Analytics, Basic Stats, and Recent Historic Context: The ABC’s of Penn State Football – Illinois Edition

Introduction

Penn State’s victory against Illinois seems to be an oddly satisfying conclusion to a bizarre, frustrating, and at times (namely the first five games) disappointing season. But if it must end, let it end with a thorough beat down of an overmatched and depleted team. By now, most people know that the team opted to not play in a bowl game this year and no one should be upset about that. For myriad reasons, this has been one of the most stressful years in recent history and the team, coaches, and staff certainly earned the right to return to some sense of normalcy with their families and friends. Not even the prestige of playing in a condiment or mortgage lending bowl was worth continuing with the restrictions placed on them. Let them come back fresh in the mid-winter and begin preparations toward a hopeful 2021 season. In the meantime, we’ll look back on 2020 over the course of the next few weeks and try to get a picture of what was and what might be moving forward.

Basic Statistics Summary

The results of the basic stats table vs. Illinois aren’t shocking. In fact, this type of domination – sans the 1st quarter — is what we all expected heading into the year. For the first time all year, the Lions won every basic statistical category – they split a few vs. Sparty, you’ll remember. The numbers below indicate just how complete of an effort this was in all facets of the game. Illinois had 138 yards and 21 points in the first quarter and only 135 yards and zero points over the final 3 quarters. The Penn State offense managed 156, 231, 71, and 123 yards, respectively, in each quarter while holding at least a 21-point lead in the entire 2nd half. 

Offensively, the 12.1 yards-per-pass-attempt is the best a Penn State offense has posted since the 2019 Buffalo game (12.7) and ranks as the 7th best since 2014. The 7.3 yards-per-play is the best of 2020 by a full 1.5 yards and is tied for 16th out of 88 games under James Franklin. Scoring 50 or more points has happened 12 times since 2014. Saturday’s scoring outburst was the 6th time the Nittany Lions reached/surpassed the 50-point barrier against Big Ten competition in the Franklin Era.

On defense, Illinois was limited to 273 yards. That’s the 5th time this season the defense has held the opposing offense to under 300 yards. The 120 passing yards allowed is the 14th best under James Franklin and 6 of the other 13 include pushovers like Idaho, Akron, Kent State, and Rutgers (3 times).

Look, we all “get” Illinois isn’t very good, lost their coach this week, and had several opt outs before heading to Happy Valley. But despite that, the Lions showed up and put a beat down on the “Fighting” Illini…and looked like they were having fun doing it. As much as anything, this visible enthusiasm HAS to be a positive sign. From an historic 0-5 hole, they fought back and salvaged the still-forgettable season and have lots to build from going into 2021. Perhaps NEXT year will be what THIS year was supposed to be? 

Advanced Stats

Today in Advanced Stats, we’re going to rank the Big Ten teams in several key statistical categories now that league play is over. Over the next couple of weeks we’ll look back on Penn State’s season and how it evolved with in-depth analysis, but for today, let’s just see where they ended up compared to the rest of the league (brace yourself for A LOT of Buckeye logos near the top).

Havoc Stats

We’ll start with the ability to cause and avoid havoc, which is a common discussion point here. As a reminder, a havoc play is one where the offense fumbles, throws an interception, allows a tackle-for-loss or a sack and Havoc Avoidance Rate (HAR) is the percentage of plays where one of those things doesn’t happen when your team has the ball. Therefore, the higher the HAR number, the better. Defensive Havoc Caused (DHC) is the percentage of plays where your team (in this case, Penn State) causes one of those outcomes versus the overall number of plays against – here too a higher number is better.

Indiana led the conference in both metrics with a HAR of 90.1% and a DHC of 18.4%. Ohio State was 2nd in DHC (17.4%) but was 12th in the conference in HAR due primarily to allowing 41 TFL in 6 games and allowing the 2nd most sacks per game. 

In both stats, Penn State was middle of the pack coming in at 9th in HAR (85.8%) and 5th in DHC (15.2%). We’ll see in coming weeks, but I’d guess the Nittany Lions’ HAR and DHC shot up a ton during the winning streak as the Good Guys improved at limiting turnovers, TFL, and sacks…and the defense became more aggressive. In terms of context, Penn State’s 2020 HAR is very much in line with how the team has performed against conference foes since 2016 with a range of 84.1% (2017) to 86.4% (2018) and the DHC value is slightly low except when compared to 2017. In 2016, 2018, and 2019 the Lions defense ranged between 16.6% and 17.3%. This may not seem like a lot but it averages out to 3 more TFL, sacks, or turnovers generated per game. So that matters…

Net Yards and Points per Possession

I’ve also been harping on this stat lately because Penn State has been good through the whole year at outgaining their opponents but then scoring at a lower rate (negative net points per possession). This was a function of turnovers, poor red zone execution, etc. The 2020 league performance is listed in the graph below. Penn State finished the season at +8.1 yards-per-possession only behind OSU’s 14.8. For how much grief we gave the offense, especially early on, they consistently moved the ball all year. But, for the first half or more of the season, they were not generating the points one should expect for all those yards gained. The gray diagonal line is the trend of the two metrics. 

