12/27/21: NET Rank v. AP Poll

[visitingcock]

NET AP

• South Carolina.........................................South Carolina
• NC State Stanford
• UNC Louisville
• Texas Arizona
• Stanford NC State
• Louisville Maryland
• Nebraska Tennessee
• Arizona Indiana
• Iowa State Michigan
• Indiana Baylor
• UConn.........................................................Uconn
• K-State Texas
• Notre Dame UGA
• UGA Iowa State
• Maryland Duke
• Tennessee Georgia Tech
• Ohio State Notre Dame
• Oregon BYU
• Mizzou State LSU
• VaTech KY
• Baylor Iowa
• Utah USF
• LSU Tx A&M
• BYU UNC
• Arkansas Ohio State

Apparently the NET hasn't settled yet. Given their schedules, UNC and Nebraska certainly shouldn't sniff the top 20 til they beat somebody. Maryland seems too low by the net.

SCar and UConn are the only teams ranked the same by humans and algorithm,

 

[Alydar]

Too bad for Scar that the NCAA's aren't starting next Friday.

 

[Sifaka]

NET AP

• South Carolina.........................................South Carolina
• NC State Stanford
• UNC Louisville
• Texas Arizona
• Stanford NC State
• Louisville Maryland
• Nebraska Tennessee
• Arizona Indiana
• Iowa State Michigan
• Indiana Baylor
• UConn.........................................................Uconn
• K-State Texas
• Notre Dame UGA
• UGA Iowa State
• Maryland Duke
• Tennessee Georgia Tech
• Ohio State Notre Dame
• Oregon BYU
• Mizzou State LSU
• VaTech KY
• Baylor Iowa
• Utah USF

• LSU Tx A&M
• BYU UNC
• Arkansas Ohio State

Here's Massey's top 25.

South Carolina

Arizona

Texas

NC State

Stanford

North Carolina

Louisville

Connecticut

Tennessee

Maryland

Nebraska

Iowa St

Baylor

Indiana

Georgia

Notre Dame

Oklahoma

Michigan

Kansas

Duke

LSU

Ohio

BYU

Texas

Oregon

 

[visitingcock]

Too bad for Scar that the NCAA's aren't starting next Friday.

We have our banner ready if there aren’t any NCAAs

 

[visitingcock]

Here's Massey's top 25.

South Carolina

Arizona

Texas

NC State

Stanford

North Carolina

Louisville

Connecticut

Tennessee

Maryland

Nebraska

Iowa St

Baylor

Indiana

Georgia

Notre Dame

Oklahoma

Michigan

Kansas

Duke

LSU

Ohio

BYU

Texas

Oregon

I don’t get Massey’s ranking of UNC given their weakness of schedule. It’s like 164th. Oh well, two weeks of conference play should see a big fall

 

[triaddukefan]

Are the NET and Massey computers based out of Chapel Hill.? .. just asking

 

[TheFarmFan]

The best explanation I can give for Massey is that his rankings heavily weigh for Margin of Victory, and at present UNC has won by an average of 33.1 points per game. SOS adjusts this down, obviously, but that's still the best in D1. Note also that Nebraska's is 28.1 (And of course, the virtue of being undefeated is that there are no games where your opponent outscored you to drag down your average.)

Unless and until those teams lose, Massey's MoV calculation will keep them ranked higher than the eye test probably suggests they warrant. That's both a feature and a bug of his approach - until someone is beaten, if they win all their games by large margins, who's to stay they won't continue to?

But since UNC faces NC State, Va Tech and Notre Dame in the first three games of the new year, and Nebraska will face Michigan, Iowa, and Indiana, those teams will either drop or they'll win out and actually probably deserve the rankings they currently have.

 

[Centerstream]

Are the NET and Massey computers based out of Chapel Hill.? .. just asking

Or are any Chapel Hill computer programming graduates coders for these algorithms???

 

[Sifaka]

The best explanation I can give for Massey is that his rankings heavily weigh for Margin of Victory, and at present UNC has won by an average of 33.1 points per game. SOS adjusts this down, obviously, but that's still the best in D1.

