Ohio State’s formula for beating Penn State: Run the ball, generate havoc
Chad Peltier via our friends at Land-Grant Holy Land
Visit their fantastic blog and read the full article (and so much more) here
This is a championship-level matchup between two elite teams — and the stats are split.
Penn State, even more than Oklahoma, is the game we’ve talked about the most all offseason.
And not just in a simple “revenge!” kind of way — this game has become a bar for where Ohio State is at as a program, for what kind of legacy
J.T. Barrett will leave behind*, and whether the offense is actually “fixed.”
This game is a really, really big deal.
The hype for the game is justified, too. Penn State’s AP ranking — second in the country — more or less matches their ranking according to the advanced stats: 2nd in S&P+, 5th in F/+, 3rd in Sagarin, 5th in The Power Rank. It’s a legitimate top-3 or top-5 matchup.
(*If you’re asking me,
J.T. Barrett’s legacy as one of Ohio State’s all-time great’s is already assured.)
Ohio State is seemingly a much more talented team than Penn State — +34% blue chip ratio! — but the Nittany Lions have played like an elite team all season. Penn State ranks third in the S&P+ rankings, and while there is a sizable gap between their actual S&P+ margins — 29.2 to 22.1, which is the same difference as between Penn State and 15th-ranked South Florida — it’s easy to see why this may be the best game of the entire season in terms of overall quality of teams.
If you measure team talent solely based on blue chip ratio, then Ohio State is definitely far superior: Ohio State has 63 four- or five-star recruits on its roster, while Penn State has 34. The difference between those (29) is nearly as many as Penn State has in total. However, even without considering equalizers like player development or scheme/playcalling, Penn State’s average recruit has a rating of 87.89, meaning that while there are a lot of three-stars on the roster, a lot of them are near-four star. Ohio State’s average is second in the country at 92.13, but that’s not as sizable a margin as it looks when just comparing blue chip ratios.
Penn State is also significantly more experienced in terms of returning production. Their overall returning production ranking are 27th vs. 72nd.
And on top of their high overall talent, incredible offensive coordinator, and experience, the Nittany Lions also are among the best in the country in turnover margin. That’s a combination that will win you a lot of games.
When Ohio State has the ball
Good news: now we have opponent-adjusted measures for just about everything, including overall success rates and IsoPPP. That means that anywhere you see the “+”, the ranking is opponent-adjusted. I still list the raw success rate percentages as well, for context.
Penn State’s offense gets most of the love because of stars like
Saquon Barkley and
Trace McSorley. That’s justified. But the less-heralded Nittany Lions defense is also top-10 despite not having many (any?) household names for the average college football fan.
The Penn State defense has four main strengths:
- The secondary: 8th in passing S&P+, 11th in passing success rate, 5th in passing IsoPPP, 3rd in defensive back havoc rate
- Creating havoc, sack rate: 7th in overall havoc rate, 6th in defensive line havoc rate, 38th in adjusted sack rate, 22nd in stuff rate.
- Preventing opponents from finishing drives: 11th in average points allowed by opponent scoring opportunities (3.29)
- Field position: first overall (opponents’ average starting field position is the 23.7 yard line).
Several of those strengths create especially bad matchups for Ohio State.
But Ohio State also has a critical advantage, too: the offense could — even should — be able to run the ball very effectively.
First, let’s talk about the passing game. Since the loss to Oklahoma, J.T. Barrett has played like one of the top-3ish quarterbacks in the country, going from people calling for him to be benched to him re-entering the Heisman conversation. Kevin Wilson’s playcalling has had a lot to do with this, as have Ohio State’s opponents, but it is undeniable that we’ve seen real improvement in the passing game, regardless of the quality of opposing defenses. J.T. is playing with confidence unlike we’ve seen since the 2014 season.
And the advanced passing stats reflect that. Until this week we’ve had to use passing stats unadjusted for opponent quality, but now passing S&P+ rates the passing attack as the 6th-best in the country. So now they have to justify that opponent-adjusted ranking by facing a truly elite secondary. Two Nittany Lions,
Grant Haley and
Marcus Allen, have been
nominated for the Thorpe award. The team is 20th in average interceptions per game.
The secondary’s job is eased by a solid pass rush as well. You may remember last year’s game, where Barrett was sacked six times -- the good news is that
Garrett Sickels and his 2.5 sacks are gone, but Penn State is still 6th in overall DL havoc and 38th in adjusted sack rate this year. Ohio State’s offensive line is ranked 66th in adjusted sack rate.
Isaiah Prince has improved dramatically at right tackle, but
Demetrius Knox is still making his second-ever start against a great Penn State defensive line.
