One new piece of analysis I'm doing this year is looking at some of key matchups and what Compu-Picks thinks the results of these matchups will say about the rest of the season. Before week one, I took a look at Alabama-Michigan, Auburn-Clemson, Boise St - Michigan St, Georgia Tech - Virginia Tech, and Texas A&M - Louisiana Tech, and after the games were over took a similar look at last week's upsets.
Next up is a similar look at some key week two games: Georgia @ Missouri, Florida @ Texas A&M, Nebraska @ UCLA, Wisconsin @ Oregon St, and an under the radar game, South Florida @ Nevada. Since we already know what happened in week one, I'll be showing four columns for everyone who's already played: the first three are for a win, a loss, and any result given what already happened in week one, and the last column is the original projection without taking into account the week one results. For Texas A&M, there's just the three columns, since their week one game at Louisiana Tech was postponed.
Let's get the ball rolling with a key early SEC matchup, Georgia @ Missouri. Both teams started 1-0, so no shocking early upsets come in and mess with the projections. This game is projected as about a tossup, and in an SEC East division that has a lot of tough games, this matchup is projected to be worth about a full win the rest of the way for both teams. Whoever wins will have a solid shot at making some SEC noise, especially if it's Georgia (who has the easier league draw and is power-rated higher).
| Georgia 2012 Season Projections based on W/L vs Missouri given 1-0 start |
| Missouri Result | W Given 1-0 | L Given 1-0 | Any Given 1-0 | Any |
| Buffalo - already played | 100% | 100% | 100% | 99% |
| Win % vs |
| Missouri | 100% | 0% | 55% | 55% |
| Fla. Atlantic | 100% | 99% | 99% | 99% |
| Vanderbilt | 81% | 65% | 74% | 74% |
| Tennessee | 90% | 80% | 86% | 85% |
| South Carolina | 53% | 37% | 46% | 46% |
| Kentucky | 93% | 85% | 89% | 89% |
| Florida | 50% | 33% | 43% | 42% |
| Mississippi | 96% | 89% | 93% | 93% |
| Auburn | 44% | 28% | 36% | 36% |
| Georgia Tech | 86% | 75% | 81% | 81% |
| AA | 100% | 100% | 100% | 100% |
| Total E(Wins) - except for Missouri | 8.94 | 7.91 | 8.48 | 8.45 |
| Total E(Wins) - including Missouri | 9.94 | 7.91 | 9.03 | 9.00 |
| Missouri 2012 Season Projections based on W/L vs Georgia given 1-0 start |
| Georgia Result | W Given 1-0 | L Given 1-0 | Any Given 1-0 | Any |
| AA - already played | 100% | 100% | 100% | 100% |
| Win % vs |
| Georgia | 100% | 0% | 45% | 45% |
| Arizona St. | 97% | 94% | 95% | 95% |
| South Carolina | 38% | 24% | 31% | 31% |
| UCF | 76% | 63% | 69% | 69% |
| Vanderbilt | 68% | 53% | 60% | 60% |
| Alabama | 26% | 15% | 20% | 20% |
| Kentucky | 94% | 88% | 91% | 91% |
| Florida | 30% | 18% | 23% | 23% |
| Tennessee | 68% | 52% | 59% | 59% |
| Syracuse | 96% | 89% | 92% | 92% |
| Texas A&M | 64% | 50% | 56% | 56% |
| Total E(Wins) - except for Georgia | 7.58 | 6.45 | 6.96 | 6.95 |
| Total E(Wins) - including Georgia | 8.58 | 6.45 | 7.41 | 7.40 |
Next up is Florida @ Texas A&M. Here again, the difference between a win and a loss for both teams is worth about a full win, a bit more in the case of Florida (since they have more tossup-type games than A&M). Florida was projected preseason as a solid favorite, and since we're just looking at wins and losses, the fact that they played very poorly against Bowling Green doesn't factor into the numbers. That said, if they don't play much better, we're heading for an upset that will shake up both of these teams' projections.
