Do You Use Comparative Scores for Your Rankings?

[No_Skill]

When evaluating where you feel teams deserve to be ranked, do you compare performances using like opponents.

ex: Team A beat team B and team B beat team C, so team A should be ranked higher than team C

I have heard some of the arguments both ways. What do you do?

[Peems]

you judge game to game. because if you used comparative scores to rank it would be a mess. imagine trying to do that at the 1-A level. florida lost to auburn who lost to arkansas who lost to USC who lost to oregon st. who lost to cal who lost to arizona who lost to LSU who lost to florida.

confusing, no?

[No_Skill]

you judge game to game. because if you used comparative scores to rank it would be a mess. imagine trying to do that at the 1-A level. florida lost to auburn who lost to arkansas who lost to USC who lost to oregon st. who lost to cal who lost to arizona who lost to LSU who lost to florida.

confusing, no?

Confusing? Hell yes. Especially when you put in a few degrees of separtation.

Personally, I wasn't thinking of taking it out that far (A beats B, B beats C,... So A would beat G). The closer the relation the better.

[*****]

margin of victory can certainly be telling as are A beat C, B beat C stuff for rankings. With a bit more knowledge (like why both beatings occured) it can be valid.

[Peems]

Confusing? Hell yes. Especially when you put in a few degrees of separtation.

Personally, I wasn't thinking of taking it out that far (A beats B, B beats C,... So A would beat G). The closer the relation the better.

i understand. but the point i was making is every game is different and to use comparative scores can get out of hand. but as always mentioned, that is why CS is the best!

[FCS_pwns_FBS]

New Hampshire got moved ahead of App State after they beat Northwestern and App State lost to NCSU (usually you don't knock someone down for losing to a IA), so I would say that many AGS voters do.

[skinny_uncle]

margin of victory can certainly be telling as are A beat C, B beat C stuff for rankings. With a bit more knowledge (like why both beatings occured) it can be valid.
But when A beats B, B beats C and C beats A, it gets very confusing. Would you like some examples from the Gateway this season?
:confused:

[JohnStOnge]

FYI, the power ratings (models) that do the best job of predicting outcomes of games are the ones that do consider margins of victory.

[Peems]

But when A beats B, B beats C and C beats A, it gets very confusing. Would you like some examples from the Gateway this season?
:confused:

look at the 1-A example i gave.

[*****]

But when A beats B, B beats C and C beats A, it gets very confusing. Would you like some examples from the Gateway this season?Usually the GFC is usually pretty easy to understand with weather and star player status for the game. So you see with that extra nugget of info then suddenly you may see why Team C beat Team A.

[skinny_uncle]

Usually the GFC is usually pretty easy to understand with weather and star player status for the game. So you see with that extra nugget of info then suddenly you may see why Team C beat Team A.
Which of these affected your prediction on the YSU-WKU game yesterday?
:D

[*****]

Which of these affected your prediction on the YSU-WKU game yesterday?I don't make predictions but YSU obviously has it going on while WKU had its problems this year.

[ngineer]

Each game is different. Comparative's have limited usefulness. Weather, what stage of the season, injuries, style of play, etc., all have differing impacts on each game.

[Jaxhen]

Another big factor is how individual teams match up. For instance, one team's strength may exploit another team's weakness. A team with a good run defense may do well against a team that emphasizes the run: whereas, that same team may really struggle with a strong passing team. Also, sometimes it's just a matter of how the ball bounces that day or key mistakes, momentum shifts during games, injuries, etc. There are so many factors that influence how a game turns out. This is why you often see a team that lost during the regular season turn around and beat that same team in the playoffs.

[SoCon48]

When evaluating where you feel teams deserve to be ranked, do you compare performances using like opponents.

ex: Team A beat team B and team B beat team C, so team A should be ranked higher than team C

I have heard some of the arguments both ways. What do you do?
Heck in some rankings (like Hansen'sxlolx ), beating a team head to head by a decent margin means nothing.

[SoCon48]

Each game is different. Comparative's have limited usefulness. Weather, what stage of the season, injuries, style of play, etc., all have differing impacts on each game.

Common opponents are one of the factors used in all statistical rankings and have similar drawbacks as any criteria.
Having said that, if team A beats team B by 30 pts. Team B beats C by 20. Needless to say, the smart money is on A when A and C meet.