Massey's top 20

[Glenn]

I would assert that the consideration of previous “years’ final ratings” (note use of plural -not year’s) does not disappear after ten games as evidenced by Oregon’s continuous over ranking last year. Either that or the algorithm places too much emphasis on “good losses” .

Oregon had a slew of good losses and no good wins but stayed in the top six. - frequently above two of the teams that swept them.

I should be careful. We could end up being a good win for Oregon during the first ten games of this year.

I agree that it probably takes more than 10 games to remove the effects of prior years, owing to the paucity of good interconference games. That said, one other factor that you might be minimizing is the fact that the Massey algorithm includes margin of victory. I think in the Oregon case it "liked" the fact that Oregon walloped mediocre teams, such as USC by 20, and barely lost to good teams--like by 2 to Stanford. So, Massey saw some "good" wins. The other difficulty last year was the lack of interconference games which is the key to being able to rank teams across conferences.

 

[Sifaka]

I agree that it probably takes more than 10 games to remove the effects of prior years, owing to the paucity of good interconference games. That said, one other factor that you might be minimizing is the fact that the Massey algorithm includes margin of victory. I think in the Oregon case it "liked" the fact that Oregon walloped mediocre teams, such as USC by 20, and barely lost to good teams--like by 2 to Stanford. So, Massey saw some "good" wins. The other difficulty last year was the lack of interconference games which is the key to being able to rank teams across conferences.

What follows is Kenneth Massey discussing his college football algorithm. I couldn't find anything specific to his approach to college basketball, which may be different or the same. In the quoted material, he makes it clear that margin of victory is not a strong determinant of rank.
source: Massey Ratings FAQ

The BCS compliant version does not use MOV at all. There is no distinction between a 21-20 nailbiter, and a 63-0 blowout.
The main version does consider scoring margin, but its effect is diminished as the game becomes a blowout. The score of each game is translated into a number between 0 and 1. For example 30-29 might give 0.5270, while 45-21 gives 0.9433 and 56-3 gives around 0.9998

The maximum is topped at 1, so the curve flattens out for blowout scores. In addition, I do a Bayesian correction to reward each winner, regardless of the game's score.

The net effect is that there is no incentive to run up the score. However, a "comfortable" margin (say 10 points) is preferred to a narrow margin (say 3 points).

In summary, winning games against quality competition overshadows blowout scores against inferior opponents. Each week, the results from the entire season are re-evaluated based on the latest results. Consistent winners are rewarded, and a blowout score has only marginal effect on a team's rating.

 

[Sifaka]

From Massey, again:

“The effect of the preseason ratings gradually diminishes each week. When every team has played a sufficient number of games to be accurately evalulated based on this year alone, the preseason bias will be phased out.

Preseason ratings are based on an extrapolation recent years' results, tuned to fit historical trends and regression to the mean. A team's future performance is expected to be consistent with the strength of the program, but sometimes there may be temporary spikes.

Other potentially significant indicators (ex. returning starters, coaching changes, and recruiting) are ignored. Therefore, preseason ratings should not be taken too seriously.“

 

[Dokey]

Massey actually has the Big 12 as the second toughest conference. I think SCar's #2 SOS is based on non-conference scheduling of Stanford, UConn, Maryland, NC State, . . .

Massey's "predictions" are 100% based on last season's results. It's a 100% analytic-based rating system. As teams play actual 2022 games the 2021 season results will count less and will disappear once a team has played 10 games (I believe).

One good use of the site right now is that 2022 schedules are available by clicking on a team name.

Thanks for that clarification. It didn’t make much sense.

 

[Dokey]

Massey actually has the Big 12 as the second toughest conference. I think SCar's #2 SOS is based on non-conference scheduling of Stanford, UConn, Maryland, NC State, . . .

Clarifications are very much appreciated. SOS is definitely linked to non-conference schedule. I rank PAC-12 as toughest conference because of competition level top to bottom. In most conferences you either have one Alpha Dog, UConn in the Big East, or two or three barkers, say SEC, but in PAC-12 every game is a question Mark.

 

[nwhoopfan]

I think in the Oregon case it "liked" the fact that Oregon walloped mediocre teams, such as USC by 20, and barely lost to good teams--like by 2 to Stanford.

Good theory...except they lost to Arizona by 16 and 20 and to UCLA by 27! So much for that theory. And Oregon St. by 11.