August brings hope for college football fans across America. Before a single game gets played, you dream about a conference title and possible berth in the College Football Playoff. A high rank in the preseason polls only adds fuel to these desires.
However, not all teams live up to the preseason hype. There are more aspiring teams than conference title game winners, and a few losses in October can lead to devastated hopes.
Here, I'll use analytics to identify overrated team that won't meet expectations. My preseason model uses team performance over a four year period, turnovers and returning starters to rank college football teams.
My preseason rankings have predicted the winner in 70.8% of games the past 3 seasons, a rate that doesn't include FCS cupcake games. In addition, these numbers have helped me go 6-0 in predicting overrated teams on Football Study Hall in 2015 and 2016.
This perfect record only guarantees that my accuracy in identifying overrated teams will decrease. In addition, I'm tempting fate this season. I will not write about West Virginia, even though my numbers agree with Bill Connelly's analysis that they don't deserve top 25 consideration.
Instead, I'll start with the Big 12 team that lacks the defense to fulfill their top 10 rank in the preseason polls. I consider this my most confident selection.
Then we'll consider 3 teams that might have been one hit wonders last season. Coaching makes one of these programs stand out against the other two. Then I'll pick between the remaining two teams, which have striking similarities.
AP 10th, The Power Rank 31st, S&P+ 20th
After the 2011 season, West Virginia smoked Clemson by a 70-33 score in the Orange Bowl. First year coach Dana Holgorsen looked like a star, and the Mountaineers started the next season 11th in the preseason polls.
West Virginia went 7-6 in 2012 and failed to live up to the hype. They epitomized the bowl bounce, or the unrealistic preseason rank after a definitive bowl win.
Oklahoma State is getting the same bowl bounce in 2017. Last season, the Cowboys notched an impressive 38-8 win over Colorado, a team that had an outside shot at the playoff heading into championship week.
There are others reasons to like Oklahoma State. They've won 10 games each of the past two seasons, and that win total should have been 11 last year if not for some unfortunate officiating against Central Michigan.
In addition, Oklahoma State returns QB Mason Rudolph, who led the 11th best pass offense by my yards per attempt adjusted for schedule. The offense should be elite.
However, the Oklahoma State defense might hold this team back from a top 10 rank. From 2014 through 2016, this unit ranked 53rd, 66th and 58th in my defense rankings by yards per play adjusted for schedule. Not top 10 type material.
The defense has the most questions in the secondary where they lose their top 3 performers from last season. They will rely on Clemson transfer Adrian Baker to take a leading role at cornerback, but many young players will play important downs.
In the pass happy Big 12, the secondary is not a good place to have question marks. Oklahoma State's defense will have to make a drastic improvement over the last 3 years to live up to a top 10 rank.
AP 4th, The Power Rank 16th, S&P+ 7th
The Trojans were two distinct teams last year:
- Team A played the first four games, got blown out by Alabama and never challenged Stanford
- Team B won the next 9 games behind QB Sam Darnold
The combination of these two teams ranked 13th in my team rankings with a rating of about 13 points. This rating implies USC is better than average FBS team by 13 points on a neutral site, a similar level of play to their previous 3 seasons.
If you isolate Team B, USC had a rating of about 20 points, which would have ranked 3rd behind Alabama and Clemson. The pollsters are looking at this team in ranking USC 4th this preseason.
The potential of Darnold is clear, and USC brings back many important players on defense. This program has the talent to play at a top 4 level.
However, this program always has the talent to play at a top 4 level. In recent memory, USC hasn't performed at that level for a full season. Analytics implores you to look over a bigger sample than 9 games, and my preseason model ranks them 16th.
I also have my concerns about the coach. A traditional power doesn't have to make an unproven hire like Clay Helton, who has a lot to prove in his second year.
A rank of 4th leaves no room for error. USC needs to win the Pac-12 and make the College Football Playoff to justify this rank. While their schedule lays out an easy path, the numbers suggest they won't.
Penn State or Washington?
These two program have many similarities.
First, both programs jumped 10 points in their rating in 2016 compared to the previous two years by my numbers. This could serve as the definition of a one year wonder team.
Second, both coaches (James Franklin at Penn State after Vanderbilt, Chris Petersen at Washington after Boise State) had the break out seasons last year that suggest their success at previous programs wasn't a fluke. These programs have the coaching to become perennial top 10 teams.
Third, both programs bring back excellent quarterbacks. Trace McSorley of Penn State and Jake Browning of Washington both had break out seasons in 2017, and their return makes regression less likely.
My preseason model has both of these teams outside the top 10 (Washington 11th, Penn State 12th). Neither team has sustained a top 10 level over more than one year, so the model questions their ability to play at this level in 2017.
So which one is overrated? It comes down to the schedule.
Washington didn't exactly schedule a murder's row of out of conference opponents in Rutgers, Montana and Fresno State. They don't play USC this year, which leaves a road trip to Stanford as their toughest game.
In contrast, Penn State travels to Ohio State and hosts Michigan. They also have tough road games at Iowa and Northwestern. They're unlikely to beat Ohio State, and slipping up in two more games makes it tough to justify a rank of 6th in both preseason polls.
The College Football Win Totals Report
Based on these preseason rankings, I have win totals for all 130 FBS programs. To grab the free report, click here.