Compu-Picks 2012 Analysis: Week Five Games

Mr Pac Ten
Posted Oct 1, 2012


2012 Compu-Picks Analysis: Looking Back at Compu-Picks' Preseason Win Total Picks After Five Weeks of Play

As many of you remember, before the season started I published win total projections as well as picks against the books' lines. This article represents a status update on how those picks are doing.

The table below represents an evaluation of each pick Compu-Picks made against the 5Dimes lines. I've evaluated and placed each pick into one of four buckets: not sure, lean, lock, and clinched. My bucketing is somewhat arbitrary, of course, but I think it's pretty reasonable at this stage. Here is how the system stands so far:

Lean/Lock/Clinched Confidence Level W L
Lean 1 5 5
Lock 1 6 5
Clinched 1 1 0
Lean 2 16 12
Lock 2 8 12
Clinched 2 4 0
Lean 3 21 15
Lock 3 18 6
Clinched 3 0 0
Lean 4 4 4
Lock 4 20 8
Clinched 4 8 0
Lean Total 46 36
Lock Total 52 31
Clinched Total 13 0
All Total 111 67

As you can see, the system has done EXTREMELY well against the preseason win total lines. Unless my evaluations of leans/locks are just totally out of whack (and I really can't see it), a 60% win rate is almost a worst case scenario at this point.

The table below is the full list of each team and an evaulation of where they stand against their preseason win total from 5Dimes (the totals above are sums of this table). It's worth noting that the five teams in the country who have already clinched over/under their win totals so far (Arkansas, Houston, Southern Miss under; Duke, UTSA over) were correct calls by Compu-Picks. In fact, two of those (Duke over 3.5 wins, Houston under 9.5 wins) were picks at the highest confidence level (4).

The table below is structured as follows. The 6 columns are: team; line (what 5Dimes posted back in late August); pick (whether Compu-Picks took the over, the under, or blank if the system passed); confidence level of the preseason pick (1 was the lowest level, 4 the highest), current estimate (win or loss on the pick); and whether the current estimate is a lean, lock, if the outcome has already been clinched.

Team Line Pick Confidence Current Estimate Lean/Lock/Clinched
Boston College 5.5 Over 4 L Lock
Clemson 8.5
Florida State 10 Over 1 ???
Maryland 4.5 Under 2 L Lean
North Carolina State 7.5 Under 3 W Lean
Wake Forest 5.5 Under 1 ???
Duke 3.5 Over 4 W Clinched
Georgia Tech 7.5
Miami (Florida) 6.5 Under 2 L Lock
North Carolina 8 Under 4 W Lean
Virginia 6.5 Under 3 W Lean
Virginia Tech 9 Under 2 W Lock
Baylor 6.5 Under 2 ???
Iowa State 4.5 Under 1 L Lean
Kansas 3.5
Kansas State 7.5 Under 2 L Lock
Oklahoma 9.5 Over 2 L Lean
Oklahoma State 8.5
Texas Christian 8 Over 2 W Lean
Texas 9 Over 2 W Lean
Texas Tech 6.5 Under 1 ???
West Virginia 8.5
Illinois 6.5 Under 1 W Lock
Indiana 3.5 Over 1 ???
Ohio State 9.5 Over 1 W Lock
Penn State 5.5 Under 2 ???
Purdue 6.5
Wisconsin 9.5 Under 3 W Lock
Iowa 7.5 Over 1 L Lock
Michigan 8.5
Michigan State 8.5 Under 2 W Lean
Minnesota 5.5
Nebraska 9 Under 1 ???
Northwestern 6.5 Under 2 L Lock
Notre Dame 7.5 Under 1 L Lean
California 6.5 Under 4 W Lock
Oregon 10.5 Over 1 ???
Oregon State 4.5 Over 3 W Lock
Stanford 7.5 Over 3 W Lean
Washington 7
Washington State 5.5 Under 2 W Lock
Arizona 5.5 Under 3 L Lean
Arizona State 5 Under 3 L Lock
Colorado 3.5 Under 1 W Lock
UCLA 6.5 Over 3 W Lean
Southern California 10.5 Under 3 W Lock
Utah 7.5 Over 3 L Lean
Florida 8 Over 2 W Lean
Georgia 9.5 Under 1 L Lean
Kentucky 5.5 Under 4 W Lock
Missouri 7
South Carolina 8.5
Tennessee 7.5 Under 3 W Lean
Vanderbilt 6.5 Over 2 L Lean
Alabama 10.5
Arkansas 8.5 Under 2 W Clinched
Auburn 7 Over 3 L Lock
Louisiana State 0
Mississippi 5.5 Under 3 ???
Mississippi State 7.5 Over 2 W Lean
Texas A&M 6.5 Under 1 L Lean
Cincinnati 7.5
Connecticut 5.5
Louisville 8.5 Under 1 L Lock
Pittsburgh 7
Rutgers 7.5 Over 2 W Lock
South Florida 8
Syracuse 5.5 Under 3 W Lock
Temple 4.5 Over 3 L Lean
Army 5.5 Over 1 L Lock
Brigham Young 8.5 Over 2 L Lean
Navy 6.5 Over 1 L Lean
Air Force 5.5 Over 4 ???
Boise State 9.5
Colorado State 6.5 Under 3 W Lock
Fresno State 7.5 Under 2 ???
Hawaii 4.5 Over 3 L Lean
Nevada 7.5 Under 4 L Lock
New Mexico 2.5
San Diego State 5.5 Over 2 L Lean
Nevada-Las Vegas 4.5 Under 3 W Lock
Wyoming 5.5 Over 2 L Lock
Central Florida 8.5 Under 1 W Lean
East Carolina 5.5 Over 2 W Lean
Houston 9.5 Under 4 W Clinched
Marshall 6.5 Under 2 ???
Memphis 2.5 Under 1 ???
Rice 3.5 Over 3 L Lean
Southern Methodist 6.5 Under 1 W Lean
Southern Mississippi 8.5 Under 1 W Clinched
Tulane 2.5 Over 4 L Lean
Tulsa 7.5 Over 3 W Lean
Alabama-Birmingham 3.5 Under 1 W Lean
Texas-El Paso 4.5 Under 1 W Lean
UMass 1.5 Over 1 ???
Akron 2.5 Under 2 W Lean
Ball State 5.5 Under 2 L Lock
Bowling Green 7.5
Buffalo 4.5 Under 2 ???
Central Michigan 4.5 Over 4 W Lock
Eastern Michigan 5.5 Under 2 W Lean
Kent 5.5
Miami (Ohio) 5.5 Over 3 W Lean
Northern Illinois 8.5
Ohio 9.5 Under 2 L Lock
Toledo 7.5
Western Michigan 8.5 Under 4 W Lock
Arkansas State 8.5 Under 2 W Lock
Louisiana-Lafayette 7.5
Louisiana-Monroe 4.5 Over 1 W Lock
Middle Tennessee State 4.5 Under 2 L Lean
North Texas 5.5 Under 1 ???
Troy 5.5
Florida Atlantic 3.5
Florida International 7.5
Western Kentucky 6.5 Over 1 W Lock
South Alabama 2.5
Idaho 4.5
Louisiana Tech 8.5 Over 1 W Lock
New Mexico State 5.5 Under 4 W Lock
San Jose State 5.5 Under 1 L Lock
Utah State 7.5 Under 1 L Lock
Texas St 4.5 Under 1 W Lean
UTSA 2.5 Over 2 W Clinched

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

Follow cfn_ms on Twitter