| Florida 2012 Season Projections based on W/L vs Texas A&M given 1-0 start |
| Texas A&M Result | W Given 1-0 | L Given 1-0 | Any Given 1-0 | Any |
| Bowling Green - already played | 100% | 100% | 100% | 97% |
| Win % vs |
| Texas A&M | 100% | 0% | 78% | 77% |
| Tennessee | 83% | 67% | 79% | 78% |
| Kentucky | 98% | 92% | 96% | 96% |
| LSU | 66% | 47% | 62% | 61% |
| Vanderbilt | 74% | 56% | 70% | 69% |
| South Carolina | 77% | 58% | 73% | 72% |
| Georgia | 63% | 44% | 58% | 58% |
| Missouri | 81% | 67% | 78% | 77% |
| La.-Lafayette | 98% | 93% | 97% | 96% |
| Florida St. | 42% | 28% | 39% | 38% |
| AA | 100% | 100% | 100% | 100% |
| Total E(Wins) - except for Texas A&M | 8.81 | 7.52 | 8.52 | 8.42 |
| Total E(Wins) - including Texas A&M | 9.81 | 7.52 | 9.30 | 9.19 |
| Texas A&M 2012 Season Projections based on W/L vs Florida |
| Florida Result | W | L | Any |
| Win % vs |
| Florida | 100% | 0% | 23% |
| SMU | 81% | 68% | 71% |
| Arkansas | 65% | 43% | 48% |
| Mississippi | 77% | 62% | 66% |
| LSU | 41% | 24% | 28% |
| Auburn | 28% | 14% | 17% |
| Mississippi St. | 33% | 18% | 22% |
| Alabama | 11% | 5% | 6% |
| Missouri | 60% | 39% | 44% |
| AA | 100% | 99% | 99% |
| AA | 100% | 99% | 99% |
| Louisiana Tech | 71% | 52% | 56% |
| Total E(Wins) - except for Florida | 6.68 | 5.22 | 5.56 |
| Total E(Wins) - including Florida | 7.68 | 5.22 | 5.79 |
Next up is another game that was rated as a tossup preseason, Nebraska going out west to visit UCLA. Nebraska played better than UCLA in week one, but their star running back may be out for the game. Given all that, a tossup projection still seems pretty reasonable, and similar to most other cases, the difference between a win and a loss is worth about another full win the rest of the way.
| Nebraska 2012 Season Projections based on W/L vs UCLA given 1-0 start |
| UCLA Result | W Given 1-0 | L Given 1-0 | Any Given 1-0 | Any |
| Southern Miss - already played | 100% | 100% | 100% | 86% |
| Win % vs |
| UCLA | 100% | 0% | 52% | 49% |
| Arkansas St. | 95% | 91% | 93% | 91% |
| Wisconsin | 85% | 73% | 79% | 76% |
| Ohio St. | 31% | 20% | 26% | 24% |
| Northwestern | 85% | 75% | 80% | 78% |
| Michigan | 59% | 45% | 52% | 49% |
| Michigan St. | 61% | 46% | 54% | 51% |
| Penn St. | 96% | 92% | 94% | 92% |
| Minnesota | 96% | 91% | 94% | 92% |
| Iowa | 65% | 48% | 57% | 53% |
| AA | 100% | 100% | 100% | 100% |
| Total E(Wins) - except for UCLA | 8.74 | 7.80 | 8.28 | 7.94 |
| Total E(Wins) - including UCLA | 9.74 | 7.80 | 8.80 | 8.42 |
| UCLA 2012 Season Projections based on W/L vs Nebraska given 1-0 start |
| Nebraska Result | W Given 1-0 | L Given 1-0 | Any Given 1-0 | Any |
| Rice - already played | 100% | 100% | 100% | 85% |
| Win % vs |
| Nebraska | 100% | 0% | 55% | 51% |
| Houston | 94% | 88% | 91% | 89% |
| Oregon St. | 79% | 66% | 73% | 70% |
| Colorado | 95% | 91% | 93% | 92% |
| California | 80% | 67% | 74% | 71% |
| Utah | 52% | 39% | 46% | 43% |
| Arizona St. | 92% | 84% | 88% | 86% |
| Arizona | 96% | 91% | 94% | 92% |
| Washington St. | 85% | 76% | 81% | 78% |
| USC | 47% | 34% | 41% | 38% |
| Stanford | 46% | 33% | 40% | 37% |
| Total E(Wins) - except for Nebraska | 8.66 | 7.68 | 8.22 | 7.81 |
| Total E(Wins) - including Nebraska | 9.66 | 7.68 | 8.76 | 8.33 |
Wisconsin at Oregon St is rated just about the same as Nebraska at UCLA, which is interesting given that the Badgers are over a touchdown favorites. Similar to the above examples, a win versus a loss is worth about another full win the rest of the way for both teams, although in this case it's a bit more than a full win for each.
