2022 College Football Season Preview: Power 5 Team Tiers
Clustering teams to get an idea of where everyone stands this season
After a summer of conference realignment, hot takes, and the transfer portal, we have finally made it to the season. This past Saturday was the last Saturday of the offseason, so it is finally time to preview the season awaiting us in just a few days. To accomplish this, we turn back to the bread and butter: our friendly neighborhood clustering algorithm!
Here are the variables that will be fed into the clustering algorithm which will then in turn group together the teams into various clusters:
Betting Win Totals (via Draftkings Sportsbook) - If you don’t partake in sports betting, a win total is the number of wins a book sets on any given team. You can then bet on whether that team will go over that set amount of wins, or under that win total. Even if you do not bet, the betting markets can provide a ton of useful information on the strength of the program.
ESPN’s FPI: First, this includes the actual FPI number, which represents how many points above or below an average team is. I also included the projected wins/losses that come from those FPI metrics, as well as a teams odds to win their division/conference/playoffs etc.
KFord Ratings - From Kelley Ford, this power rating is scaled from 0-100 with 50 representing the FBS average. These ratings are designed to be purely predictive, or forward-looking.
Returning Production - As we learned from a previous data exploration, returning production metrics are indicators of whether a team will improve/regress from their previous season. The numbers included are from ESPN’s Bill Connelly’s returning production numbers, plus the returning usage of offensive skill position players (QB, RB, WR, TE). Finally, we also include the returning Predicted Points Added (Think EPA’s cousin) from said returning players.
A Few 2021 Metrics - The last few variables deal with how the team did last season. This includes raw advanced metrics like EPA/Play or Success Rate, as well as opponent adjusted metrics like SP+ and F+.
We have our variables. Now all that is left to do is get to clustering. Once we have our clustered groups, we can arrange them based off the groups average preseason win total in order to get a more tiered list.
Power 5
The “Contenders” Tier
Let’s start with the Power 5. Alabama, Ohio State, Clemson and Georgia being in the contenders tier should come as no surprise to anyone. Each of these teams has a greater than 57% chance to reach the playoff according to FPI. It is almost a forgone conclusion that these teams will reach the playoff. Why almost? the one question mark among these four teams lies with the Clemson offense. Tigers QB DJ Uiagalelei had a season to forget last year: At -0.08 EPA/Play, Uiagalelei ranked 110th among 124 qualifying QB’s last season, and his 66.3 Offensive PFF Grade ranks well into the bottom half of returning FBS QB’s. If Clemson can figure out how to get the most out of Uiagalelei’s tools, they should make a return to the playoffs.
The “Next Teams Up” Tier
The “Next Teams Up” are teams that could potentially spoil the contenders seasons and get into the playoff if things go right. Notre Dame should continue to be strong even as they transition into the Marcus Freeman era. Texas A&M’s win total is 8.5 this year, but that doesn’t feel like it matches the expectations of the Aggie faithful. Jimbo Fisher hasn’t reached the promise land yet, but it feels like a potential title window is creaking open. NC State beat Clemson last year for the first time since 2011. While that didn’t lead to an ACC title appearance, the Wolfpack feel as though their time to take the conference is now. QB Devin Leary returns to NC State for one final season, and his play will dictate how high NC State can finish this season. If Clemson can’t fix their offense, all eyes turn to NC State in the ACC Atlantic.
The “Most Interesting Teams In CFB” Tier
If you’re someone who is wondering what to watch for since the playoff teams are basically penciled in, the “Most Interesting Teams in CFB” cluster is exactly for you. Quinn Ewers looking to elevate Texas back to its glory days, the Cristobal era in Miami, another year of Hendon Hooker in the Heupel four verts offense, and so much more. This cluster is jam packed with storylines. This tier returns the most production on average compared to the rest of the tiers, as well as the second most passing PPA behind the contenders tier.
