EXPLANATION OF DATA AND
PRESENTATION FORMATS

As of the end of play on March 6, 2005, all Ninety-six SEC games are complete, and I have predicted the winner correctly in 64[67%] of these games. 

The thirty-two games I have missed are:

1.  Alabama @ Arkansas, 50111. 
Alabama +1 W's and -1 L's
Arkansas -1 W's and +1 L's

2.  Mississippi State @ Tennessee, 50112
Tennessee +1 W's and -1 L's
Mississippi State-1 W's and +1 L's

3.  USC @ LSU, 50112
LSU +1 W's and -1 L's
USC -1 W's and +1 L's

4.  Mississippi @ Auburn, 50115
Mississippi +1 W's and -1 L's
Auburn -1 W's and +1L's

5.  Arkansas @ Mississippi State, 50115
Mississippi State +1 W's and -1 L's
Arkansas -1 W's and +1 L's

6.  Arkansas @ LSU, 50119
LSU +1 W's and -1 L's
Arkansas -1W's and +1 L's

7. Tennessee @ Florida, 50119
Tennessee +1 W's and -1 L's
Florida -1 W's and +1 L's

8.  Vanderbilt @ Georgia, 50122
Georgia +1 W's and -1 L's
Vanderbilt, -1 W's and +1 L's

9.  USC @ Mississippi State, 50122
Mississippi State, +1W's and -1 L's
USC, -1 W's and +1 L's

10. Kentucky @ Arkansas, 50129
Kentucky, +1 W's and -1 L's
Arkansas, -1 W's and +1 L's

11. Mississippi State @ LSU, 50129
LSU, +1 W's and -1 L's
Mississippi State, -1 W's and +1 L's

12. Florida @ Mississippi State, 50201
Mississippi State, +1 W's and -1 L's
Florida, -1 W's and +1 L's

13.  Auburn @ Mississippi, 50202
Mississippi, +1 W's and -1 L's
Auburn, -1 W's and +1 L's

14.  Mississippi State @ Auburn, 50205
Auburn, +1 W's and -1 L's
Mississippi State, -1 W's and +1 L's

15.  Auburn @ LSU, 50209
LSU,+1 W's and -1 L's
Auburn, -1 W's and +1 L's

16.  Vanderbilt @ Mississippi State
Mississippi State,+1 W's and -1 L's
Vanderbilt, -1 W's and +1 L's

17.  Auburn @ USC
Auburn,+1 W's and -1 L's
USC, -1 W's and +1 L's

18.  LSU @ Mississippi State
LSU, +1 W's and -1 L's
Mississippi State, -1 W's and +1 L's

19.  Arkansas @ Alabama
Alabama,+1 W's and -1 L's
Arkansas, -1 W's and +1 L's

20.  Florida @ LSU
LSU, +1 W's and -1 L's
Florida , -1 W's and +1 L's

21. USC @ Alabama
Alabama,+1 W's and -1 L's
USC, -1 W's and +1 L's

22.  Auburn @ Georgia,
Georgia, +1 W's and -1 L's
Auburn, -1 W's and +1 L's

23.  Arkansas @ Vanderbilt
Vanderbilt,+1 W's and -1 L's
Arkansas, -1 W's and +1 L's

24.  Alabama @ LSU,
LSU,+1 W's and -1 L's
Alabama, -1 W's and +1 L's

25.  USC @ Tennessee
Tennessee +1 W's and -1 L's
USC, -1 W's and +1 L's

26.  LSU @ Auburn
LSU,+1 W's and -1 L's
Auburn, -1 W's and +1 L's

27.  Florida @ USC
Flordia, +1 W's and -1 L's
USC, -1 W's and +1 L's

28.  Mississippi State @ Arkansas, 50129
Mississippi State, +1 W's and -1 L's
Arkansas, -1 W's and +1 L's

29.  LSU @ Mississippi
LSU,+1 W's and -1 L's
Mississippi, -1 W's and +1 L's

30.  USC @ Vanderbilt
Vanderbilt +1 W's and -1 L's
USC, -1 W's and +1 L's

31  Arkansas @ Auburn
Auburn, +1 W's and -1 L's
Arkansas -1 W's and +1 L's

32.  Vanderbilt @ LSU,
LSU,+1 W's and -1 L's
Vanderbilt, -1 W's and +1 L's

These incorrect predictions change the predicted final standings to:

Current
East Division

 

 

 

Florida  -2 4

12

4

USC  -7 7

7

9

Kentucky  +1 1

14

2

Vanderbilt -1 5

8

8

Tennessee +3 3

6

10

Georgia  +2 2

2

14

 

Total for Division -3 11

 

 

49

 

47

 

Modified Predicted
East Division

 

 

 

Florida  -2

12

4

USC  -7

7

9

Kentucky  +1

14

2

Vanderbilt -1

8

8

Tennessee +3

6

10

Georgia  +2

2

14

 

Total for Division

 

 

49

 

47

 

Current
West Division

 

 

 