Knox looked solid against Nebraska, but Nebraska’s defensive line is also just 106th in adjusted sack rate. According to CFB Film Room, the guard spots are the most concerning in pass protection:
So there are two critical questions that the Ohio State offense has to solve. First, can the guards especially hold up in pass protection against a disruptive Penn State defensive front? Second, if Barrett does have time to pass,
as Ross Fulton said, “stretch the defense horizontally to exploit the intermediate zones between the run-focused front and three deep defenders”? Can they effectively constrain the Penn State secondary with screens and play action that provide “a vertical hi-lo stretch on the remaining deep zone defenders”, as Fulton again wrote? These concepts have worked extremely well against sub-par competition, but now they must work against elite defenders.
Two more related Penn State strengths: field position and finishing drives. The Penn State’s defense country-leading average starting field position is due in large part to their offense’s ability to move the ball: they rank 22nd in first down rate (84%) and 19th in FEI’s offensive drive success rate, meaning that the Nittany Lions offense rarely puts its defensive in a bad place field position-wise. Then, after forcing opponents to drive an average of 76.3 yards to score a touchdown, they are excellent at forcing field goals instead of allowing touchdowns. Penn State allows an average of just 3.29 points per opponent scoring opportunity, which is a first down inside the opponents’ 40 yard line. When the field gets shorter, Penn State’s defense gets tougher (much like Ohio State’s).
Combine an elite secondary with a disruptive defensive line, that often forces offenses to drive further on average then any other defense, and then limits offenses to field goals... then you’ve got a recipe for a more low-scoring game.
The Ohio State offense does have one key advantage over the Penn State defense, and that’s on the ground. Penn State ranks 50th in opponent-adjusted rushing S&P+. This is where opponent adjustments are key, because while the Nittany Lions are 13th in rushing success rate, allowing just a 33.8% success rate and not more than 3.77 yards per carry in any game this season, they’ve also not faced any great rushing offenses. Michigan found sporadic success with a few big plays, but still averaged just 2.45 yards per carry.
Penn State’s run defense is a little more complicated than just that, though. Even just looking at the opponent-unadjusted numbers, they rank 13th in success rate, 25th in opportunity rate, and 22nd in stuff rate — but 47th in adjusted line yards. This suggests that Penn State makes a significant number of tackles behind the line, but also allows a fair number of successful plays too — especially explosive plays. They rank 30th in rushing IsoPPP, but also 37th in allowing runs of 10+ yards, even against the relatively poor rushing teams they’ve faced.
All of this means that Ohio State should be able to run on Penn State.
J.K. Dobbins should pick up a number of explosive runs — and he’s a much different style player than anyone the Buckeyes had last year. Occasionally Ohio State has veered away from its running backs and relied on quarterback runs in big games: against top-15 S&P+ teams in the last three years, Ohio State’s RB/QB run rate is just 1.13. In their four losses, it is 1.01. Dobbins especially should get the ball often, and the Buckeyes should find explosive success relying on him and Weber.
The nightmare scenario is where Penn State’s secondary is able to effectively contain the passing game, allowing the Nittany Lions to commit more defenders to stopping the run, or vice versa, where the defensive line is so disruptive and creates so many negative plays on its own that the secondary gets more help.
But the stats suggest that while the Nittany Lions may get a few tackles for loss, they shouldn’t get many — Ohio State is second in allowing the fewest run stuffs. And they should also rip off a number of 10+ yard runs, too. The strong run game could be the key to Ohio State’s win.
It is true that
Penn State hasn’t played any elite offenses. It’s difficult to pick out the best offense they’ve faced, even.
When Penn State has the ball
The Penn State offense gets a lot of deserved love from the national media. Between Saquon Barkley,
Trace McSorley, and Joe Morehead’s playcalling, there’s a lot of talent on that side of the ball.
It’s easy to run through their strengths. Barkley averages 8.2 highlight yards per opportunity (meaning he’s very explosive), and is still efficient with a 44.4% opportunity rate. McSorley is an underrated runner, averaging 5.1 yards per carry but equaling Barkley’s opportunity rate. McSorley has improved in his efficiency too, now with a 46% passing success rate (up from 42.4%, and ranked 25th now). This is clearly a top-10 offense. And Ohio State’s secondary still really hasn’t been challenged since the Oklahoma game — the opponent-adjusted metrics for the passing offense have drastically improved, but Ohio State’s pass defense still ranks just 49th in passing S&P+.
But there are some clear weaknesses that the Ohio State should be able to take advantage of:
- The PSU offensive line: 117th in stuff rate (OSU is 7th in stuff rate), 105th in adjusted sack rate, 50th in adjusted line yards
- Finishing drives: Penn State averages only 4.66 points per scoring opportunity, which is 45th
Ohio State should really be able to exploit the PSU offense in these two areas, because they rank 5th in finishing drives, allowing just 3.03 points per scoring opportunity on average, 2nd in adjusted line yards, 7th in stuff rate, 7th in adjusted sack rate, and 8th in havoc rate. In short, Ohio State’s defensive line especially should create a ton of negative plays -- that, plus the defense getting more efficient the closer they get to their own endzone — gives the Ohio State defense a blueprint for slowing down the Nittany Lions.
