TeamRankings.com
Get Smart About Sports
September 2, 2015 - by Seth Trachtman
The time of year college football fans have been waiting for has finally arrived. It’s Week 1 of college football season, and there’s a full slate of games beginning with North Carolina vs. South Carolina on Thursday night.
On Sunday night, we rolled out our college football preseason rankings for 2015. We also launched our college football projected standings page, which will update every day between now and the end of the season.
The projected standings bring some very interesting observations. We’ve included some information on our performance and methodology below, or you can jump straight the projections:
ACC | Big Ten | Big 12 | Pac-12 | SEC | All other conferences
A fine question. The honest answer is that we haven’t yet done an extensive analysis of their historical accuracy, or their performance vs. benchmarks like Vegas futures or other sites. We are going to get better at that.
In the meantime, some circumstantial evidence:
As we noted in our 2015 college football preseason rankings blog post, after studying years of historical data, we’ve identified a set of team-level metrics that have predictive value when it comes to projecting how good a team will be in an upcoming season. We built a model that takes in these factors as inputs, and computes a numerical preseason power rating for all 128 FBS teams.
Then, using the 2015 college football schedule and our preseason team power ratings, we simulate the 2015 season game-by-game, thousands of times. Thanks to randomness, each season simulation plays out differently. Occasionally, an unheralded team like Minnesota or Utah makes a run, catches some luck, and wins its conference.
Over thousands of simulation runs, though, trends in the results begin to emerge. The numbers you see below represent the averages of all the season simulation runs we conducted.
It’s important to understand what we mean when we project a team to have, say 7.6 wins this season.
As mentioned above, that 7.6 number is the average number of wins the team had in the thousands of season simulations we ran. So take note:
Human college football “experts” — some of them, at least — can get to be pretty good at projecting the future performance level of a team.
In our experience, though, most sports pundits aren’t good at understanding probability or the potential impacts of randomness, especially when trying to evaluate how the dynamics of a team’s schedule are likely to impact its end of season record.
For example, many people tend to underestimate a great team’s odds of losing to a mediocre or bad team. While it’s true that a team like Ohio State is very unlikely to lose to a team like Penn State or Indiana, upsets do happen, and those probabilities keep adding up game after game. So you can’t discount them, especially when conference championships can be decided by just one win.
Running thousands of season simulations using all these probabilities is a much more accurate and objective way to evaluate the likely impact of a team’s schedule on its expected results.
Below are our end of season standings projections for all five major conferences, based on our preseason predictive power ratings and thousands of computer simulations of the 2015 season.
We update these projections update every single day; below is what they were on September 1, 2015.
The rest of the conferences, along with our most recent projections for all FBS teams, can always be found on our college football projected standings page.
As they do every season, key player injuries, suspensions and other unexpected events that occur as the season goes on are sure to make a good number of these initial forecasts less trustworthy over time.
| Projection | Playoffs | ||||||
|---|---|---|---|---|---|---|---|
| ATLANTIC | conf W | L | overall W | L | Bowl Eligible | Win Conf | Undef |
| Clemson | 5.9 | 2.1 | 9.0 | 3.0 | 88.5% | 20.4% | 7.1% |
| Florida St | 5.6 | 2.4 | 9.0 | 3.0 | 91.7% | 16.0% | 5.9% |
| Louisville | 4.7 | 3.3 | 7.0 | 5.0 | 71.0% | 6.8% | 1.6% |
| NC State | 3.2 | 4.8 | 6.5 | 5.5 | 65.3% | 2.4% | 0.9% |
| Boston Col | 2.9 | 5.1 | 4.9 | 7.1 | 43.9% | 1.7% | 0.3% |
| Syracuse | 2.7 | 5.3 | 4.8 | 7.2 | 41.9% | 0.9% | 0.2% |
| Wake Forest | 1.7 | 6.3 | 3.3 | 8.7 | 22.0% | 0.3% | 0.0% |
| COASTAL | conf W | L | overall W | L | Bowl Eligible | Win Conf | Undef |
| GA Tech | 5.5 | 2.5 | 8.3 | 3.7 | 84.4% | 17.9% | 4.4% |
| VA Tech | 5.2 | 2.8 | 7.9 | 4.1 | 79.1% | 10.9% | 3.0% |
| N Carolina | 4.5 | 3.5 | 7.5 | 4.5 | 76.4% | 5.9% | 2.3% |
| Miami (FL) | 4.1 | 3.9 | 7.1 | 4.9 | 70.4% | 7.7% | 2.1% |
| Pittsburgh | 4.1 | 3.9 | 6.6 | 5.4 | 64.2% | 4.8% | 1.2% |
| Duke | 3.8 | 4.2 | 7.1 | 4.9 | 71.3% | 3.5% | 1.6% |
| Virginia | 2.1 | 5.9 | 3.2 | 8.8 | 22.0% | 0.9% | 0.1% |
Champion Pick: Is this finally the year Clemson overtakes Florida State? Our projections think so, but the numbers are still very close. It’s worth noting that futures odds do favor Florida State by a few percentage points (as of 9/2 leading sportsbooks were offering +236 payout odds for FSU to win the ACC, vs. +298 for Clemson).
Most Improved: After a mediocre 6-6 regular season in 2014, Virginia Tech is projected just shy of eight wins during the regular season — even with a matchup vs. #1 Ohio State in Week 1.
Biggest Decline: It’s tough to match last season’s ACC Championship and College Football Playoff appearance for Florida State. After going undefeated in conference and a perfect 13-0 during the season, we give the Seminoles an overall projection of 9-3 and only a 16 percent chance to win a fourth consecutive ACC title.
Biggest Loser: Virginia (most likely 3-9)
| Projection | Playoffs | ||||||
|---|---|---|---|---|---|---|---|
| Team | conf W | L | overall W | L | Bowl Eligible | Win Conf | Undef |
| TX Christian | 7.3 | 1.7 | 10.1 | 1.9 | 96.9% | 34.4% | 14.5% |
| Baylor | 6.7 | 2.3 | 9.7 | 2.3 | 96.0% | 22.2% | 9.8% |
| Oklahoma | 5.7 | 3.3 | 8.2 | 3.8 | 84.7% | 13.2% | 3.4% |
| W Virginia | 5.0 | 4.0 | 7.7 | 4.3 | 81.3% | 6.3% | 1.7% |
| Oklahoma St | 5.0 | 4.0 | 7.7 | 4.3 | 80.5% | 8.8% | 2.7% |
| Kansas St | 4.6 | 4.4 | 7.3 | 4.7 | 75.3% | 6.9% | 1.7% |
| Texas | 4.5 | 4.5 | 6.4 | 5.6 | 63.2% | 5.2% | 0.7% |
| Texas Tech | 3.4 | 5.6 | 5.2 | 6.8 | 46.8% | 2.4% | 0.3% |
| Iowa State | 2.0 | 7.0 | 3.6 | 8.4 | 23.8% | 0.5% | 0.1% |
| Kansas | 0.7 | 8.3 | 1.6 | 10.4 | 6.3% | 0.1% | 0.0% |
Champion Pick: After last season, the big question is which team will finish on top in the TCU vs. Baylor rivalry. Our projections favor TCU, and it certainly helps that this year’s matchup will be taking place in Fort Worth.
Most Improved: Oklahoma State, 6-6 in the regular season in 2014, is projected to pick up almost another 2 wins each this year, and has the fourth best odds to win the conference.
Biggest Decline: We expect a dropoff of 1.7 wins for Kansas State, after the Wildcats went 9-3 during the regular season in 2014. And can it get any worse for the Kansas Jayhawks? Apparently. They finished last season 3-9, but we project only 1.6 wins and a pitiful 6.3 percent chance of being bowl-eligible. Vegas has set the over/under at 1.5 wins.