Arkansas -8 8

6

10

Alabama +2 4

12

4

Miss. St .  +1 9

9

7

Auburn  -3 7

4

12

Mississippi  +1 3

4

12

LSU  +10 10

12

4

 

Total for Division  +3 23

 

47

 

49

 

 

Modified Predicted
West Division

 

 

 

Arkansas -8

6

10

Alabama +2

12

4

Miss. St . +1

9

7

Auburn  -2

4

12

Mississippi  +1

4

12

LSU  +10

12

4

 

Total for Division

 

47

 

49

 

As the tables above show, the team on the significantly plus side of the ledger, relative to my pre-season predictions is LSU [+10].  The big loser thus far is Arkansas [-8] and USC [-7].  The other nine teams are within one to three games of my pre-season predictions:  Within one game [Four Teams] of the predicted pace, within two gamesd [Four Teams] of my predicted pace, or within three games [One Team].

 

The original article with the predictions follows:

2005 SEC Will Be
Very Competitive and Exciting

by
Richard "The Professor" Cheeks

Since Adolph Rupp began coaching at Kentucky , Kentucky has dominated SEC basketball, winning or sharing the SEC regular season championship 42 times , and winning the SEC post-season tournament 10 of the last 13 seasons.   Other members of the conference have risen to challenge Kentucky , but after a few seasons in the hunt, fall back into the pack.  

In the 1950s, Mississippi State rose to prominence under the leadership of Babe McCarthy.   Tennessee provided the principal challenge in the late 1960s and early 1970s during Coach Ray Mears' tenure.   Coach Brown led LSU to this position in the late 1970s and 1980s.   Arkansas followed Nolan Richardson to the glory land of college hoops in the 1990s.   In recent years, Coach Donovan's Gators have emerged to threaten Kentucky 's dominance.   However, through each of these challenges, Kentucky has remained the standard by which all others compare their status and progress.  

The 2004-05 edition of Wildcats will again compete for top conference honors, but they will not have a cakewalk to a regular season or tournament championship this year.   Six of the twelve SEC teams are legitimate NCAA tournament material, and this balance will produce a highly competitive and exciting 2005 SEC season.  

SEC East:

The Eastern Division of the SEC will be very competitive this season, with four of the six teams legitimately challenging for top honors.  

Coach Odom has assembled a strong team at South Carolina , continuing the fine coaching tradition he established while leading the Wake Forest program.   During the pre-season, USC has earned an impressive 8 3 record, with road loses to Kansas (4 pts) and Pittsburgh (4 pts), and a home overtime loss to in-state rival Clemson (1 pt).   USC is 9 points from a perfect record through the strongest pre-conference schedule of all SEC teams this season.   USC will compete for top honors in the East, and they will at least share those honors with either Kentucky or Florida .

Unlike USC, Florida has played the second weakest pre-conference schedule in posting an identical 8 3 record.   However, Florida lost twice at home ( Louisville and Miami of Florida) and an embarrassing 13-point road loss to a weak, but rival Florida State .   Florida did defeat a good Providence College team in November on a neutral court by 18 points.   Florida will compete for top honors, especially if Coach Donovan can mold his team into a cohesive unit, absent the strife that has marked recent Florida squads.

Kentucky has traditionally played one of the toughest pre-conference schedules in the country and in the SEC.   However, this season, Coach Smith altered his scheduling philosophy and added several weak opponents to the schedule to provide an opportunity for his young Cats to develop for the conference wars.   In the past, the tougher pre-conference schedule has benefited Kentucky teams once SEC play begins.   However, Kentucky 's pre-season schedule is not void top-flight opponents.   Kentucky has defeated a strong Louisville team in Freedom Hall , Indiana on a neutral court, and lost to North Carolina in Chapel Hill .   Kansas remains on the schedule at Rupp Arena on Sunday, January 9.    However, games against Coppin State (143), Ball State (131), Georgia State (152), Tennessee Tech (252), Morehead State (183), William & Mary (221), and Campbell (261) allowed Kentucky to post impressive statistics while building a 9 1 pre-season record.    Despite the impressive record, the schedule weakness must moderate any forecast for the remainder of Kentucky 's season.   Nonetheless, Kentucky will compete for the SEC East with USC and Florida .

Vanderbilt has played the second toughest pre-conference schedule on its way to a 9 4 record.   Vanderbilt will be a very tough test for the conference contenders in Nashville this season, but at the end of the day, Vanderbilt will secure only a fourth place finish in the rugged SEC East.   If Vanderbilt could play in the West, it would fare much better.

Georgia and Tennessee bring up the rear of the SEC East this season.  

SEC West:

In the West, Arkansas and Alabama will compete head to head with the big four of the East.   Auburn and Mississippi State have solid teams, which have posted better than average non-conference records and performance while Mississippi and LSU bring up the bottom.   This forecast breaks with the conventional wisdom which says that Mississippi State , and not Arkansas will compete at the top.   Arkansas has posted an impressive 12 1 record against an admittedly weak non-conference schedule, and their only loss came to Illinois , a legitimate national champion contender this season.   Mississippi State has a similarly impressive 12 2 record, with loses to highly regarded Syracuse and Arizona .

