Race, the South, and the NFL

I was reading this post on The Blue-Gray Sky, and it got me thinking about the effects of race on football. Big 10 Commissioner Jim Delaney recently wrote a long ranting letter about how his conference rocks and comparing it to the SEC. In doing so, he basically called out the SEC, saying "It seems premature for us to lower our admission standards or give up on the tremendous talent pool in the Midwest."

While I don't dispute that the Midwest has talent, I wanted to look at the prevalence of NFL players by state. Thanks to this valuable site and Census Bureau estimates, I was able to rank the states according by per pcapita NFL participation. Here is the list:

State NFL Players per Million Residents
DC 17.20
MS 16.15
LA 14.46
SC 11.34
AL 10.00
FL 9.34
GA 8.97
TX 7.36
VA 7.33
HI 6.22
NE 6.22
AR 6.05
NC 5.98
WA 5.79
OH 5.66
CA 5.43
US 5.42
OK 5.31
PA 5.22
NJ 5.04
MO 4.96
ND 4.72
AK 4.48
MI 4.46
IA 4.36
MD 4.27
AZ 4.22
KS 3.98
TN 3.81
CO 3.79
OR 3.78
WI 3.78
IL 3.66
UT 3.53
IN 3.33
KY 3.09
WV 2.75
ID 2.73
MA 2.64
DE 2.34
MN 2.13
MT 2.12
NY 1.76
CT 1.71
NM 1.53
SD 1.28
NV 1.20
RI 0.94
NH 0.76
ME 0.00
VT 0.00
WY 0.00

As you can see, the list is dominated by the south. Taking it one step further, I plotted these numbers against each state's percentage of African-American residents. The results:


(Click on chart to enlarge)

I knew the correlation would be pretty strong, but I didn't expect quite so powerful. For the statistically disinclined, the equation says that all other things being equal, a state with no black people will have 1.87 NFL players per 1,000,000 residents and one with all blacks will have 28.6. It also says that the percentage of black residents explains 65.3% of the variance between states in per capita NFL players.

Incidentally, Canada has 11 NFL players. These are the things I do at work when I have free time.