In other news (apolitical thread)...

[Gigem]

I was looking through some team records the other day.  There are teams out there that are 6-1 and 7-0, now maybe 7-1 that are not ranked, and teams that are 6-2 that are ranked.  

[utee94]

I was looking through some team records the other day.  There are teams out there that are 6-1 and 7-0, now maybe 7-1 that are not ranked, and teams that are 6-2 that are ranked. 

SOS matters.
But of course, there are also teams ranked #3 that have a lower SOS than teams ranked #20.

[betarhoalphadelta]

I get that the conference vs OOC thing is realistic, but I don't think it is very useful here.

We all want the games to matter, but when h2h is 8.3% of a season, how is holding it over 91.7% of the season wanting the games to matter?

Again, this is why I asked you to bound the question. Please be more specific. You're making a claim that someone is holding H2H over something, perhaps "bad losses".

My questions... WHO is holding H2H over other factors unfairly? In what circumstances? With what negative effects?

If you can't be more specific, we can't meaningfully engage your question.

I.e. if you made the charge, "I believe the CFP selection committee is overweighting H2H results and here are multiple examples", we can have a very fruitful discussion.

But for now it's just a nebulous accusation that someone, somewhere, is weighting H2H over more important but barely named factors. That's not enough to engage because none of us can actually be sure what the claim even is.

[OrangeAfroMan]

Again, this is why I asked you to bound the question. Please be more specific. You're making a claim that someone is holding H2H over something, perhaps "bad losses".

My questions... WHO is holding H2H over other factors unfairly? In what circumstances? With what negative effects?

The consensus seems to be to hold h2h more than any other outcome, even when things aren't equal.  But when teams have equal records, the masses tend to treat that as all else being equal (a separate issue).

It's hard to make it more specific when the idea is held by the masses, lol.  

When we have a tie at the top of a conference, the h2h winner tends to have a worse loss.  That means their performance has more variance than the other team:  higher ceiling AND lower floor.  Is a higher-variance team more likely to have had a better season?  Isn't it better to be a lower-variance team (ie - consistent, reliable)?  Why not reward that team with the tie-breaker?