The differences between these two teams are revealed in the efficiencies of their respective offenses and defenses against their non-conference opponents.   In posting its 12 1 record, Arkansas has an average offensive efficiency of 0.96 points per possession (ppp) while limiting their opponents to less than 0.7 ppp.   In contrast, the corresponding Mississippi State efficiencies are 0.87 and 0.77 ppp for offense and defense, respectively.   This difference will be significant once SEC play begins.   Arkansas ' 0.26 point per possession advantage over its opponents is among the best in the SEC this year, while Mississippi State 's 0.11 ppp advantage is slightly below average.

For these reasons, Arkansas will win the SEC West this season.   The number two team in the SEC West will be Alabama , followed closely by Mississippi State .   Auburn will finish fourth in the West, and Mississippi and LSU bring up the bottom.

Methodology:

The SEC season consists of 96 basketball games, eight on each of the 12 home floors.   The non-conference offensive and defensive efficiencies, normalized for variable schedule strengths provide an indication of the outcome of each of these games.   This analysis estimates the game score, the margin, and identifies the predicted winner.   While objective, this analysis cannot account for normal game-to-game variations in the level of play for a team.   It is this game-to-game variation that produces upsets every year in every conference.  

For example, over the last 10 years, Kentucky 's offensive efficiency has averaged about 0.9 ppp.   By definition, about half of the Kentucky games have offensive efficiencies below the mean and about half above the mean.   Some of that variation is due to variable opponent strength.   However, part of that variation is also due to variable performance.   The Standard Deviation of the typical season data has been about 0.11 ppp, which means that offensive efficiency falls in a range of 0.79 ppp to 1.01 ppp for about 2/3 of all games.   Similar statistical properties result for defensive efficiency and Net Game Efficiency.

The Prediction:

This analysis results in the following predicted conference standings.   The following table provides the predicted conference wins and loses for each team.   As this analysis indicates, the East is stronger than the West this season, and Eastern Division teams will win 23 of the 36 interdivisional games.   The analysis also indicates that 3 points or less will decide 44 of the 96 games.   In contrast, 8 or more points will decide only 23 games.  

To see how well the predictions compare with actual outcomes, click here.

East Division

 

 

 

Florida

14

2

USC

14

2

Kentucky

13

3

Vanderbilt

9

7

Tennessee

3

13

Georgia

0

16

 

Total for Division

 

 

53

 

43

 

 

West Division

 

 

 

Arkansas

14

2

Alabama

10

6

Miss. St .

8

8

Auburn

6

10

Mississippi

3

13

LSU

2

14

 

Total for Division

 

43

 

53

 

Will upsets occur?   You betcha!   However, over the course of the full conference season, a team like Kentucky will win one or two games it is now picked to lose, and vice versa.   This is true for all teams.

The upcoming SEC season will be competitive and exciting.   The SEC should send at least five teams (USC, Florida , Kentucky , Arkansas , and Alabama ) to the big dance in March, and could get a sixth team (Vanderbilt).

FOOTNOTES:

http://www.ukfans.net/jps/uk/Statistics/TeamSECChampions.html visited January 4, 2005 .

http://www.ukfans.net/jps/uk/Statistics/TeamSECTournamentChampions.html visited January 4, 2005 .

Kentucky must play Kansas on January 9, 2005 to round out its non-conference play.   Kansas is a legitimate contender for the 2005 national championship, and while this game will be on the Rupp floor, Kentucky will be hard pressed to defeat the Jayhawks, and the final Kentucky non-conference record will probably be 9 2.

The numbers in parentheses are the RPI rankings on January 4, 2005 .   I did not include Indiana in this litany of weak opponents even thought IU's current RPI rating is 134 because Indiana is a traditional Kentucky opponent, and their program is in an uncharacteristic down period.   However, the scheduling of Indiana over the years has earmarked the toughness of Kentucky 's non-conference scheduling.

Vanderbilt has played four teams with current RPI ratings of 50 or better, and they lost all four games.   Vanderbilt's non-conference schedule is the second strongest in the SEC this season.

Efficiency as expressed herein measures the relative skill of teams.   When a team has higher offensive efficiency than defensive efficiency, it will win more games than it loses.   The greater the spread, the higher the eventual winning percentage.   The difference between the measure, Net Game Efficiency, is directly proportional to winning percentage.   Arkansas has a Net Game Efficiency for its non-conference games of 0.26 ppp whereas Mississippi States corresponding Net Game Efficiency is about 0.11 ppp.   Both played similarly strong (weak) non-conference schedules.   When conference play begins, the level of competition will be higher.   The average Net Game Efficiency among SEC schools for non-conference play is about 0.13.   Therefore, Mississippi States performance is slightly below average while Arkansas ' performance has been the second strongest of all SEC teams this year.

As described in FN6, Net Game Efficiency is the Offensive Efficiency minus the Defensive Efficiency.   Winning teams will have positive Net Game Efficiencies and losing teams will have negative values.

 

Click Here To Return to the Data

 

Copyright 2005 Richard Cheeks
All Rights Reserved