Computer Ranking - Pythagorean Poll (PYTH)
Pyth after week 12 RK Prev TEAM PYTH SOS adjPYTH W/L% SCORE 1 1 Iowa 0.7878 0.444 0.7440 0.917 0.830 2 2 Arizona 0.6749 0.451 0.6421 0.833 0.738 3 3 USC 0.6954 0.438 0.6519 0.750 0.701 4 5 Notre Dame 0.5486 0.535 0.5676 0.583 0.575 5 4 Alabama 0.4939 0.438 0.4630 0.583 0.523 6 8 Arkansas 0.5142 0.465 0.4963 0.500 0.498 7 7 Miami 0.4604 0.500 0.4604 0.500 0.480 8 6 Ohio State 0.4734 0.563 0.5029 0.417 0.460 9 10 Oklahoma 0.4411 0.542 0.4595 0.333 0.396 10 9 Florida State 0.3722 0.535 0.3851 0.333 0.359 11 11 Texas 0.4411 0.472 0.4288 0.250 0.339 12 12 UCLA 0.0972 0.618 0.1086 0.000 0.054
TEAM WK1 WK2 WK3 WK4 WK5 WK6 WK7 WK8 WK9 WK10 WK11 WK12 PYTH ProjW ProjL Alabama 0.08 0.25 0.95 0.58 0.70 0.62 0.07 0.71 0.00 0.97 1.00 0.000 0.4939 5.93 6.07 Arizona 0.04 0.91 0.65 1.00 0.62 1.00 0.69 0.62 0.64 0.69 0.76 0.475 0.6749 8.10 3.90 Arkansas 0.69 0.15 0.35 0.43 0.30 0.66 0.63 0.84 0.74 0.34 0.03 1.000 0.5142 6.17 5.83 Florida State 0.53 0.09 0.26 0.02 0.00 0.88 0.20 1.00 0.81 0.03 0.26 0.394 0.3722 4.47 7.53 Iowa 0.96 0.85 0.74 0.92 0.54 0.98 0.80 0.29 0.88 0.97 0.74 0.801 0.7878 9.45 2.55 Miami 0.92 0.37 0.62 0.08 0.46 0.12 0.04 0.16 0.75 0.59 0.82 0.606 0.4604 5.53 6.47 Notre Dame 0.26 0.75 0.38 0.26 0.38 0.63 0.97 0.97 0.12 0.66 0.65 0.525 0.5486 6.58 5.42 Ohio State 0.74 0.22 0.59 0.98 1.00 0.02 0.96 0.22 0.26 0.31 0.18 0.199 0.4734 5.68 6.32 Oklahoma 0.47 0.78 0.26 0.42 0.54 0.38 0.37 0.03 0.36 0.06 0.97 0.656 0.4411 5.29 6.71 Texas 0.39 0.88 0.41 0.57 0.46 0.34 0.93 0.38 0.19 0.41 0.00 0.344 0.4411 5.29 6.71 UCLA 0.31 0.12 0.05 0.00 0.01 0.00 0.03 0.00 0.25 0.03 0.35 0.007 0.0972 1.17 10.83 USC 0.61 0.63 0.74 0.74 0.99 0.37 0.31 0.78 1.00 0.94 0.24 0.993 0.6954 8.34 3.66
Pythagorean Scoring Poll (PYTH): This poll ranks each team by applying the Pythagorean formula to their Points For (PF) and Points Allowed (PA). For football, statistic experts have found that an exponent of 2.37 seems to work best for predicting expected record for college football. So the Pythagorean Theorem for football looks like this: PF^2.37 / (PF^2.37 + PA^2.37).
- Some would argue that blowouts, especially extreme blowouts can artificially inflate or deflate a team's Pythagorean score depending on whether or not they received or doled out the beating. The solution for this margin of victory issue? The Pythagorean percentage is computed on a game by game basis, then each week's score is added together and divided by games played. This way each game is counted the same and the effect of blowouts is lessened.
- The initial Pythagorean score is weighed against your team's Strength of Schedule (SOS) rating to take into account that it may be inflated or deflated based on your opponents. The way this works is your SOS is compared to a .500 SOS to determine your SOS weight % (e.g. if you have an SOS of .600, your weight % would be 110%, thus slightly increasing your Pythagorean score). [This was added to the formula on 11/7/2010, a week before the first BCS rankings were published]
- In the end, football is not just about scoring, so your Pythagorean score is averaged with your W/L % to get a final score for ranking purposes.
- The team with the highest final score gets 1.000 and the rest of the teams divide their final score by the final score of the team with the highest score to determine their BCS component %.
The great thing about this poll is that it encourages teams to play their best throughout the season. Teams that slack off at the end of the season could potentially end up hurting their Computer ranking.