Computer ratings have more bias

At least when the writers and coaches underrate you, they can make a major adjustment when it’s wrong. Sagarin and RPI started us so low that even after dominating OU, OU is still ranked much higher than UH in both ratings. Hopefully the committee sees through that bias when determining the top 4 if UH finishes undefeated.

YEP! I Tweeted on Sagarin last night. It’s pure garbage. Not is it logical or rational, as many people assume it to be. There are a number of highly subjective assumptions that lie at that base of this rating. We will likely never know WHAT they are but we can certainly infer their existence from the behavior if the ratings themselves.

What, if anything, was his response to how OU is 4 now, and UH is 17?

Computers tend to rank based off the entire picture rather than one game. For example, Sagarin takes in to account future SOS and may look at projected victories (I know some do and some don’t).

The remaining SOS is a concern as only Louisville may be considered a “good” matchup. The AAC is definitely shaping up to be much weaker this year.

well then they’re giving it too much weight, which shows bias…

Agreed.

Massey has us at #7. Lowest ranking is ESPN’s RPI that has us at #30. Moore and England have us highest at #2.

http://www.masseyratings.com/cf/compare.htm

Edit: Check that, Massey has us at #16, but the Composite has us at #7.

It’s not ranking off the “entire picture” if SOS is still based off of what he projects the teams will do and not what they have done. In other words, when P5 teams lose a few players, he considers that they are reloading and they get the same rating as what they had at the end of last year or get an upgrade. Teams that aren’t in the P5, he considers losing starters and down grades all the teams. So the SOS is severely bulked up for the P5 teams and is severely underrated for all the G5 teams.

Several years ago I went through a few years of looking at his final rankings for one season and the preseason rankings for the next. He reranks teams as if they are starting from scratch each year. It’s not fair to the successful G5 programs that work their way into the top 25 each year but start the next season 10 or 15 stops back only to earn their way to same spot they earned the previous year.

1 Like