Monday, March 29, 2010

Top 12: AP Poll, Power Points, and Percentages

When all teams have played an equal number of games, combining a team's winning percentage and opponents' winning percentage will produce the same standings as the Power Points System. When all teams have not played an equal number of games, the two systems will produce different standings with the Power Points System (PPS) favoring teams that play more games and the Winning Percentage System (WPS) favoring teams that play fewer games. Listed below are the number of AP top 12 teams that finished the regular season ranked in the top 12 of the PPS or WPS for each season since 1978. The AP rank of the teams that did not finish in the top 12 of either the PPS or WPS are listed after the dash. Also listed below are the number of teams (out of 32) with each AP rank to finish in the top 12 of the PPS or WPS.

1978 11 - 12
1979 11 - 12
1980 10 - 5 11
1981 10 - 11 12
1982 11 - 12
1983 12 -
1984 11 - 12
1985 11 - 12
1986 10 - 6 10
1987 11 - 9
1988 11 - 12
1989 11 - 12
1990 11 - 12
1991 11 - 12
1992 10 - 8 12
1993 11 - 12
1994 11 - 12
1995 11 - 12
1996 10 - 9 11
1997 11 - 11
1998 11 - 11
1999 10 - 4 12
2000 11 - 12
2001 11 - 12
2002 10 - 6 10
2003 10 - 6 12
2004 11 - 7
2005 12 -
2006 11 - 12
2007 11 - 6
2008 12 -
2009 12 -

1 32
2 32
3 32
4 31
5 31
6 28
7 31
8 31
9 30
10 30
11 27
12 13

Round 4: Objectivity vs. Subjectivity

Objectivity (Power Points System) vs. Subjectivity (NCAA Selection Committee)

OVERALL RECORD
Objectivity 45-15. Subjectivity 40-20.

GAMES BETWEEN COMPETING HIGHER SEEDS
Objectivity leads 8-3.

SHARED UPSET LOSSES
Objectivity leads 8-1-3 when it comes to seeding the matchups closer.

DIFFERENT TOP 32 TEAMS
Objectively favored teams finished 7-7 with three teams advancing to the Sweet Sixteen. Subjectively favored teams finished 1-7.

TEAMS WITH SEEDS UPGRADED BY 4+ PLACES
These teams finished 8-8 with three teams advancing to the Sweet Sixteen. Only two of eight losses were to an opponent seeded lower by the Power Points System.

TEAMS WITH SEEDS DOWNGRADED BY 4+ PLACES
These teams finished 1-6.

OBJECTIVITY'S UPSET LOSSES
Only three of the 15 games won by teams seeded lower by the Power Points System involved a seed difference greater than three.

Saturday, March 27, 2010

Top 16: AP Poll, Power Points, and Percentages

When all teams have played an equal number of games, combining a team's winning percentage and opponents' winning percentage will produce the same standings as the Power Points System. When all teams have not played an equal number of games, the two systems will produce different standings with the Power Points System (PPS) favoring teams that play more games and the Winning Percentage System (WPS) favoring teams that play fewer games. Listed below are the number of AP top 16 teams that finished the regular season ranked in the top 16 of the PPS or WPS for each season since 1978. The AP rank of the teams that did not finish in the top 16 of either the PPS or WPS are listed after the dash. Also listed below are the number of teams (out of 32) with each AP rank to finish in the top 16 of the PPS or WPS. For example, all 32 number one teams in the AP finished ranked in either the PPS or WPS.

1978 13 - 12 15 16
1979 12 - 12 13 14 15
1980 14 - 11 16
1981 13 - 12 15 16
1982 13 - 13 14 16
1983 15 - 13
1984 11 - 12 13 14 15 16
1985 15 - 14
1986 14 - 14 16
1987 14 - 14 15
1988 16 -
1989 15 - 16
1990 14 - 15 16
1991 15 - 15
1992 14 - 12 16
1993 15 - 15
1994 13 - 12 15 16
1995 14 - 13 16
1996 15 - 11
1997 13 - 14 15 16
1998 15 - 11
1999 15 - 12
2000 14 - 12 14
2001 13 - 14 15 16
2002 15 - 15
2003 13 - 12 15 16
2004 15 - 16
2005 15 - 15
2006 15 - 15
2007 14 - 15 16
2008 13 - 13 14 15
2009 15 - 16

1 32
2 32
3 32
4 32
5 32
6 32
7 32
8 32
9 32
10 32
11 29
12 23
13 26
14 22
15 15
16 15

Friday, March 26, 2010

Round 3: Objectivity vs. Subjectivity

Objectivity (Power Points System) vs. Subjectivity (NCAA Selection Committee)

OVERALL RECORD
Objectivity 43-13. Subjectivity 38-18.

GAMES BETWEEN COMPETING HIGHER SEEDS
Objectivity leads 7-2.

SHARED UPSET LOSSES
Objectivity leads 8-1-2 when it comes to seeding the matchups closer.

DIFFERENT TOP 32 TEAMS
Objectively favored teams finished 7-7 with three teams advancing to the Sweet Sixteen. Subjectively favored teams finished 1-7.

TEAMS WITH SEEDS UPGRADED BY 4+ PLACES
These teams finished 8-8 with three teams advancing to the Sweet Sixteen. Only two of eight losses were to an opponent seeded lower by the Power Points System.

TEAMS WITH SEEDS DOWNGRADED BY 4+ PLACES
These teams finished 1-6.

OBJECTIVITY'S UPSET LOSSES
Only three of the 13 games won by teams seeded lower by the Power Points System involved a seed difference greater than three.

ELITE EIGHT
Based on the actual bracket, Objectivity advanced five teams to the Eight Eight and Subjectivity advanced four teams.

Sunday, March 21, 2010

Round 2: Objectivity vs. Subjectivity

Objectivity (Power Points System) vs. Subjectivity (NCAA Selection Committee)

OVERALL RECORD
Objectivity 38-10. Subjectivity 32-16.

GAMES BETWEEN COMPETING HIGHER SEEDS
Objectivity leads 7-1.

SHARED UPSET LOSSES
Objectivity leads 6-1-2 when it comes to seeding the matchups closer.

DIFFERENT TOP 32 TEAMS
Objectively favored teams are 7-4 with three teams remaining. Subjectively favored teams finished 1-7.

TEAMS WITH SEEDS UPGRADED BY 4+ PLACES
These teams are 8-5 with three teams remaining. Only one of the five losses was to an opponent seeded lower by the Power Points System.

TEAMS WITH SEEDS DOWNGRADED BY 4+ PLACES
These teams finished 1-6.

OBJECTIVITY'S UPSET LOSSES
Only two of the 10 games won by teams seeded lower by the Power Points System involved a seed difference greater than three.

SWEET SIXTEEN
Objectivity seeded 11 remaining teams to reach the Sweet Sixteen based on actual bracket. Subjectivity seeded 8 remaining teams to reach the Sweet Sixteen.

Friday, March 19, 2010

Round 1: Objectivity vs. Subjectivity

The Power Points System is not a predictive system nor does it make any claims about which teams are better than others. It is simply a rules system that ranks teams according to the value of their season based on the chosen criteria. Simply put, higher ranked teams are not predicted to beat lower ranked teams nor are they said to be better teams. Their seasons are simply more valuable under the prescribed rules.

That said, higher seeded teams in the Power Points System's (PPS) statistical equivalent are 26-6 through the first round of the NCAA tournament. Higher seeds according to the NCAA selection committee (SC) are 22-10. There have been six games where the two systems disagreed on the higher seed and the PPS holds a 5-1 advantage in those games. The two systems share five upset losses and the PPS seeded the game closer in four of the five.

Also, with the NCAA considering expanding its tournament to 96 teams which would presumably include 32 teams having byes, the two systems agreed on 25 of the top 32 teams. Where they disagreed, the PPS Seven finished 4-3 in the first round with all three losses to other PPS top 32 teams and only one loss qualifying as an upset in the PPS. The SC Seven finished 1-6 in the first round with all six losses qualifying as upsets according to the SC's own seeds.

Thursday, March 18, 2010

2010 March Madness

Here is the NCAA field if determined by a statistical equvalent to the Power Points System. This equivalent matched 28 of 34 at-large selections. Actual NCAA seeds are listed second and the six teams in the actual field that missed our field are listed at the bottom.

Kansas 1 1
Kentucky 1 1
Syracuse 1 1
Duke 1 1
New Mexico 2 3
West Virginia 2 2
Butler 2 5
Temple 2 5
Northern Iowa 3 9
Purdue 3 4
Kansas State 3 2
BYU 3 7
Baylor 4 3
Saint Mary's College 4 10
Villanova 4 2
Cornell 4 12
Ohio State 5 2
UTEP 5 12
Pittsburgh 5 3
Tennessee 5 6
Siena 6 13
Murray State 6 13
Gonzaga 6 8
Utah State 6 12
Georgetown 7 3
Richmond 7 7
Vanderbilt 7 4
Wisconsin 7 4
Xavier 8 6
Maryland 8 4
Texas A&M 8 5
Michigan State 8 5
Old Dominion 9 11
San Diego State 9 11
Texas 9 8
UNLV 9 8
California 10 8
UAB 10 NS
Washington 10 11
Florida State 10 9
Rhode Island 11 NS
Oklahoma State 11 7
Memphis 11 NS
Wichita State 11 NS
Virginia Tech 12 NS
Oakland 12 14
Clemson 12 7
Kent State 12 NS
Missouri 13 10
Notre Dame 13 6
Wofford 13 13
Sam Houston State 13 14
New Mexico State 14 12
Morgan State 14 15
North Texas 14 15
Vermont 14 16
Montana 15 14
UC Santa Barbara 15 15
Robert Morris 15 15
Lehigh 15 16
Ohio 16 14
Houston 16 13
East Tennessee State 16 16
Winthrop 16 16
Arkansas-Pine Bluff 16 16

