I'm working on a pseudo power ranking formula for next season to help determine who the most impressive teams have been. I am thinking of two versions, one a ranking that shifts with each game (i.e. The Jets beat the Raiders by 20-10) and one that is based on a teams overall season long rankings (The Jets are 2-0 and winning by an average of 12.0 points per game). The basic criteria that I believe will be important for the game to game one is win/lose, team ranking differential (#12 team beats a #4 team for instance), point differential and home/away. The first question I have been wondering is if there is any category I might be missing that could be important on a game by game basis. There will be slightly different categories based on the overall season ranking but that will come later. Separately I have questions of which is the most important outcome in each of the categories listed above. Obviously a win is more impressive than a loss and a win on the road is more impressive than a win at home. But what about the other two? What is more impressive, winning by 20 or winning by 2? At first glance I would say blowing everyone out is more impressive but there is something about winning close games that is important. On the other hand, you might just be squeeking them out and not really proving anything. Also, what is more impressive, a #32 beating a #10 or a #31 beating a #9? Each is 22 spots higher in the ranking but what would hold more weight, the winning team's ranking or the losing team's ranking? In other words is it more impressive that a certain team lost or that a certain team won? Thanks for your help, lets see where this takes us.