This is taken from Massey's FAQ.

“Explain how your system uses margin of victory (MOV).

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.”

 

[Alydar]

I don’t get Massey’s ranking of UNC given their weakness of schedule. It’s like 164th. Oh well, two weeks of conference play should see a big fall

I guess it's because it's an objective rating that isn't influenced by comments from coaches.

 

[TheFarmFan]

This is taken from Massey's FAQ.

“Explain how your system uses margin of victory (MOV).

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.”

Hmm, I think you may have grabbed the part of the FAQ that discussed how his ratings were used when the College Football BCS system incorporated computer rankings, including his. As Massey himself notes at the top of that page, those FAQ responses are out of date (RIP BCS System, almost a decade gone). But from the same page:

First, to avoid confusion, be aware that I publish two different sets of rankings:

• the "Massey Ratings", which utilize actual game scores and margins in a diminishing returns fashion
• the BCS compliant version, which do not use the actual score

Massey Ratings are what we're all looking at for WCBB. And I'm pretty dang certain that MOV is the major innovation in Massey's ratings versus RPI and NET ratings, because I've read about this as a key difference in a number of places. And as far as I know, ELO is the only other rating that takes MOV into account.

Edited to add: Elsewhere, where Massey provides a description of the theory behind how the calculation works, he makes clear he takes score into account:

Only the score, venue, and date of each game are used to calculate the Massey ratings.

 

[EdSaid]

The only in common opponent for unc and UConn is Minnesota. Unc beat them by six UConn by 30. Unc has played no ranked teams. I like banghart and what she’s done and doing but but that makes no sense

 

[darkstar1]

The most accurate barometer of the likelihood of a team to win the natty is the Vegas futures odds. As of today, South Carolina is +210 and UConn +240. Third is Stanford at +600. No other team is less than +1000. There’s your 1,2,3. I trust the Las Vegas oddsmakers more than the pollsters. They actually have something real at risk.

 

[visitingcock]

The most accurate barometer of the likelihood of a team to win the natty is the Vegas futures odds. As of today, South Carolina is +210 and UConn +240. Third is Stanford at +600. No other team is less than +1000. There’s your 1,2,3. I trust the Las Vegas oddsmakers more than the pollsters. They actually have something real at risk.

Actually, they don’t set odds based on who they think will win and by how much. They set spreads based on what they think will spur betters on both sides to bet.

Bookies put out spreads that will induce betting. They don’t bet against you if they can avoid it. They get other people to bet against you by moving the spread to where people will bet both sides. They take 10% (juice) of all bets

 

[Alydar]

Actually, they don’t set odds based on who they think will win and by how much. They set spreads based on what they think will spur betters on both sides to bet.

Bookies put out spreads that will induce betting. They don’t bet against you if they can avoid it. They get other people to bet against you by moving the spread to where people will bet both sides. They take 10% (juice) of all LOSING bets.

Edited. Bookies don't charge winners. Maybe sportbooks do but not bookmakers.

 

[darkstar1]

In any event, in order to spur betting on both sides, Sportsbook lines tell who the favorite is and by how much. The presumed better team is favored to win. Very different than a ranking, where a superior team may be lower in the rankings yet favored to win.

 

[visitingcock]

No. They make odds and handle bets. They are not predictors of outcomes.

 

[Vowelguy]

B does not follow from A

 

[Alydar]

The only in common opponent for unc and UConn is Minnesota. Unc beat them by six UConn by 30. Unc has played no ranked teams. I like banghart and what she’s done and doing but but that makes no sense

We've seen Baylor play much worse OOC schedules early and stay highly ranked in the AP. If they are a paper tiger then the ACC schedule will expose it.

 

[Alydar]

The bookies set lines that will produce an equal amount of wagering on either team. Lines get adjusted or not based on the betting. If you hear of a bookie who "got killed" on a certain game then it's likely that it wasn't a bookie but a gambler who sat on some wagers. That isn't bookmaking.

 

[Alydar]

I have little actual experience betting on sports. My hobby is betting on horses but I have known several bookies fairly well over my life. I also worked in the gaming industry for 40 years. Mostly horse racing.