Better Know A BCS Computer: Jeff Sagarin Ratings

by — August 3, 2010 at 2:16 pm | 0 comments


[Editor's note: In August of 2010, BCS Know How took a look at each of the six BCS computers using all available information and a multitude of sources. Some formulas are more public than others. I aimed to make the most inclusive collection of facts about all of the computers, with a varying degree of success.]

If you asked the casual college football fan to name one computer ranking, they’d probably come back with the Sagarin Ratings, run by the USA Today and Jeff Sagarin.

Sagarin has had to modify his popular formula over the years, especially in light of the BCS’s campaign against margin of victory calculation. His original formula included the variable, but in the ratings he submits to the BCS, he has taken out margin of victory to conform with the BCS’s wishes.

The ratings formula is a well kept secret, but the major variables are wins, losses and venue of game.

Sagarin’s strength of schedule is implicit in his rankings, as each rating output compounds both records of opponents and records of opponents’ opponents.

Sagarin includes a preseason ranking system and weighs it through the first few weeks.

However, when the first BCS rankings are released, Sagarin removes the preseason ranking, allowing for each team to be ranked as if they was not a preseason ranking — all starting at the same spot.

The ratings are basically a predictive rating, as opposed to Massey’s ratings, which are simply a measure of what has occurred before.

In order to predict a game, all a viewer must do is subtract the rating distance from the favorite to the underdog, and add three points to the home team. The distance between the two teams should be an accurate predictor of final spread when game time rolls around.

Because Sagarin views road wins as more difficult (three points more difficult), road wins garner more weight in his process.

Sagarin also seems to favor undefeated and single-loss teams more than some of the other ranking systems.



Comments are closed.