Falling exactly on the line means a team is generating the “right” amount of net points for their net yardage performance. Teams far above the line are outscoring expectations, and teams below the line are underperforming. 2020 Penn State is FAR below the line. So, despite being 2nd in net yards, the Lions should’ve scored A LOT more. On a positive note, they ended the year 6th in net points and a positive value of 0.17. This improved over the back half of the year, thanks largely to strong play on both sides of the ball. 

Explosivity

Explosivity was a very weak point early in the season. The offense really lacked an ability to deliver big plays…but it steadily improved throughout the season. The calculation for Explosivity is Yards x Points/ 100-plays/Game. How productive a team is in a given number of plays. I multiply yards and points because even if a team averages 20 yards per play, if they don’t score, it’s irrelevant. For context, in Big Ten games since 2016, seven squads have had an explosivity value over 2.3 – 2017 OSU (3.07), 2018 OSU (2.59), 2019 OSU (3.27), 2020 OSU (3.09), 2016 PSU (2.44), 2017 PSU (2.43), and 2019 Minnesota (2.36). The league average since 2016 is 1.46. 

This year’s league values are here and Penn State actually finished 4th in the conference with an above-average 1.63. Yesterday’s 75-yard touchdown on the first play and 100-yard Lamont Wade kick return helped this metric considerably. When you consider how methodical Penn State’s offense was in October and November, the final and respectable Explosivity Rating by season’s end was quite a feat.

Still though, the gap between Ohio State and the rest of the league is substantial. Explosive teams are typically good teams and good teams are often explosive. For Penn State – or anyone, really – to take the Buckeyes crown, that gap must shrink. 

Offensive, Defensive, and Overall Effectiveness

We calculate offensive and defensive effectiveness through a mix of ball control (yards generated per time) and scoring and generate three stats: Off-Eff, Def-Eff, and Overall Effectiveness. For Off-Eff, higher numbers are better and indicate that a team is controlling the ball through yard generation and scoring at a reasonable rate. Def-Eff is the opposite and shows how well a team limits these things…here, smaller values are better. The Overall Effectiveness is the difference of Off-Eff and Def-Eff (higher is better; zero is exactly average). In the graph below, we plot Off-Eff on the x-axis, Def-Eff on the y-axis, and the label is the Overall Effectiveness (OE). 

We see Ohio State dominating the Off-Eff category and the traditional defensive stalwarts (Iowa, Northwestern, and Wisconsin) being the leaders in Def-Eff. If you look at the combination of Off-Eff and OE you see just how good Northwestern and Wisconsin’s defenses were. Both rank in the bottom 4 of Off-Eff but managed to finish 4th and 5th in OE. Iowa had the league’s best defense and finished 3rd in Off-Eff to have the league’s 2nd best OE. 

Penn State finished the season with the 5th best Off-Eff (30.0), 6th best Def-Eff (27.6), and 6th best OE (2.44). All of these values are the worst of the last five years! Prior to this season, PSU finished with the following OE’s: 2016 (20.4), 2017 (23.6), 2018 (10.6), and 2019 (16.5). But again, to finish the year in the top half of the league is still an accomplishment after the 0-5 start.

Summary

2020 was a season that started with incredible hope – Big Ten championship aspirations, possible playoff run, shiny new offense, lots of returning production – that quickly turned to despair during the first month-plus where the team looked lost, disinterested, and just bad. But looking back on the full year, you absolutely should give Penn State credit for digging out of the terrible start. It would have been soooo easy to quit and just pack it in, but I think we learned a lot about the players, coaches, and staff. Granted, the competition over the last four games (3 of the worst 4 teams in the league, and Michigan) wasn’t as strong as the first five games (OSU-1st, Iowa-2nd, Indiana-3rd in OE) but the two losses against Nebraska and Maryland showed that the Nits weren’t only failing against the best in the conference. They improved and now have something to build on going forward. Here’s to a more normal 2021!