Tuesday, February 14, 2006

Homefield Advantage in the SEC

A few weeks ago I took a look at the cummulative Big 10 conference record of every team since 2000 to try and determine who had the biggest homefield advantage. So here’s a follow up with the same thing for the SEC teams. Again, I chose only conference records because each team has different non-conference schedules (both in quality and quantity). The sample size of 6 seasons gives us 48 total games (24 home and 24 road) for each team and helps eliminate some of the volatility of a conference season that does not include every opponent. It should be noted that Florida and Georgia screw up the data by playing one neutral site game against each other, so they will have only 42 games apiece. First up, the home record of each SEC team since 2000 from best to worst.

Auburn 20-4
Georgia 17-4
LSU 19-5
Florida 16-5
Tennessee 15-9
Alabama 13-11
Mississippi 13-11
South Carolina 13-11
Arkansas 12-12
Mississippi St. 9-15
Kentucky 4-20
Vanderbilt 3-21

Now the road warriors.

Georgia 17-4
Tennessee 18-6
Auburn 16-8
LSU 16-8
Florida 13-8
Alabama 11-13
South Carolina 11-13
Arkansas 9-15
Mississippi 9-15
Kentucky 4-20
Vanderbilt 3-21
Mississippi St. 1-23

Here are the leaders in delta points at home. Delta points are net points. Florida’s +331 means they have outscored their home opponents by 331 points (roughly 15.8 per game).

Florida 331
Auburn 279
Georgia 269
LSU 181
Tennessee 178
Alabama 123
Arkansas 64
South Carolina 42
Mississippi 2
Mississippi St. -217
Kentucky -228
Vanderbilt -270

Now the leaders in road delta points.

Georgia 217
LSU 198
Florida 141
Tennessee 124
Auburn 110
Alabama 101
Arkansas -113
South Carolina -126
Mississippi -181
Kentucky -322
Vanderbilt -422
Mississippi St. -477

Next up is a ranking of 'relative homefield advantage' by difference in home/road winning percentage. Mississippi St. has the largest discrepancy in home and road play in this category, winning at a .375 clip at home (9-15) versus an atrocious road record of 1-23 (.042) for a difference of .333.

Mississippi St. .333
Auburn .167
Mississippi .167
Florida .143
Arkansas .125
LSU .125
Alabama .083
South Carolina .083
Georgia 0
Kentucky 0
Vanderbilt 0
Tennessee -.125

Some observations. In the past half-decade Mississippi St. has been simply awful on the road. Their only road win in the 2000’s was in 2000 at Kentucky. They have won an unbelievable 9 times as many home conference games as road games in this time span. Surprisingly, Georgia, Kentucky, and Vanderbilt are just as likely to win at home as they are on the road. Of course this does not mean these teams similar. Georgia (along with Auburn and LSU) has been the class of the conference since 2000 and are a very good road team. Their homefield advantage is only relatively smaller because they win so many road contests. Kentucky and Vanderbilt on the other hand have been terrible both at home and on the road. Another surprise has Tennessee with a better road conference record in this six year span than home ocnference record.
Finally, here is a ranking of 'relative homefield advantage' by difference in delta points at home and on the road. Mississippi St. again has the largest discrepancy in home and road play in this category. They have -217 delta points at home and -477 on the road for a difference of 260 delta points.

Mississippi St. 260
Florida 190
Mississippi 183
Arkansas 177
Auburn 169
South Carolina 168
Vanderbilt 152
Kentucky 94
Tennessee 54
Georgia 52
Alabama 22
LSU -17

From this second set of data, the shocker is that LSU actually has more delta points on the road than at home. What this should mean is that they have a better record in close games at home than on the road. A quick stroll through the data reveals just that. At home in conference games decided by one score (8 points or less) they are a remarkable 10-1 since 2000. In similar road games, they still managed a winning record, but it was only 5-3. This is nothing endemic to this team’s character, they have probably just gotten the lucky bounces at home and not gotten as many on the road. This also shows that they have played more close games at home than on the road. Not something you would expect from the team that plays in a place called ‘Death Valley’. As usual, all coments and questions are welcome. Big 12 and Pac 10 studies coming soon.


Anonymous said...

You might be interested in this article from the Journal of Quantitative Analysis in Sports. (Free registration required.)

If you have a problem accessing the article, send a message to the "List owner" email address of my Web site and I'll send it to you as a PDF.

Anonymous said...

ESPN's Ivan Maisel reports that the Pac-10 is going to become easier to study because "the Conference presidents voted for a 12th game only on the condition that it be a conference game. The league will play a round-robin schedule for the first time since it expanded in 1978."

matt said...

Hopefully the Pac 10's nine game schedule will last for a while and not be a short-lived fad.