Assume the BCS continues with the two top teams: The human polls would still be used. The number 1 team would receive one point. However, every other team would have a “power” rating by dividing the number of points it receives by the votes received by the number one team. Example, assume Team “A” is number one with 100 votes, Team “B” is second with 125 votes and Team “C” is third with 150 votes. Team “B” would have a rating not of “2” as it now does, but of 1.25 (125 divided by 100) and Team “C” would have a rating of 1.5 (150 divided by 100). This system better represents the consensus of the voters on the teams’ relative strengths. Now, the tenth ranked team is considered 10 times worse than the number one team, which its not. Computer polls would be used, but limited to the five polls that have been the best predictors over the previous three years. Not to get too complicated those polls with the smallest standard deviation from the actual game scores for the previous three years. (Games for the previous three years are compared to the power ratings of each computer poll. Those polls that are closest to the actual scores would have a smaller standard deviation.) We would then know that we were using those polls that have been shown to be most valid and reliable. The power ratings would be converted and averaged. The number 1 ranked team would automatically have one point, with the other teams ranked on a per centage basis based on their ratings compared to the number 1 ranked team. There would be no strength of schedule. The way now done is screwy, but also unnecessary. It gives too much weight to strength of schedule. The human voters, on a subjective basis, and the computer polls on a more “objective basis", take this into account and it should not be considered twice. There will be no "quality wins" and playing a team twice would simply be factored in as if the teams were playing the first time. Margin of victory would be allowed up to a certain score range (say somewhere between 40 and 50 points). Margin of score does make a difference. Suppose Team “A” and Team “B” each play in the same three period Teams “C”, “D” and “E”. Team “A” beats each by three points; Team “B” by 21 points. Does anybody not agree that Team “B” is better than Team “A”? Now, the two teams are equally ranked. Advantages: Only the five most reliable and valid computer polls are used, increasing validity; margin of victory used, but no incentive to run up the score; most people could at least understand the system. Whaddaya think?
Let me take a whack at it: 1) I say you totally do away with the entire low to high BCS scoring system. Start from 0 and go up. The highest point total is 1st, the lowest point total is last. 2) I like MOV. It should be used. It makes no sense that Ohio State was barely beating marginal teams and getting credit the same way OU, LSU, or USC was dominating teams. I say you use the Las Vegas odds to determine the predicted MOV, and then you award or punish teams based on the outcome (relative to the projected MOV). For every 3 points that you go over or under the projected MOV, you give or take away 1 point. For instance, OU projected to win by 31 points and wins by 38, gets credit for 2 points since they won by 7 points above the projected MOV. You place a cap at 5 points awarded for going over to avoid 77-0 scores. There would be no cap for going under. Sorry Iowa State, Vanderbilt, SMU and all other deadbeat programs. 3) SOS is mandatory. You can't have the TCU situation again where they would have finished somewhere in the top 10 (probably 7th) in the BCS had they gone undefeated and they would not have beaten anyone in the top 45 of the BCS (Southern Miss would be about 46th in the BCS had they lost to TCU). In fact 9 of their 11 opponents are 65th or worse in the BCS. That is unacceptable. The current SOS ranking system is fine, except you would have to re-do the scoring to fit my scoring system. There are currently 117 1-A teams. The team with the best SOS would get 117 points, the 2nd would get 116, etc. 4) QW needs a total revamping. You should get credit for QW against any team in the top 25 of the BCS, not just the top 10. Additionally, you get the credit for the position the team was in when you played them, not at the end of the year. For example, when Auburn beat Arkansas in early October and Arkansas was ranked in the top 10, Auburn should get credit for the top 10 ranking Arkansas had then, not the #31 ranking they have now. You give 1 point for each position from 25 to 1. A win against a #1 team gives you 25 points. And yes if you play a team for a 2nd time and beat them twice, you get the credit for both wins. 5) Every win you get 15 points. Every loss you deduct 5. You cap the win total at 11 games. Teams who play more than 11 games, you still count those games for MOV, SOS and QW. 6) Div II-A do not count at all for any points. 7) Screw all the polls
You mean 64 team....they wouldn't let those boys play than many games and it would extend the season all the way into february and into the spring semester. 8 teams is the most I can see they would go with.
Well, maybe you weren't...but I sure was...And the only reason I thought you were joking is that like islstl said, that is unreasonable....Talk about injuries, exhaustion, and an unrealistically longer season....
