Analytics, Basic Stats, and Recent Historic Context: The ABC’s of Penn State Football – The 2020 Offense Retrospective

Introduction

Before we turn the page to 2021, let’s take one last look back on the 2020 Penn State Nittany Lions. What trended well? What were the consistent strong/weak points? And, finally, what needs to be fixed in the offseason. Today, it’s the offense’s turn under the microscope. As we always do, this season-ending summary will be a thorough mix of basic and advanced stats within the broader context of the James Franklin era at PSU. 

Basic Statistics Summary

Our first graph shows the yards generated in each game. Passing yards (blue), rush yards (orange), and total yards (red) .

With the exceptions of Indiana and Nebraska, Penn State really struggled to run the ball early in the season and relied heavily on the pass game to generate yards against Ohio State, Maryland, and Iowa which were (probably not) coincidentally the three biggest losses. Penn State’s lack of a rushing attack against good defenses like the Buckeyes and Hawkeyes –two teams that typically have strong, opportunistic defensive lines – left the Lions offense vulnerable to sacks and turnovers. If we look at passing yards as the % of total yards generated (Pass Share) in the graph below, you can see how much the offense skewed towards passing, especially early on.

In three of the first five games, Penn State generated at least 78% of its yards through the air. In the other two games (the closer losses) the Nittany Lions were much more balanced. In their first two wins, Penn State flipped the script and became fairly run dominant with at least 60% of the total yards coming on the ground. In Kirk Ciarrocca’s most successful years (2019 Minnesota, 2015 and 2016 Western Michigan) his teams have been between 54-59% pass share. Therefore, I don’t think Penn State’s pass-heavy start to 2020  was by design. For the year, Penn State’s average Pass Share was 61% — a higher percentage than Ricky Rahne (54%) but lower than John Donovan (66%) or Joe Moorhead (63%). Interestingly though, a high Pass Share does not always mean a loss is imminent. Since 2014, Penn State has had a Pass Share of 70+% in 29 games and is 20-9 in those contests. 

The yardage wasn’t really the problem, though. Have a look at the average yards generated in conference play since 2014. The 2020 team produced at a rate that nearly rivaled the 2016 and 2017 years, which we all look back on very fondly. No, the 2020 team had other issues and we’ll cover them in due time…

Before we dig deeper, let’s look at scoring. All of the yards in the world don’t mean a thing if you don’t generate points. After the Indiana game, the offense failed to generate more than 30 points over the next six games, averaging only 23 points from Week 2 to Week 7. 

The scores of 35, 39, and 56 are much better and when taken as a whole, it was comparable to 2018 and 2019 (see below). In 2018 Big Ten games, Penn State averaged 28.4 PPG. The next year, that average crept up to 30.1. Yes, this year’s 29.8 PPG was lower than peak 2016 and 2017 seasons, but wasn’t SO LOW that a sub-.500 record should have been expected. There must be something else that was different this year…some other trend that torpedoed things.

Advanced Stats

So, the total yardage generated with Ciarrocca at the helm  was basically at Joe Moorhead levels, and the scoring was at least as good as what Penn State posted under Ricky Rahne. So what gives, man? Tomorrow we’ll cover the challenges that the defense had this year, but how much of that was a function of the offense? 

Possession Efficiency

I’ve covered Possession Efficiency nearly ad nauseum at this point, but it’s worth mentioning again. For the newbies, Possession Efficiency is a team’s ability to consistently generate yards and points each possession. Today, we’ll start with a baseline of Big Ten performance from 2016-2020 in the graph below. You see a good correlation to yards-per-possession (x-axis) to points per possession (y-axis). Today, we’re going to use the equation of the trendline to measure Penn State’s game-by-game performance. The trendline equation is below and it says that a team should generate points at a rate of: Pt./Poss. = 0.103* Yard/Poss. – 1.03. 

Looking at the Penn State actual game-by-game performance, we get the following…which shows the yardage and scoring efficiency with the total points as the label. 

But how do these actual outcomes compare to what Penn State should’ve scored? The graph below shows how many more or fewer points Penn State should’ve scored based on its yardage efficiency. You can see that for three games (Indiana, Ohio State, and Iowa) they were pretty close to expectation. Sadly, for OSU and Iowa, expectation was bad and these were just poor days for the offense. But against Maryland and Nebraska they underscored expectation by 12 and 17 points respectively (Penn State really should’ve beat Nebraska). Against Michigan and Rutgers they should’ve scored another touchdown or field goal. In the final two weeks of the season, the Michigan State and Illinois games, Penn State beat its points projection by at least two scores. 

