Analytics, Basic Stats, and Recent Historic Context: The ABC’s of Penn State Football – The Jim Knowles Effect

Sponsor: Looking for last-minute seats to Saturday’s Big Game? Go with a trusted source, STUBHUB. Be sure to use our FTB STUBHUB LINK so that we too can wet our beaks in the deep pool of Service and Convenience Fees.

Introduction

If you haven’t heard, Penn State welcomes its toughest opponent of this year (and seemingly every year) as Ohio State visits Beaver Stadium for a Saturday Gusgasm-filled nooner. The leaf sticker wearers have brandished a lethal offense for close to a decade now. Normally, that potent attack has been complemented by a defense full of future multi-millionaires performing as a cohesive, stubborn unit. I say “normally” because 2020 and 2021 were certainly exceptions to the standard.

Enter Jim Knowles.

The bearded bastion of defense started his career as the DC at Duke from 2010-2017 before moving to OSU-Central (Oklahoma State) from 2018-2021. For various reasons, both of those posts come with a litany of challenges – Duke with its talent deficiencies and Okie State plays in the Big XII (nuf said). And yet, at both spots, Knowles somehow made it work. And though the 2022 season is far from complete, thus far, Knowles has taken Ohio State’s full cupboard of talent and orchestrated a substantial defensive turnaround…evidenced in the various basic and advanced stats we’re about to present.

Jim Knowles History

For this section, let’s examine how Knowles’ defenses have done over the years vs. FBS averages. Specifically, we’ll focus just on defensive efficiency (yards and points per drive) and then overall effectiveness. The red diamonds below represent the defense commanded by Knowles with the associated label. The line going across the chart is the global FBS average and every gray dot represents one FBS team from that year.

For yardage efficiency, we see that an average FBS defense allows roughly 32 yards per drive. In Knowles first few years at Duke, his defenses finished below average in this particular metric. BUT, from 2013 until his final season in 2017, he got the Blue Devils at, or in the case of 2017, much better than average in yards per drive. That season, the Blue Devils finished 7-6 capped with a Quick Lane Bowl win over Northern Illinois inside that NFL Stadium Up North. So, during Knowles’ time in Durham, his defenses gradually improved – a recurring theme throughout his coaching career.

When Knowles inherited the DC whistle at Oklahoma State in 2018, his inaugural defense once again finished below the FBS average in yards-per-drive. Very quickly, though, Knowles flipped the script. In 2021 – his final act – the Cowboys posted the 2^nd best yardage efficiency mark in the country (behind only Wisconsin). Considering the Big XII doesn’t feature car-on-cinderblock offenses like Iowa, Northwestern, Illinois and Rutgers, there’s no shame in finishing second, Ricky Bobby be damned.

What’s scary is that through 7 games, the 2022 Buckeyes have been even better in this key metric, surrendering a measly 20 yards/possession.

It’s pretty much the same story when we move to points/drive – tough start at Duke but he eventually got them into shape, mediocre first year at Oklahoma State, grand finale with a top-10 national ranking, and then even MORE improvement in his first year at Ohio State. Turns out the learning curve required to master Knowles’ scheme shrinks dramatically when roster talent overfloweth, like it does at Ohio State.

Same trajectory when measuring Defensive Effectiveness (labeled Final Calc in the table below). In 2021, the Oklahoma State defense was 5^th nationally (Baylor was the next best Big XII team at 16^th). The 2017 Duke team was 26^th (Duke – in football!). And now, we have Ohio State sitting at 4^th nationally – the rich get richer…

Ohio State and Advanced Stats

Compared to the rest of college football, Ohio State hasn’t had bad defenses recently. With that said, the Buckeyes 2020 and 2021 defenses certainly didn’t look or perform like an Ohio State defense. As we find in the graph below – which charts the yards allowed/drive for OSU over the last 14 seasons – Knowles’ impact has been immediate as the Buckeyes are back to being one of the top defensive units in FBS. They are currently 2^nd in yards allowed per drive at 19.00 (trailing Illinois at 18.5). For those scoring at home, Ohio State’s 19.99 yards/drive is the 5^th best of all FBS teams (1,700+ squads) since 2009. So, yeah, that’s impressive.

