Facts. Try not to think of it as a poll but more like a periodic table. They designate a rating to each team to determine the associated spread that should be assigned for each matchup. For instance, this upcoming week #4 USC (117.8) plays #9 Oregon (111.3). I would have to see more games played between ranked teams, but it may be simple subtraction. 117.8 - 111.3 = 6.5, then you factor in the game being played in Oregon (minus 3 pts). The current spread is 3.5 or 4 depending on where you look. One more #2 Texas (119.6) plays at #15 OK St (108.7). 119.6 - 108.7 = 11.6. Current spread is 9. Roughly discounting the home field advantage.
this system seems totally screwy to me and i cant believe they actually use it because it doesnt account for matchups. quite often a #10 team will have a much better chance against a #3 team than a #8 team because of matchups.
Dont quote me on this, but this poll does account for matchup issues, thats why the big nuber (110.8 or whatever it is) can change week to week. If the oddsmakers feels like LSU matches up welll against Bama then the number will reflect that. I like the comparison to a periodic table. Dont worry about the numerical ranking but the larger number it tells the entire story of how this poll is made up. It can also change if betting tendencies go one way or the other, if an assload of people bet on LSU then our big numer would change to reflect the spread. It is totally set up for the spread. Got this from a friend who bets alot so it is only as accurate as my source, a compulsive gambler. Kinda like the drug dealer in The Hangover.