Exceeding scoring expectations vs. Michigan State and Illinois makes sense when you consider the special teams touchdowns (which is good). But really the Maryland and Nebraska games stick out as the heights of frustration for fans. Poor red zone play and turnovers kept points off the board and were big reasons for these large deviations from expectation. Speaking of Red Zone…

Red Zone Offense

“They’re going to get to the 10 and then fail to score,” I yelled at my TV during 4th quarter of the Nebraska game. You could see it coming. The offense moved the ball extremely well all game but then threw fade after desperate fade into the end-zone. The offense’s red zone production is shown in the plot below with the x-axis being total red zone attempts, the y-axis being the average points per red zone attempt and the label as the scoring-% in the red zone (either TD or FG).

Iowa just beat Penn State this year. The Lions offense made the most of their single red zone opportunity but it wasn’t nearly enough. The two games in the bottom right of the graph are what sticks out most to me. Against Indiana (which was a one point game, mind you) Penn State had six red zone attempts and only scored on 3 of them. Against Nebraska, the Lions averaged less than a field goal on six red zone attempts. Since 2014, Penn State has averaged some type of score on 88% of red zone opportunities and achieved 5.1 points per RZ attempt (PPRZA). Average production in either of these categories likely means both Indiana and Nebraska are wins, not losses.

Overall, 2020 (81% scoring and 4.7 PPRZA) was the worst performance for a Penn State offense since 2014 (80% RZ scoring and 4.1 PPRZA). These metrics obviously go hand-in-hand but you can score at a lower rate as long as your generating touchdowns OR you can have a very high scoring rate even if settling for field goals. Being down in both is death.

Havoc Avoidance Rate (HAR)

Havoc avoidance is an offense’s ability to prevent bad things (sacks, TFL’s, fumbles, interceptions) from happening as a function of total plays. An average Big Ten offense will have a HAR of 86% — meaning that 86% of the time they didn’t have a disaster. Since 2016, this is a metric in which Penn State’s offenses have been consistently average (2016 – 86%, 2017 – 84%, 2018 – 86%, 2019 – 86%). For comparison, Ohio State tended towards 88+% over the same time (except 2020 at 85%). And as I’ve mentioned before, this 2-4% difference may not seem like a big deal but when an offense runs 70 plays per game, 3% is 2 more turnovers or drive killing sacks or TFL’s. And when games are decided by one or two possessions, well, it matters a lot.

Penn State’s game-by-game HAR is marked below along with the havoc plays allowed. Starting with Michigan (HAR of 96%) the team generally took better care of the ball and limited bad plays. Even in the losses to Indiana and Nebraska, HAR was above average. However, In the other three losses, the HAR was 80% or less, which is dreadful. A team needs to consistently maintain high HAR and RZ Efficiency for success.

Explosivity

Offensive Explosivity in calculated by: Yard x Point/Play/100 and is an indicator of an offense’s ability to move quickly down the field and score points. Early in the season, this was a metric where the PSU offense struggled considerably. But the Lions managed to work their way to average towards the back half of the season. The plot below shows all Big Ten offenses in explosivity versus scoring from 2016-2020 on a per game basis. The 2020 PSU squad is in the box with the label. The two outlier (in a good way) PSU logos are the 2016 and 2017 teams. As you can see, the 2020 Lions didn’t come close to matching those two squads’ explosivity, but they finished in the same neighborhood as the 2018 and 2019 teams both in terms or points scored and explosivity. 

The challenge for the 2020 team wasn’t scoring points…it was the time it took to score them. They finished with an Explosivity Rating of 1.62, but through 7 games that rating was a meager 1.29 – including the clunker Maryland (0.89) and Iowa (0.98) performances. While the numbers themselves are somewhat intangible, you can see from the graph that anything above 2.0 separates a team from the pack of the conference and anything <1.0 is pretty bad. It took the 2.26 (Michigan State) and 4.06 (Illinois) ratings to achieve end-of-season respectability. This will be something (along with the red zone production) to work on in the offseason. This offense was VERY hesitant to take shots downfield. That has to change.

Overall Offensive Effectiveness (OffEff)

To summarize all of this, we have the overall offensive effectiveness (OffEff) value where an average team is 27 and this is calculated by a lot of what is described in this article. Against Big Ten competition, this year’s Penn State offense finished with a value of 30.0 (5th in the conference). Somehow, Penn State’s offense wasn’t that bad when taken as a whole. As discussed, there’s substantial dividing line – a shift or awakening, if you will — from the first five games to the last four, and those end-of-year performances really bolster a lot of these numbers. Since 2016, this ranks as Penn State’s 4th best offense and is miles away from 2016 and 2017 (41.1 and 42.1 respectively). This squad was more in line with the 2019 (31.4) and 2018 (29.7) offenses. And for context, over the same time period, the conference benchmark of Ohio State had a low of 40.3 in 2016 and a high of 55 in 2019. We, as fans, all want to see Penn State win another conference title and to get there, Penn State’s offense must compare to the best in the league. And in that regard, there is still some distance to go. 

Pattern photo created by jcomp – www.freepik.com