So Knowles has limited offenses’ productivity…but how? What additional stats have contributed to these lofty mid-season FBS defensive rankings? Here at the ABC’s, we lean heavily on Red Zone scoring, explosivity prevention, and HAVOC when examining and comparing different defenses. In each of these metrics, the Buckeyes are markedly better than they were last year. We start with points/Red Zone attempt (a FG is 3 and a TD is 7). The FBS average is 4.9 points allowed per Red Zone trip. In this metric, the 2020 and 2021 Buckeyes were far too generous, finishing well below average. This year, they’re allowing a full point less per Red Zone visit – which is a substantial improvement.

But the points/RZ attempt only tells half of the story. To gain a fuller perspective, we need to look at a measure of total RZ attempts. We analyze that by calculating RZ points allowed per game (RZ attempts * points per RZ attempt). In this metric, the Buckeyes are better than they’ve ever been with 7.1. This means that opposing offenses are only getting into the RZ about 1.5 times per game versus an average FBS defense, which allows 3.8 RZ attempts per game. Again, impressive.

To measure explosivity – our in a defense’s case, explosivity prevention — we like to use a homemade stat that calculates TD points scored from outside of the RZ – what we call TORZ. An average FBS team allows 9.2 TD points from outside the red zone (TORZ). Somewhat surprisingly, the 2022 Buckeyes are allowing a 7 TORZ (1 explosive TD) per game. That’s actually a step back from 2021. Not quite sure how to explain this. Considering Penn State’s vertical passing game finally showed a pulse vs. Minnesota, this might be one area where the Nittany Lions could find some success against Knowles’ defenders.

Regular FTB readers know we’ve beaten HAVOC to death over the past three weeks in our examinations of Manny Diaz’s defense. The Nittany Lions have been historically good in this metric, thanks to an ungodly number of PBUs. For the newbies, a HAVOC play is a turnover, tackle-for-loss, sack, or pass break-up. HAVOC percentage is the sum of the HAVOC plays divided by the total plays against. Penn State is generating about 21 HAVOC plays per game. The Buckeyes are slightly better than average at 14.1, which is down from their top-end total of 19.8 from 2019. Of course, total HAVOC plays can be a bit misleading…because it’s hard to generate a ton of HAVOC plays if your defense isn’t on the field much.

The Buckeyes are only defending 57 plays per game (FBS average is 70). This is the 7^th fewest plays per game for an FBS defense since 2009. Except for 2011 Alabama, the other 5 teams ahead of 2022 OSU were service academies – which makes sense since they don’t pass the ball, meaning the clock rarely stops, thus shrinking the game.

Taking HAVOC plays with total plays we see that the 25% HAVOC rate for the 2022 Buckeyes defense is well above average (actually, the 2019 OSU had the best HAVOC rate of any FBS team since 2009). Anything over 25% is top-10% of all FBS teams since. So even though the Buckeyes D isn’t on the field much, when do step between the white lines they tend to make life hell for opposing offenses.

When we put it together and look at their points/drive allowed we see they’re a very good 1.24, which isn’t the best mark from an OSU team during the past decade but is still amazing (42^nd best in FBS since 2009) and greatly improved versus 2020 and 2021.

I want to end with a comparison between the Penn State and Ohio State defenses as it currently stands. Michigan game aside, Penn State has performed pretty well on defense (ranked 13^th in Overall Effectiveness). But that’s still well below what the Buckeyes have done. In TORZ allowed, Penn State has a slight edge. The HAVOC rates are similar but Penn State faces 72 plays per game, giving them 15 more opportunities to make a sack, recover a fumble, etc. And for the point and yard per drive efficiencies, the Nittany Lions are better than average but not close to Ohio State.

Does this mean that the outcome of this game is a foregone conclusion and the only suspense isn’t whether Ohio State will win but by how much? No, absolutely not. But for Penn State to keep it close (or pull off an upset) the defense will have to “break serve” a half-dozen or so time, be it via turnover or punt, and the offense will need to generate a half-dozen explosive plays and protect the football. Wouldn’t hurt if 107,000 slam a few Red Bulls to replicate last Saturday’s 7:30 p.m. atmosphere at Noon. It’s unlikely but it not out of the realm of possibility either (see 2016). We Are.