Of course, as multiple coaches have said, Saquon Barkley is a transcendent talent. He rushed for 194 yards against Ohio State as a freshman, then averaged 8.25 yards per carry against the Buckeyes last season. And the Nittany Lions do rank 9th overall in rushing S&P+ and 12th in opportunity rate, meaning that they’re still an elite rushing team. But as you can see with their other offensive line stats, they still frequently allow negative plays. If Ohio State is going to win here, then the Buckeye defense will have to fulfill this expected advantage, stopping Barkley in the backfield while still remaining sound and disciplined.
The big concern for the Ohio State defense is still against the pass. Outside of
Denzel Ward, a second shutdown corner still hasn’t emerged, and we’ve seen slot receivers burn both cornerbacks and mismatched linebackers. And Penn State last year, and Indiana and Oklahoma both showed the ability to complete sideline passes in one-on-one coverage, too. This is a real, real concern for Ohio State — probably the most critical area for the Buckeyes to win.
Jordan Fuller was effective in the slot against Nebraska, so maybe he’s part of the answer. Ultimately I don’t expect the secondary to shut down the Penn State passing game or the defensive line to completely shut down Barkley — but they have to create enough negative plays to keep the Nittany Lions behind the chains.
The one good thing about the passing game is that Penn State is less explosive than they were last year, down from 4th to 49th in passing IsoPPP. In 2016 they had an explosive pass rate of 23.5% (which is the percentage of 15+ yard passes), but that is down to 21.7% this year. So both the rate and the magnitude of explosive pass plays are down this year.
Saquon Barkley is a big part of the passing game, almost equaling Daesean Hamilton’s team-leading 481 receiving yards despite receiving six fewer targets. Barkley vs.
Jerome Baker (or whoever else is responsible for him in coverage) is an absolutely key matchup. Hamilton averages 17.2 yards per catch too, so he’s still a major explosive threat.
One more note on the offense —
a lot has been made about Penn State’s fast starts. In fact, getting out to an early commanding lead over Michigan last week had a huge effect on the game. But in terms of efficiency, Penn State’s offense is excellent for the entire game. They rank 11th in first quarter offensive S&P+, but are 21st, 16th, and 3rd in S&P+ in the other three quarters. Interestingly enough, the defense takes a big dip in the fourth quarter, going from an average of 14th in the first three quarters to 62nd in the fourth. My guess is that could be due to defensive depth as players get tired. But that’s just a guess.
Intangible factors
There are also a few less-quantifiable factors that could affect the game. I don’t think any/all of these are enough to significantly affect the game’s outcome either way, but they might nudge either team up or down within their possible range of performances:
- Revenge factor for Ohio State
- PSU coming off big White Out win
- OSU’s bye week and extra prep time
- Alternate uniforms
I don’t know if these will have any effect at all, or if they do, who they benefit necessarily (I’d guess they’d
slightly benefit Ohio State if they have any impact), but I thought they were worth mentioning nonetheless.
Takeaways
So if you want the tl;dr, here are the key matchups that should determine the game:
- Ohio State’s pass defense is just 38th in passing S&P+, against the 9th-ranked passing S&P+ offense. Ohio State’s second corner and slot coverage have been notable weak points this season, and McSorley has gotten even more efficient (but also less explosive).
- But Ohio State should be able to create a lot of negative plays, both in run stuffs (7th vs. 117th) and sacks (7th vs. 105th).
- Another huge key will be taking advantage of Penn State’s relative weakness against the run (5th vs. 50th in rushing S&P+). Dobbins has to have an explosive game with more than 12 carries.
- The little things will matter a lot: turning Penn State scoring opportunities into field goals instead of touchdowns (5th vs. 45th in finishing drives, in OSU’s favor), field position (Penn State is 1st compared to OSU at 36th in offensive average field position), and turnovers always make a huge difference. I haven’t really discussed turnovers here, but obviously any turnovers could significantly swing the game either way -- and Penn State has a slight advantage in average turnover margin per game.
Picks
- S&P+: Ohio State 33.2, Penn State 23.5, 71.1% win probability
- F/+: Ohio State by 3, 56.8% win probability
- Adj. S&P+: Ohio State by 8.6, 69% win probability
- Power Rank: Penn State by 1.6, 45% win probability
- My pick: Ohio State 31, Penn State 27
As you can see, the stats are split on predictions. S&P+ is high on the Buckeyes, FEI much less so, and Power Rank favors the Nittany Lions. I fall somewhere in the middle, but there’s also a part of me that, like Doug Lesmerises said on
BuckeyeTalk (listen if you’re not already!), thinks that Ohio State rolls.
In the distribution of possible game outcomes, I think there are a fairly wide range of outcomes with a decent likelihood — most likely is a touchdown or so Buckeye win, and then nearly equal chances of a narrow Ohio State win or narrow Penn State win, and then much less likely is a Penn State win by more than 10 points or so.
Continue reading...
/cdn.vox-cdn.com/uploads/chorus_asset/file/9543269/10_26_2017_10_52_36_AM.jpg)
