[stupac2]

I dont know if I completely agree, because there is a psychological aspect to college ratings that you dont have in the NFL. I think we get more "credit" for beating a 6-0 and an 0-6 team, then for beating two 3-3 teams. People like to see "quality wins". And unrelated to the Low APR U/UCLA game, I think we benefit  when a Pac12 team we play beats one that we dont play that year, because I assume it makes our SOS stronger. So while I agree in general with what you say, that we always want the south to beat the north because it helps us win the division,  I dont think it's black and white when you take into account selection for the Playoffs.

Okay sure, say we win out. ESPN predicts UCLA and Cal to both be 8-4 (rounding UCLA up and Cal down, which I think is fair since we're assuming we beat Cal). How much more impressive will it be for Cal to be 9-3 and UCLA 7-5? Or the other way around? Does it really matter? Either way we've beat one team that has a pretty good record and another that has an okay record. Who cares which team is which?

What I really don't get about this line of thought is that it's all so marginal. If we don't win out the records of the teams we beat are completely and utterly meaningless (I don't think anyone disagrees with this). If we do win out, we'll have wins over a highly ranked ND, a highly ranked South champion (whoever it is will be ranked), a probably-back-to-ranked-by-then Oregon, and presumably some other teams we've beaten will be ranked, though it's hard to say who will be. Will beating #15 Cal then #10 ND then #5 Utah really look that much better than beating #25 Cal then #10 ND then #5 Utah? Why? Because you need to compare it to the other Power 5 team we're trying to jump, and in that case either:

1) We're obviously not going to jump them (they're undefeated or a 1-loss SEC team).
2) We're obviously going to jump them (1-loss teams from any conference besides the SEC).

So, again, we're talking about something that only can possibly matter ~13% of the time (according to FPI, as good as anything for estimating this), and even then is so marginal as to provide no real of gain, according to the evidence we have.


Given the fact that there's an 80-something percent chance that none of this matters and that no matter what we want to make the CCG for a shot at the Rose Bowl, we want to root for the south over the north. Outside of that I think we want to root against upsets, insofar as that helps SOS on the margin, but no matter what we absolutely want to root for the south.

(I know that it doesn't actually matter, since who we root for doesn't affect anything, but I think if we're going to be cheering for a team besides Stanford we should do it for good reasons.)

[stupac2]

I dont know if I completely agree, because there is a psychological aspect to college ratings that you dont have in the NFL. I think we get more "credit" for beating a 6-0 and an 0-6 team, then for beating two 3-3 teams. People like to see "quality wins". And unrelated to the Low APR U/UCLA game, I think we benefit  when a Pac12 team we play beats one that we dont play that year, because I assume it makes our SOS stronger. So while I agree in general with what you say, that we always want the south to beat the north because it helps us win the division,  I dont think it's black and white when you take into account selection for the Playoffs.

Okay sure, say we win out. ESPN predicts UCLA and Cal to both be 8-4 (rounding UCLA up and Cal down, which I think is fair since we're assuming we beat Cal). How much more impressive will it be for Cal to be 9-3 and UCLA 7-5? Or the other way around? Does it really matter? Either way we've beat one team that has a pretty good record and another that has an okay record. Who cares which team is which?

What I really don't get about this line of thought is that it's all so marginal. If we don't win out the records of the teams we beat are completely and utterly meaningless (I don't think anyone disagrees with this). If we do win out, we'll have wins over a highly ranked ND, a highly ranked South champion (whoever it is will be ranked), a probably-back-to-ranked-by-then Oregon, and presumably some other teams we've beaten will be ranked, though it's hard to say who will be. Will beating #15 Cal then #10 ND then #5 Utah really look that much better than beating #25 Cal then #10 ND then #5 Utah? Why? Because you need to compare it to the other Power 5 team we're trying to jump, and in that case either:

1) We're obviously not going to jump them (they're undefeated or a 1-loss SEC team).
2) We're obviously going to jump them (1-loss teams from any conference besides the SEC).

So, again, we're talking about something that only can possibly matter ~13% of the time (according to FPI, as good as anything for estimating this), and even then is so marginal as to provide no real of gain, according to the evidence we have.


Given the fact that there's an 80-something percent chance that none of this matters and that no matter what we want to make the CCG for a shot at the Rose Bowl, we want to root for the south over the north. Outside of that I think we want to root against upsets, insofar as that helps SOS on the margin, but no matter what we absolutely want to root for the south.

(I know that it doesn't actually matter, since who we root for doesn't affect anything, but I think if we're going to be cheering for a team besides Stanford we should do it for good reasons.)