Last year I told you to bet on Beckman, which was correct... in a sense
I thought a fourth kid had put me into full-fledged retirement, but apparently there are few of you degenerates out there who still think this is good advice to put your [theoretical] dollars behind. Here it is, your 2015 Stock Watch.
Before we get into this season, the annual transparent review of the prior year’s predictions:
I may not be sold on Tim Beckman, but my numbers are high on [no longer] his Illinois team this year. 762 out of 1,000 scenarios run have the Illini exceeding their projected 4.5 wins this year, with over half putting Illinois in line for a bowl bid.
A big ball of mediocre. Even more pronounced than the Big Ten, I have everyone but BC (over 4.5 wins) within 1.2 games of the Vegas win total. On top of that, 9/14 teams are predicted between 3-5 and 5-3 in conference.
The Pac 12
the numbers like Cal a lot more and Colorado a lot less than the projected win totals.
One of my biggest sells of the season are the Spartans with only 5.5% of simulations seeing the Spartans exceed their 9.5 win projection.
The model is predicting about 9 wins and a virtual tie with Oklahoma, right behind predicted frontrunner Baylor.
After picking Tennessee to breakout in previous years, the model has given up on the Volunteers this season, along with Les Miles’ LSU squad. Two teams projected to overachieve, are league favorite Auburn, which despite a brutal schedule, the model pegs at 10.5 wins, a full 1.5 wins above Vegas along with the rebuilding Kentucky Wildcats.
2015: The Season At Hand, and Other Obvious Subtitles
Buying one, selling the other Pac-12 opponent
Every year in the offseason I test my preseason prediction model, tweaking the coefficients to match the model with the most accurate prediction. Usually it’s just a small move here or there, not really amounting to much. This year I looked at a new variable I called MVP effect. MVP effect looks at the points per play for all QBs and RBs on an offense. Each players’ PPP on their carries+passes is compared versus what the PPP for the team on all other players. This is the gap. The gap is then multiplied by the numbers of plays that the player was responsible for to get their total contribution. I plugged in the MVP stat and took out any stats dealing with returning QBs+RBs and saw a 2.3% reduction in total offense prediction error, and a 2.0% reduction in total error. A pretty big jump for a well-established model. This is part of the reason you’ll see some of these teams as buys or sells.
Return a lot of key pieces on offense and the defense was pretty bad and just hired a new head coach who’s not too bad at coaching that side of the ball.
Selling: Virginia Tech
I know they were young last year, but it’s been a long time since this was a good offensive team and don’t know that the defense can get them over 8 wins.
Buying: Illinois, again
See 2014 notes. For the second straight year they have a 50/50 shot at a bowl game. And they don't have Beckman. Congrats Illini fans?
Selling: Indiana, Wisconsin
See the MVP effect. Tevin Coleman is gone and who is going to generate the Indiana offense? Corey Clement will surely pile up the stats again, but there was a big gap between Melvin Gordon and Corey Clement in PPP and I can’t see the QB play making up the difference. The West is thoroughly mediocre but I have serious doubts about Wisconsin’s ability to stay above the fray.
Selling: West Virginia, kind of TCU
Like Virginia Tech, I don’t see how West Virginia is getting past 8 wins. I think TCU will be good, my season simulations have them as the 3rd most likely team to reach the playoff. The problem is that they are the least talented team in the top ten and despite massive productivity last season, I think Trevone Boykin is due for some serious regression. Definitely a chance I am wrong on TCU but I think they have the most downside risk of any of the major preseason contenders.
Buying: Oregon St
Here’s the MVP effect working in reverse. Sean Mannion was not good for the Beavers last year and now he is gone. A redistribution of the offense should bode well for Oregon St.
Michigan’s other Pac-12 opponent falls on the other side of the ledger. I don’t think Utah was as good as their record last year and another team who is going to fighting uphill in terms of talent for most of their schedule.
Going back to the Malzahn well one more time. Elite talent, elite offensive scheme. 8.5 wins is very doable, even with a tough schedule.
Selling: Arkansas, kind of Texas A&M
Arkansas’s weird season last year has been well documented as several services have vastly overcorrected heading into 2015. Taking the over on 8 wins means you are predicting the Hogs to go at least 5-3 against the SEC West schedule. A&M has a weird setup. Like A&M there is a lot of potential and risk on the roster, I don’t think A&M win total is that far off (they’re the only team on this list where my pick is within 1.5 games of Vegas) but at +200 on the under, sign me up.
Last year I brought in one of my favorite heuristics: your national champion will be on the short list of most talented, experienced rosters. It is now 11 straight years that the national championship has ranked in the top 10 for roster talent (recruiting rankings adjusted for age) and 9 of 11 where the winner has been top 4 on at least one side of the ball. Last season OSU paid at 25/1 by checking in at #4 overall and top 4 on offense.
Here are this year’s top 10 with Top 4 O/D noted.
- Ohio St (O/D)
- Michigan (O/D)
- Alabama (D)
- USC (O)
- Auburn (O)
- Florida St (D)
- Notre Dame
My season simulations have Baylor, TCU, Oregon and Michigan St all with good shots at making the playoff (and odds much higher than several of the teams on this list). It’s not impossible that one of those four wins it all, but it would be the first time in a long time that it’s happened.
Will this one simple trick turn your under achieving team into a contender?
My model loves talented, under-achieving teams (see Texas, Michigan) and it has had some of its biggest misses on teams like this. Michigan will be a big test this season. There are two general ways it can go, depending on what your underlying opinion of the team is.
1. Brady Hoke was a terrible evaluator and developer of talent and the talent Michigan has on paper is a mirage and it’s going to take a couple years to get back on top.
2. Brady Hoke was just a terrible coach and the talent on roster is there, but as yet untapped, especially on offense. It’s less about a Harbaugh turnaround than it is about a loss of Hoke and anything from Harbaugh is gravy this year.
I tend to side with #2. It’s hard to believe that all the talent on the roster were misses. Add to that the defense was pretty good already and you have an opportunity waiting to be exploited and the perfect coach to do so. All preseason predictions tend to take last year’s record as status quo, adjust for the general consensus of returners versus departures and everyone ends up in roughly the spot they started it, adjusting for maybe a game or two in the standing.
If there is ever a case to throw out last year’s record as a starting point it’s this situation. A veteran team, low hanging fruit on turnovers and special teams, a proven defense and an offense that has talent but not production and a coach who has excelled on that side of the ball.
Put me down for 9 wins and a 1-1 record against Michigan State and Ohio State. It’s optimism, it’s the model, it’s the hope that #4 will get decent quarterbacking out of the team and the rest of the team can showcase the talent and experience they have on paper.