If QB’s are your thing, this tier features SIX QBs that finished in the top 15 of the QB Composite (Purdue’s Aidan O’Connell: 3rd, Wake’s Sam Hartman: 7th, UVA’s Brennan Armstrong: 9th, Louisvilles Malik Cunningham: 12th, Utah’s Cameron Rising: 13th, Miss States Will Rodgers: 15th). That doesn’t even include Kentucky’s Will Levis and Miami’s Tyler Van Dyke, who finished outside of the top 15 but have garnered NFL draft interest. This tier is going to prove why shifting your focus away from the playoff will make your CFB viewing experience a lot more positive this season.
The “A Lot To Prove” Tier
The average win total difference between the previous tier (7.5) and this tier (7.22) is relatively small (Oklahoma’s win total is still higher than Texas Sooner fans, nothing to worry about here). The biggest difference between this tier and the one above it is simply returning production. This tier ranks last among all tiers in overall returning production and returning pass PPA. Pittsburgh (Kenny Pickett), Ole Miss (Matt Corral), UNC (Sam Howell), Iowa State (Nebraska) lost their QB’s to the Draft. Nebraska, Arizona State, Virginia Tech has both lost and replaced their starting QB’s via the transfer portal. LSU, Oregon, Florida, USC, and Virginia Tech ring in the new season with a new head coach at the helm. This tier has a lot of new faces, and im excited to see how some of these programs respond to changes.
The “Proof Of Concept/Foundation Building” Tier
In every coaches time there comes a point when its time to see a proof of concept of what they’re building. That point seems to have come for both Auburn’s Bryan Harsin and Florida State’s Mike Norvell. For Harsin, he was almost ousted after his first year at the helm for Auburn. Their 6-7 record left a lot to be desired, especially when you look at their win probability in the second half of their losses:
Harsin to Auburn always seemed like an odd fit, so the chair could be pulled at any time during year two if things appear to be heading south. For Mike Norvell and FSU, its bowl season for bust. FPI gives them an 80% chance to reach 6 wins, and their win totals seem to hover around 7 wins. Ask any Seminoles fan and the expectation would climb to 8+ wins, but it seems like anything in the bowl eligible range will keep Norvell safe till 2023. The schedule does feature some challenges and a gauntlet of good ACC QB’s, but that doesn’t matter, the expectations remain high. Syracuse’s Dino Babers has shown a proof of concept before, reaching 10 wins in 2018. After two 5-7 seasons and a 1-10 season, however, its time to see if this is going to get turned around. If not, we could see Syracuse making a change sometime soon.
The rest of the teams (aside from Kansas State) are looking to build some foundation to their programs. South Carolina’s Shane Beamer guided the Gamecocks to a 7-6 season last year, which far exceeded most expectations. Now he has Oklahoma transfer QB Spencer Rattler, who is looking to prove he is in fact that elite QB everyone expected him to be. For a team that constantly has to punch above their weight class year after year, there is certainly some bright spots for the Gamecocks. Its been a long time since Boston College had an elite college QB or elite offense. Now that Phil Jurkovec is healthy, we could see a level of QB play we haven’t seen since Matt Ryan was on campus. Finally, the Red Raiders might finally be coming back to their Air Raid, juggernaut offense days with former WKU now TTU OC Zach Kittley. New HC Joey McGuire seems to be a very good fit for Texas Tech, so they will be a team to keep an eye on in the future.
The “Just Not It” Tier
Its just not your year. The average win total is 3.45 and none there just isn’t really much going on here. These teams are rated poorly by advanced metrics, and aren’t favored in many games. The biggest highlight is Georgia Tech, who have this gauntlet of a schedule:
In a must win year for Geoff Collins, that is 100% not the schedule you want to see.
Conclusion
For those that are interested, here is how the tier averages look on some of the more notable metrics:
Those are the power 5 tiers! Later on in the week we will tackle the group of 5 to round out the 2022 College football season preview. It is TIME to get excited for College Football!
If you want to dive in to the data like I do, check out @CFB_Data and @cfbfastR on Twitter, where you can learn how to get started in the world of College Football data analysis!
If you want to see more charts and one off analysis, follow my twitter page, @CFBNumbers