clock menu more-arrow no yes

Filed under:

Ohio State blew out Northwestern, but let’s nitpick with analytics anyway

New, 1 comment

You have to dig deep to find holes in their performance — but there are some!

NCAA Football: Ohio State at Northwestern David Banks-USA TODAY Sports

Ohio State’s 52-3 win over Northwestern was a more dominant performance than I expected heading in to the game — something I’ve thought every week since week 1.

No one gave (or should have given) Northwestern’s offense much of a chance to do anything, but there was still a world where the Wildcats make the game ugly enough to put the game in doubt for a little while. While the Wildcats’ defense did find some success in the first half, crazy third down efficiency, amazing red zone passing, and several very explosive runs ensured a blowout.

OSU vs. NW EPA percentages

Stat OSU offense OSU defense
Stat OSU offense OSU defense
Avg EPA 94% 82%
EPA SR 66% 90%
Avg EPA Rush 77% 54%
EPA SR Rush 25% 73%
Short-Yd Rush EPA 58% 49%
Avg EPA Pass 92% 91%
EPA SR Pass 94% 96%
RZ EPA 85% 81%
RZ EPA SR 60% 95%
Std Down EPA 23% 66%
Pass Down EPA 97% 76%

Let’s do some quick bullet points before moving on to Wisconsin. EPA refers to expected points added,

  • In the table above, I’ve translated the game EPA scores to normalized percentiles. So a 94 percent in offensive average EPA means that there was only a six percent chance of getting the EPA they Ohio State did, based on all other offensive performances this season. That is equal to 1.52 standard deviations above the mean for average offensive EPA.
  • The difference in average EPA and EPA success rate (94% vs. 66%) is primarily due to the major difference between average rushing EPA and rushing EPA success rate -- 77 percent compared to 25 percent success rate. Only one in four runs was efficient, adjusted for down, distance, and field position -- but the successful runs were very successful.
  • For example, the three long runs -- Master Teague’s 73-yarder, Marcus Crowley’s 53, and Dobbins’ 68 yard run -- combined for 12.23 expected points added. That’s pretty unreal and speaks to the value of explosive plays (even if the most important predictor of explosiveness is efficiency).
  • While those long runs were great, and it’s hard to be too critical of a 45-point offensive performance where the starters were pulled for the last quarter, the Buckeyes will face a number of really good run defenses this season and they will need to be a little more efficient when the opposing offense might actually be capable of scoring in return. Twenty-two percent of the Buckeyes’ runs were stopped at or behind the line of scrimmage.
  • The problems mainly seemed to come on second down, where they had just a 25 percent success rate, compared to 45 percent on first down and 57 percent on third down.
  • Compare the rushing EPA numbers with the passing performance. At 92 percent and 94 percent, the offense was both efficient and explosive relative to expected down, distance, and field position.
  • As expected, Justin Fields’ insane third-and-15 completion to K.J. Hill was the top passing EPA play of the night (+3.93 EPA), followed by the touchdown passes to Dobbins and Chris Olave.
  • The final thing to note offensively is the disparity between standard and passing downs EPA, which were 23 percent and an astounding 97 percent. Their actual standard vs. passing down success rates weren’t as divided though, at 33 percent vs. 44 percent. The good news is that this is an atypical performance for the Ohio State offense, because relying on completing third-and-long passes is unsustainable. (Note that Ohio State’s third down efficiency is typical, as they lead the country in third down efficiency, but usually they’re more efficient on standard downs too and so don’t have as far to go on third.) That was kind of what happened with Clemson and Alabama in last year’s national championship game. The Tigers improbably converted nearly every third-and-long in a way that they might be unlikely to consistently repeat if the two teams played 10 times.
  • Ohio State’s defense obviously played well, as you would expect against a really poor Northwestern offense. But Baron Browning’s absence was definitely felt at middle linebacker, as the Buckeyes’ average rushing EPA was only 54 percent.
  • I think part of that relatively poor run defense was due to missing Browning, but it’s also reasonable to be a little concerned with Wisconsin coming to Columbus next. Ohio State hasn’t faced an offense that was really capable of exploiting any holes in the run defense. And in all honesty, they haven’t faced an elite offense of any kind, but the Badgers definitely have a strong rushing offense (albeit one that can be slowed, as Illinois showed).