Thursday, September 23, 2010

Game Watchability Score – 2010 Week 4

I came up with a method that weighs the quality of teams and closeness of teams equally (with a home field adjustment thrown in). Games will score well if the teams are ranked high and the teams are closely matched.

The list is ranked by the ‘Game Watchability Score’, with the teams and their rankings. Remember it's early, the ratings used for this are based on 2-3 games per team. Still a fun method to look at it objectively. I’ll post these each week.

I’ll explain the methodology below in case you are interested.

Please click the “read more” link to see the scores. The rankings listed do not include games from Thursday/Friday and they combine FBS and FCS.

3.07 Oklahoma (18) @ Cincinnati (12)
2.92 Oregon (3) @ Arizona State (34)
2.90 Temple (21) @ Penn State (11)
2.88 Virginia Tech (26) @ Boston College (23)
2.84 Central Michigan (16) @ Northwestern (40)
2.73 UCF (24) @ Kansas State (46)
2.73 Alabama (1) @ Arkansas (36)
2.72 Cal (27) @ Arizona (13)
2.61 Florida International (32) @ Maryland (54)
2.44 Miami (FL) (9) @ Pitt (51)
2.32 Stanford (2) @ Notre Dame (49)
2.09 West Virginia (47) @ LSU (20)
1.99 Northern Colorado (48) @ Michigan State (78) - take note of the higher ranked team
1.95 South Carolina (31) @ Auburn (68)
1.90 Georgia (53) @ Mississippi State (80)
1.82 Kentucky (56) @ Florida (25)
1.78 Wake Forest (64) @ Florida State (44)
1.76 UCLA (62) @ Texas (38)
1.66 Liberty (55) @ James Madison (7)
1.54 Oregon State (58) @ Boise State (6)
1.45 NC State (5) @ Georgia Tech (75)

Which is the least interesting game? Austin Peay (#231) @ Wisconsin (#33), scores –3.38.

To figure Game Watchability Score:

For the quality rating, take the ratings of the two teams and multiply them. Figure that for all games.

Take the game quality rating and subtract the average game quality rating. Divide by the standard deviation of all games.

For the closeness rating, take the absolute value of home team -
road team rating (with a HFA built in of +.025/-.025).

Take the average closeness factor and subtract the game closeness
factor. Divide by the standard deviation of all games.

This puts each factor on the same scale. Add the standard deviations
together and you get your game rating.

No comments: