Analytics, Basic Stats, and Recent Historic Context: The ABC’s of Penn State Football – Sean Clifford vs. Bo Nix, Relative Performance Index

Sponsor: Looking for the perfect Whiteout shirt for Saturday!?!?! Well, you’re SOL because there’s no way FANATICS is gonna get it to you by then. Show you’ve learned your lesson, procrastinator, by ordering Penn State gear for the remaining 5 home games using our Fanatics link HERE.

*Each FANATICS purchase helps support For The Blogy.

Introduction

Ah, Whiteout week. What a time to be on social media. Since Sunday the trash talk has flowed like stock footage of a mountain stream in a 1980s Busch beer commercial (speaking of trash talk, note to Auburn fans: James Franklin has been spouting his 1-0 mantra for years. Get over it). Crowd noise. Stadium size. Speed of athletes. Those trivial topics, and many more, have littered the Twitterverse for days now. One topic, however, that we’ve found to be not only interesting but decently researched has been the Jekyll/Hyde home and away statistical splits of Auburn QB Bo Nix.

The hyperlinked article (above) concludes with the following sentence: Penn State’s pass defense has given up 411 yards in 2 games and ranks no better than 68th nationally against the pass. Let’s see if Nix can take advantage. Fact check: OK that’s technically true. But the context of this stat is lacking. Against FBS competition, the Penn State defense has faced 41 pass attempts per game, which puts them tied for 15^th most in the country. The Nittany Lions have allowed 5.0 yards per pass attempt — which is tied for 16^th-best nationally — and opposing QBs have posted a collective 95.8 Passer Rating (15^th).

So, sure, they’ve given up some yards but – who cares? If Bo Nix throws for 205 yards at 5 YPA, Penn State will win Saturday. That being said, let’s take a real look at the differences between the home and away splits of both QBs spotlighted on college football’s grandest stage – Nix and Sean Clifford. How have Nix and Clifford been in a real context against their opposition? You’re about to find out!

Relative Performance Index

Here, we introduce a new concept of Relative Performance Index. It is a concept that has been used by analysts of the other football but I haven’t seen it applied as we’re about to apply it. For American football QB RPI, we will consider only three individual stats: completion-%, Yards-per-Attempt (YPA), and Passer Rating We will compare the game-by-game performance of each QB against that of their opponents’ averages using home-away splits. For example, in 2019 at Minnesota, Sean Clifford had a 53.5% completion rate, 7.9 YPA, and a Passer Rating of 113.6. The Gophers home averages allowed were 52.6%, 6.9 YPA, and 111.8 respectively. So, we can convert that to a percentage and find that Sean performed 2% better in Comp-%, 15% better in YPA, and 2% better in Passer Rating. If we apply this to all games, we can possibly start to gain a better understanding of how well a QB does at home versus away compared to the opponent expectation. Got it? Good.

Completion Percentage

Let’s start with the graph dujour. Below we have four columns of data – two for Bo Nix (left) and two for Sean Clifford (right). The “A/N” is the QB’s relative performance in away or neutral site games (as a percentage) against the opposing defenses’ performances at home. The “H” is for home games and uses the opponents’ Away/Neutral stats. The line in the light/dark gray split is the median average. And the total box is the total IQR (Interquartile Range) – 25^th to 75^th percentiles. The colors of the dots are season of play.

OK, so what does it mean? Well, What we can roughly infer is that during A/N games, Bo Nix averages a completion rate that’s ~98% of the opponents’ average and we can expect him to be somewhere between 78% – 113% of the opponent average. Sean Clifford, in home games, averages slightly higher than 100% of the opponent performance and should deliver between 90% – 107% of the expectation. This implies that Bo Nix might be more volatile in road environments than Sean Clifford. The Nix-H and Clifford-A/N splits don’t matter for Saturday but are there for reference.

So far in 2021, against FBS opponents, the PSU defense has allowed a comp-% of 63.4% and Auburn connected on 81.5% of its total throws. We would therefore expect Nix to complete between 50% and 71% of his passes. Since Auburn only played 1 game (against Akron) I don’t know what to make of the prediction for Clifford. If we use Auburn’s 2020 average where they allowed 63% completions, Clifford should end up between 57% – 67%.

YPA

When we turn YPA into an expected output percentage, we see some differences emerge between Bo Nix and Sean Clifford that could hint at what to expect on Saturday. In Away games, both QBs struggle, only outperforming the opponent average four times in their careers. At home, however, Nix and Clifford excel. And since the 2021 PSU/Auburn contest will be held in Happy Valley, location definitely favors Clifford.

When playing at home, Clifford exceeds opponents’ usual YPA allowed, posting a YPA percentage of 123% with a range of 99 -133%. In 2020, Auburn allowed 7.1 YPA (same as the single game in 2021) and 6.4 YPA in 2019. We’ll use 6.4 for baseline which implies Clifford’s floor should be 6.4 with a max of 8.5. Thus far, through the first two games, Penn State is averaging 7.8 YPA on offense so I think the range is very reasonable.

Bo Nix has an A/N average of 90% with a range of 82-107%. PSU’s defense, as mentioned, allows 5.0 YPA in 2021 and gave up 6.7 in 2020 in a shortened, Big Ten-only slate. For the sake of argument, let’s average these two seasons. When we do, this gives Bo Nix an expected range of 4.8-6.3 YPA.

Passer Rating

Finally, let’s look at Passer Rating through the same lens. Here again, Bo Nix performs worse on the road (IQR range of 77% -105%, exceeding opponent average allowed four times in his career) and Sean Clifford performs very well at home (IQR range of 103% -125%, exceeding opponent average allowed in every home start but two).

Penn State’s defenses in 2020 and 2021 averaged a Passer Rating allowed of 114. So, from this data, expect Nix to have a Whiteout Passer Rating somewhere between 88-120. This would rank in the bottom quintile of FBS.

Auburn’s defense gave up a Passer Ratings of 139 in 2020 and 126 in 2019 (just for the sake of comparison since 2020 was a dumpster fire). Giving Auburn the benefit of the doubt and using 126 as the baseline, Sean Clifford’s expected Passer Rating on Saturday should fall between 130-157.5.

Summary

Yes, Bo Nix performs worse away from Jordan-Hare whereas Sean Clifford does better in the friendly confines of Beaver Stadium. This week, in the battle of two similarly hot-and-cold QBs, the game is in Happy Valley and will be one of the craziest environments that college football can deliver. Therefore, history and data dictate that Sean Clifford will outplay Bo Nix this weekend.