for 47 yards and a 67% success rate. Northwestern will have to be disciplined about pushing this outside rather than diving at running backs they can't reach.
Option elements were notably lacking: while both Haskins and running quarterback Tate Martell sometimes ran out fakes after handing off, it's questionable whether these were true reads, as on a number of occasions the unblocked player seemed to commit too far inside but was not punished by a pull. They did run speed option once; Haskins pitched on the early side and their tight end blew a perimeter block, allowing MSU to stop the play for a loss.
The passing game suffered from two things. First, MSU did well at coming up and tackling on short routes. OSU runs a lot of mesh and shallow cross concepts; against MSU's zone defense, while these were reasonably successful at earning first downs, they typically resulted in a quick tackle before the receiver could extend the gain. This contrasts sharply with the way they shredded Michigan's primarily man defense last Saturday.
The second problem was simple poor execution. Haskins missed some throws outright, and on a number of plays he left the pocket and threw poorly on the run without being pressured. OSU's receivers dropped some passes, most notably a deep cross that should have gone for a long gain.
I don't want to take too much away from Michigan State's defense, which played their usual stout run defense, tackled well after the catch, came close to intercepting a few balls, and hit Haskins more than most of OSU's opponents, helping to encourage errant throws and poor movement in the pocket. If the Northwestern defense plays as well as MSU did, I will be quite pleased, and the team should have a real shot at winning. But OSU easily could have broken through on offense in this game with just a few better throws and catches, and with conditions indoors better than the cold, windy day in East Lansing on which this game was played, I expect that Ohio State's execution will be sharper on Saturday than it was for this one.
The blueprint for Northwestern is clear and basically the same as the usual plan: prevent big plays, stiffen in the red zone, and hope to come up with a few turnovers. For the most part, Ohio State is happy to work the ball down the field slowly, so I expect a number long, clock-chewing drives culminating in dramatic red zone confrontations. However, OSU should feel confident in their ability to keep Haskins upright, and their receivers present serious problems for an athletically suspect secondary. Furthermore, a strategy that keeps possessions down likely plays in Northwestern's favor. This suggests that Ohio State may be tempted to take more than their usual number of shots down the field. If they can connect on just a few, the game could get out of hand early; if NU can keep the downfield plays in check, things should be interesting.