DID NOT QUALIFY
Marquette NS 6
Wake Forest NS 9
Georgia Tech NS 10
Louisville NS 9
Florida NS 10
Minnesota NS 11

Saturday, March 6, 2010

50/50 and 60/40

When all teams have played an equal number of games, the Power Points System will be equal to a 50/50/0 version of college basketball's Ratings Percentage Index. The original version of the RPI is 25% winning percentage, 50% opponents' winning percentage, and 25% opponents' opponents' winning percentage. When teams have not played an equal number of games, competitive advantages/disadvantages result. The Power Points System favors teams that have played more games while the revised RPI favors teams that have played fewer games. I determined the top two teams (minimum 10 FBS games played) for the Power Points System and the revised RPI for the past 32 seasons. The results that came closest to matching the AP poll or BCS Standings are listed below with games removed from a match listed in parentheses. The results for a 60/40/0 version of the Power Points System and revised RPI also follow. For the 60/40 version of the Power Points System, each win and loss is worth an additional 2.75 points in 11 game seasons and an additional 3.00 points in 12 game seasons ('02-03, '06-09).

50/50 RESULTS
1978-Southern California vs Penn State (1)
1979-Florida State vs Brigham Young (4)
1980-Florida State vs Pittsburgh (4)
1981-Clemson vs Pittsburgh (4)
1982-Georgia vs Penn State (0)
1983-Auburn vs Nebraska (1)
1984-Florida vs Brigham Young (2)
1985-Penn State vs Oklahoma (2)
1986-Penn State vs Miami-Florida (0)
1987-Miami-Florida vs Florida State (2)
1988-Notre Dame vs Southern California (1)
1989-Notre Dame vs Colorado (2)
1990-Colorado vs Texas (1)
1991-Miami-Florida vs Washington (0)
1992-Miami-Florida vs Alabama (0)
1993-Florida State vs Nebraska (0)
1994-Nebraska vs Penn State (0)
1995-Florida vs Nebraska (0)
1996-Florida State vs Arizona State (0)
1997-Michigan vs Tennessee (1)
1998-Tennessee vs Florida State (0)
1999-Florida State vs Alabama (1)
2000-Oklahoma vs Florida State (0)
2001-Miami-Florida vs Nebraska (0)
2002-Miami-Florida vs Ohio State (0)
2003-Oklahoma vs Louisiana State (0)
2004-Oklahoma vs Southern California (0)
2005-Southern California vs Texas (0)
2006-Florida vs Michigan (1)
2007-Louisiana State vs Virginia Tech (1)
2008-Oklahoma vs Florida (0)
2009-Alabama vs Texas (0)

60/40 RESULTS
1978-Southern California vs Penn State (1)
1979-Florida State vs Brigham Young (2)
1980-Florida State vs Pittsburgh (1)
1981-Clemson vs Pittsburgh (2)
1982-Georgia vs Penn State (0)
1983-Auburn vs Nebraska (1)
1984-Brigham Young vs Florida (1)
1985-Penn State vs Oklahoma (1)
1986-Penn State vs Miami-Florida (0)
1987-Miami-Florida vs Syracuse (1)
1988-Notre Dame vs Southern California (1)
1989-Colorado vs Notre Dame (1)
1990-Colorado vs Texas (1)
1991-Miami-Florida vs Washington (0)
1992-Miami-Florida vs Alabama (0)
1993-Florida State vs West Virginia (1)
1994-Nebraska vs Penn State (0)
1995-Florida vs Nebraska (0)
1996-Florida State vs Arizona State (0)
1997-Michigan vs Nebraska (0)
1998-Tennessee vs Florida State (0)
1999-Florida State vs Virginia Tech (0)
2000-Oklahoma vs Florida State (0)
2001-Miami-Florida vs Nebraska (0)
2002-Miami-Florida vs Ohio State (0)
2003-Oklahoma vs Louisiana State (0)
2004-Oklahoma vs Southern California (0)
2005-Southern California vs Texas (0)
2006-Ohio State vs Florida (0)
2007-Louisiana State vs Oklahoma (1)
2008-Oklahoma vs Texas (1)
2009-Alabama vs Texas (0)