Biggest Loser: Kansas (most likely 1-11, or less likely 2-10)
| Projection | Playoffs | ||||||
|---|---|---|---|---|---|---|---|
| EAST | conf W | L | overall W | L | Bowl Eligible | Win Conf | Undef |
| Ohio State | 7.4 | 0.6 | 11.1 | 0.9 | 99.2% | 43.9% | 29.7% |
| Michigan St | 6.0 | 2.0 | 9.5 | 2.5 | 95.1% | 14.1% | 7.2% |
| Michigan | 4.8 | 3.2 | 7.8 | 4.2 | 80.6% | 6.4% | 2.6% |
| Penn State | 4.6 | 3.4 | 7.9 | 4.1 | 82.6% | 3.0% | 1.3% |
| Indiana | 2.7 | 5.3 | 5.3 | 6.7 | 49.2% | 0.6% | 0.3% |
| Maryland | 2.0 | 6.0 | 4.2 | 7.8 | 31.5% | 0.2% | 0.0% |
| Rutgers | 1.4 | 6.6 | 4.0 | 8.0 | 28.5% | 0.2% | 0.0% |
| WEST | conf W | L | overall W | L | Bowl Eligible | Win Conf | Undef |
| Wisconsin | 6.6 | 1.4 | 9.9 | 2.1 | 97.5% | 16.3% | 9.3% |
| Nebraska | 5.5 | 2.5 | 8.6 | 3.4 | 86.7% | 8.6% | 4.7% |
| Iowa | 4.1 | 3.9 | 6.7 | 5.3 | 66.0% | 2.2% | 1.1% |
| Minnesota | 3.8 | 4.2 | 6.5 | 5.5 | 65.1% | 2.8% | 0.5% |
| Northwestern | 3.0 | 5.0 | 4.7 | 7.3 | 40.3% | 0.7% | 0.2% |
| Purdue | 2.4 | 5.6 | 4.1 | 7.9 | 32.5% | 0.7% | 0.1% |
| Illinois | 1.6 | 6.4 | 3.6 | 8.4 | 25.8% | 0.3% | 0.1% |
Champion Pick: The Ohio State University is our #1 ranked team in the preseason, and has the best chance in the nation to go undefeated during the regular season. We give them nearly 30 percent higher odds than Michigan State to win the conference.
Most Improved: How much of a difference will new head coach Jim Harbaugh make for the Wolverines? Our projections see a team on the rise, forecasting a 2.8 win improvement after going 5-7 last season, and a better than 80 percent chance of becoming bowl eligible after failing to make a bowl last year.
Biggest Decline: Illinois is projected to plummet from a 6-6 regular season in 2014 to a four- or three-win regular season this year. Betting odds are pessimistic about the Illini as well; after head coach Tim Beckman was fired last week, leading sportsbooks were offering worse than a 1:1 payout for betting that the Illini will win fewer than 4.5 games.
Biggest Loser: Illinois (4-8 or 3-9)
Projection Playoffs
NORTH conf W L overall W L Bowl Eligible Win Conf Undef
Oregon 7 2 9.3 2.7 93.8% 25.3% 7.8%
Stanford 6.5 2.5 8.9 3.1 87.9% 18.4% 6.5%
California 3.5 5.5 5.4 6.6 49.9% 2.5% 0.4%
Washington 3 6 4.4 7.6 36.2% 1.5% 0.1%
Wash State 2.2 6.8 4.1 7.9 30.8% 0.5% 0.1%
Oregon St 2.1 6.9 3.4 8.6 22.1% 0.3% 0.0%
SOUTH conf W L overall W L Bowl Eligible Win Conf Undef
USC 6.2 2.8 8.6 3.4 87.2% 16.2% 5.1%
UCLA 6 3 8.8 3.2 90.4% 11.6% 5.2%
Arizona St 5.5 3.5 7.8 4.2 80.4% 9.6% 2.9%
Arizona 5.4 3.6 8.1 3.9 84.4% 7.8% 2.7%
Utah 5 4 7.4 4.6 75.7% 5.9% 1.8%
Colorado 1.6 7.4 4.3 8.7 21.8% 0.3% 0.0%
Champion Pick: The Pac-12 is tight at the top, but Oregon projects as the most likely conference champion, although we see their odds of doing so at only around 1-in-4. Both Stanford and USC are nipping at the Ducks’s webbed feet.
