Calculating wins vs ranked teams.

[NDQuebec]

There appears to be two ways to calculate a team's win-loss record vs ranked teams.

The first method is based on where the team is ranked at the time the teams face each other. The second method is based on where the team is ranked at the end of the season and not at the time the teams play each other.

Which one of the two do most NCAA football sites use, which one seems to be the most objective, and is one of them used officially by the NCAA, ESPN, etc.? Thanks.

 

[IrishTusker]

Definitely the one where you look at how the team did at the end of the season. Plenty of teams are hyped before the season and end up unranked.

Also, it probably makes sense to look at Sagarin or some system that ranks all the teams. Beating a team ranked 25 shouldn't count much more (if it should count for more at all) than beating a team ranked 26. This year, a bunch of teams we played were higher in Sagarin than their final AP ranking. (Louisville, USC, Army, Navy, Texas A&M.)

 

[ACamp1900]

I actually like taking point in time a bit more,… if it’s a top ten match mid season it shouldn’t not count because the team that lost loses a couple more and drops out. Where if you beat an unranked team early but they then get hot and finish 25th largely because of schedule,… eh. I get both sides tho

 

[MrNimbus]

The problem starts with preseason rankings imo. And it's an interesting question. At what point did fsu and michigan stop becoming ranked wins this season? Because teams that beat them got an undeserved bump in polls from playing them early. I wish they would scrap preseason polls altogether. They just put a ton of sec and big10 teams in it anyways.

 

[NDRock]

Whichever one makes Notre Dame look better.

 

[Bane]

I like both, think both ways have merit, and think that they add context to a team’s season. At the end of the day the polls are all to extent arbitrary and self reinforcing.