| Wisconsin 2012 Season Projections based on W/L vs Oregon St. given 1-0 start |
| Oregon St. Result | W Given 1-0 | L Given 1-0 | Any Given 1-0 | Any |
| AA - already played | 100% | 100% | 100% | 99% |
| Win % vs |
| Oregon St. | 100% | 0% | 46% | 45% |
| Utah St. | 91% | 85% | 88% | 87% |
| UTEP | 96% | 90% | 93% | 92% |
| Nebraska | 31% | 18% | 24% | 24% |
| Illinois | 89% | 78% | 83% | 82% |
| Purdue | 63% | 47% | 54% | 54% |
| Minnesota | 87% | 78% | 82% | 82% |
| Michigan St. | 59% | 43% | 50% | 50% |
| Indiana | 87% | 76% | 81% | 81% |
| Ohio St. | 34% | 19% | 26% | 26% |
| Penn St. | 79% | 66% | 72% | 72% |
| Total E(Wins) - except for Oregon St. | 8.16 | 7.00 | 7.53 | 7.48 |
| Total E(Wins) - including Oregon St. | 9.16 | 7.00 | 7.99 | 7.93 |
| Oregon St. 2012 Season Projections based on W/L vs Wisconsin given 1-0 start |
| Wisconsin Result | W Given 1-0 | L Given 1-0 | Any Given 1-0 | Any |
| AA - already played | 100% | 100% | 100% | 98% |
| Win % vs |
| Wisconsin | 100% | 0% | 55% | 55% |
| UCLA | 38% | 21% | 30% | 30% |
| Arizona | 79% | 66% | 73% | 73% |
| Washington St. | 87% | 79% | 83% | 83% |
| BYU | 22% | 11% | 17% | 17% |
| Utah | 44% | 29% | 37% | 37% |
| Washington | 40% | 23% | 32% | 32% |
| Arizona St. | 92% | 85% | 89% | 88% |
| Stanford | 20% | 12% | 16% | 16% |
| California | 81% | 70% | 76% | 75% |
| Oregon | 10% | 4% | 7% | 7% |
| Total E(Wins) - except for Wisconsin | 6.12 | 4.99 | 5.62 | 5.56 |
| Total E(Wins) - including Wisconsin | 7.12 | 4.99 | 6.17 | 6.10 |
Miami at Kansas St is a game rated similarly to Florida at Texas A&M, where one team was projected as a solid favorite. It's interesting that this didn't dramatically change even after MIami scored a key win at BC in a game that Compu-Picks also called them solid dogs for. If they can win this one, though, they'll have gone from an original 5 win projection to over an 8.5 win projection, a major movement in just two weeks. Even if they can't pull it off, simply splitting their road trip puts them over a full win in front of where they were originally projected.
For Kansas St, on the other hand, a loss here would be a definite blow to their postseason hopes. They play in a very tough league, and if they can't sweep a pretty light non-conference slate, it'll be an uphill slog to hit six, much less seven or more wins.
| Miami (Fla.) 2012 Season Projections based on W/L vs Kansas St. given 1-0 start |
| Kansas St. Result | W Given 1-0 | L Given 1-0 | Any Given 1-0 | Any |
| Boston College - already played | 100% | 100% | 100% | 24% |
| Win % vs |
| Kansas St. | 100% | 0% | 35% | 23% |
| Georgia Tech | 49% | 29% | 36% | 24% |
| N. Carolina St. | 85% | 76% | 79% | 68% |
| Notre Dame | 43% | 23% | 30% | 19% |
| North Carolina | 91% | 83% | 86% | 78% |
| Florida St. | 24% | 13% | 16% | 9% |
| Virginia Tech | 60% | 42% | 48% | 34% |
| Virginia | 83% | 65% | 72% | 60% |
| South Florida | 63% | 49% | 54% | 41% |
| Duke | 69% | 53% | 59% | 45% |
| AA | 100% | 99% | 99% | 97% |
| Total E(Wins) - except for Kansas St. | 7.67 | 6.33 | 6.80 | 5.00 |
| Total E(Wins) - including Kansas St. | 8.67 | 6.33 | 7.14 | 5.23 |
| Kansas St. 2012 Season Projections based on W/L vs Miami (Fla.) given 1-0 start |
| Miami (Fla.) Result | W Given 1-0 | L Given 1-0 | Any Given 1-0 | Any |
| AA - already played | 100% | 100% | 100% | 99% |
| Win % vs |
| Miami (Fla.) | 100% | 0% | 77% | 77% |
| North Texas | 95% | 88% | 93% | 93% |
| Oklahoma | 6% | 3% | 6% | 5% |
| Kansas | 88% | 77% | 85% | 85% |
| Iowa St. | 73% | 58% | 70% | 70% |
| West Virginia | 33% | 17% | 29% | 29% |
| Texas Tech | 72% | 56% | 68% | 68% |
| Oklahoma St. | 46% | 31% | 42% | 42% |
| TCU | 19% | 10% | 17% | 17% |
| Baylor | 60% | 40% | 55% | 55% |
| Texas | 9% | 3% | 8% | 8% |
| Total E(Wins) - except for Miami (Fla.) | 6.01 | 4.82 | 5.74 | 5.71 |
| Total E(Wins) - including Miami (Fla.) | 7.01 | 4.82 | 6.51 | 6.48 |
There are a few important notes and caveats I need to make about this model:
1) Compu-Picks does not endorse implicitly or explicitly any form of illegal gambling. Compu-Picks is intended to be used for entertainment purposes only.
2) No guarantee or warranty is offered or implied by Compu-Picks for any information provided and/or predictions made.
3) This preseason model is primarily based on the main compu-picks model. Essentially, it attempts to predict how well a team will rate given its rating history, as well as a number of other data points, such as returning starters, draft talent lost, turnovers, recruiting, etc. This means, among other things, that the rankings are power rankings based on how good a team projects to be, as opposed to a more cynical (though accurate) model that attempts to project how the BCS will rank a team by making adjustments to favor those with easy schedules and punish those with tough schedules.
2012 Compu-Picks Blog
Questions, comments or suggestions? Email me at cfn_ms@hotmail.com