Most Improved: Although they’re still expected to finish in the cellar, Colorado’s win total projection (4.3) is more than double their number of 2014 wins. Stanford also looks like a strong contender for this honor, with an expected 1.9 win improvement after a 7-5 regular season where they ended up near the bottom of our 2014-15 luck rankings.
Biggest Decline: After going 8-5 overall during the regular season last year, Washington is projected to go 4-8 or 5-7 this year, with bowl eligibility odds of only about 36 percent.
Biggest Loser: Oregon State (3-9 or 4-8)
| Projection | Playoffs | ||||||
|---|---|---|---|---|---|---|---|
| EAST | conf W | L | overall W | L | Bowl Eligible | Win Conf | Undef |
| Georgia | 5.4 | 2.6 | 8.7 | 3.3 | 89.8% | 15.0% | 5.3% |
| Missouri | 4.7 | 3.3 | 8.1 | 3.9 | 82.8% | 7.1% | 3.5% |
| Tennessee | 4.2 | 3.8 | 7.6 | 4.4 | 78.4% | 6.7% | 1.9% |
| Florida | 3.7 | 4.3 | 7.0 | 5.0 | 72.7% | 4.0% | 1.5% |
| S Carolina | 3.2 | 4.8 | 5.6 | 6.4 | 53.8% | 2.3% | 0.6% |
| Kentucky | 2.3 | 5.7 | 5.2 | 6.8 | 48.1% | 1.0% | 0.4% |
| Vanderbilt | 1.9 | 6.1 | 4.4 | 7.6 | 35.8% | 0.8% | 0.2% |
| WEST | conf W | L | overall W | L | Bowl Eligible | Win Conf | Undef |
| Alabama | 5.8 | 2.2 | 9.6 | 2.4 | 94.4% | 23.5% | 10.5% |
| Auburn | 4.4 | 3.6 | 8.1 | 3.9 | 84.5% | 8.6% | 3.5% |
| LSU | 4.4 | 3.6 | 8.2 | 3.8 | 85.1% | 7.3% | 3.4% |
| Texas A&M | 4.3 | 3.7 | 7.7 | 4.3 | 79.8% | 6.9% | 2.8% |
| Mississippi | 4.1 | 3.9 | 7.8 | 4.2 | 82.7% | 5.9% | 2.2% |
| Arkansas | 4.0 | 4.0 | 7.7 | 4.3 | 80.1% | 6.1% | 2.3% |
| Miss State | 3.5 | 4.5 | 7.2 | 4.8 | 76.8% | 4.7% | 1.7% |
Projected Champion: Can Alabama repeat in the SEC? The conference is stacked with contenders, but we give the Crimson Tide the best chance to win it in 2015. Although they only have about 24% odds to pull it off, Bama still has a decently better shot at the conference title than either in-state rival Auburn (roughly 9%) or projected SEC East champion Georgia (15%).
Most Improved: While still projected near the bottom in a loaded SEC West division, we expect a two-conference-win improvement from Arkansas over last year (4 vs. 2). In the SEC East, Tennessee is projected at 7.6 wins after going 6-6 during the regular season last year.
Biggest Decline: After Missouri went 10-2 during the regular season in 2014, our projections see a nearly two win regression for the two-time defending SEC East champions. Mizzou is still expected to finish 2nd in the East, but Georgia has significantly better odds to win the conference (15% vs. 7% for Missouri).
Biggest Loser: Vanderbilt (4-8 or 5-7)
Enjoy the 2015 college football season, and remember, plenty of these projections will end up being wrong, for lots of possible reasons. Our goal is the overall accuracy of the system, and even if system as a whole ends up being highly accurate this year, the projections for any one specific team could still be WAY off.
The value we bring to the table is a sophisticated, systematic season projection model that can precisely evaluate dynamics that human experts are generally terrible at understanding — a system that is not driven by subjective opinions, but by objective data that we can demonstrate has predictive value.
Printed from TeamRankings.com - © 2005-2024 Team Rankings, LLC. All Rights Reserved.