Northern Illinois and Fresno State

[Spence]

The problem is that cfb is mostly regional and lots of people want it to be national. Scheduling is mostly regional. It is impossible to compare teams and be right.

[Duke1632]

There is a downside if you only have two teams that make it. That is the problem with a team that plays a weak schedule. It doesn't even have to be a team from a non AQ conference. Ohio State's schedule is weak. I try to compare them to the other top teams, but it is hard. The truth is that if Ohio State played in the MAC and played the same schedule they have played this year, they would be ranked like NIu is now. The reason they are rank 4 is they are a blue blood program. I hate to admit that, but it is true. Ohio State could be good, but they could be a product of their schedule too. I think we will find out a lot about the Buckeye Offense if/when they play Michigan State in the B-10 championship game. I don't think we will find out about the Buckeye D until the bowl game. That is what scares me.

But how is that a downside? It seems to be evidence for the point I made, not against it. According to CCR, there's about the same difference between Baylor's schedule (#36) and FSU's schedule (#78) as there is between FSU's and NIU's. No one is suggesting that FSU ought to be left out because of their relatively soft schedule. In fact, quite a number of people seem to think FSU is the best team in the country, despite their relatively soft schedule this year. Why? because they are undefeated, and they have looked great winning given their schedule. Things that are also true (to some extent at least) of NIU.

[billybud]

CCR is way off...

Who has Baylor beaten?

FSU wiped out two teams that were top 10 and are still Top 15 in BCS.

Baylor has played really nobody other than their 10 point win over unranked KSU.

There is a reason that the computers have FSU #1.....

[Spence]

Because you want to get the two best teams. It doesn't matter who they are, you want a fair system to determine it, if you are playing for a national championship. I think CFB should go back to the old bowls and there should be only regional championships and exhibition bowls. It has gotten out of hand and it is no more fair today than when it started.

[Duke1632]

The problem is that cfb is mostly regional and lots of people want it to be national. Scheduling is mostly regional. It is impossible to compare teams and be right.

Agreed. Hence, a rational actor will select the best tool available for making these comparisons. We happen to be on the board that appears to be the best tool (I looked around too before joining). Hence, teams like Baylor (and, yes, your Buckeyes) are very legit according to the best tool we have for making this impossible comparison. It's not perfect, of course, and we are human with our own perceptions and biases (and reverse biases), but it clearly means something significant--about as significant as anything else we can come up with. Hence, I for one have no problem accepting Fresno St. and NIU are both in the top 15 right now, where CCR puts them.

[Duke1632]

Because you want to get the two best teams.

But as you yourself pointed out above, that's an impossible task. No one knows or could possibly know who the two best teams are. Besides, reality probably wouldn't let you do that anyway. For example, let's say, somehow, we know that bama and FSU are the two best teams. But even the best team can lose on occasion. So what if both of them drop their last game of the season. They are both still the best team, but would you still advocate leaving out Oregon and OSU (both unbeaten) in that case?

Ultimately, no one can know who the best teams are or what the rankings ought to be, therefore, that which demonstrates predictive success for what actually happens on the field (e.g., an accurate model of reality) is miles above the subjective belief of anyone, including you, including me. A team that loses in November almost definitely should not be in the NCG, even if everyone thinks they are the best team in the world and even if they have the toughest schedule of all. Objectively, that team did not earn it that year. That's one reason why W/L is and must be much more valuable than SoS.

[Duke1632]

CCR is way off...

Who has Baylor beaten?

FSU wiped out two teams that were top 10 and are still Top 15 in BCS.

Baylor has played really nobody other than their 10 point win over unranked KSU.

There is a reason that the computers have FSU #1.....

Both schedules have 12 teams on them, including games not yet played. That's the SoS. Baylor has 4 very tough games coming up, and if they lose, they will (rightly) be dropped in the rankings, but it won't affect their SoS much. Same thing goes for FSU, which has 10 other teams on their schedule beyond the two impressive wins they've already garnered that you mention. In the aggregate, I do not see any reason to doubt the values CCR has right now. By the end of the season, Baylor will have finished a tougher schedule than FSU, but that does not in any way mean Baylor is a better team. Their W/L record at the end of the season will speak more of that than anything else. Same applies to NIU, et al.

[Spence]

Because you want to get the two best teams.

But as you yourself pointed out above, that's an impossible task. No one knows or could possibly know who the two best teams are. Besides, reality probably wouldn't let you do that anyway. For example, let's say, somehow, we know that bama and FSU are the two best teams. But even the best team can lose on occasion. So what if both of them drop their last game of the season. They are both still the best team, but would you still advocate leaving out Oregon and OSU (both unbeaten) in that case?

Ultimately, no one can know who the best teams are or what the rankings ought to be, therefore, that which demonstrates predictive success for what actually happens on the field (e.g., an accurate model of reality) is miles above the subjective belief of anyone, including you, including me. A team that loses in November almost definitely should not be in the NCG, even if everyone thinks they are the best team in the world and even if they have the toughest schedule of all. Objectively, that team did not earn it that year. That's one reason why W/L is and must be much more valuable than SoS.

I agree that it is impossible to know for sure, but that doesn't mean you shouldn't try. But I would rather have the old bowl system of exhibition games. If we have to have a playoff, then I think it should be a tournament of champions and take the winner of every conference regardless of strength. I think that would at least be fair. If you want to be the best, win your conference and prove it. That wouldn't guarantee the two best teams, but at least the way you got to the tournament ould be fair and what you do from there is up to you.

[Duke1632]

I agree that it is impossible to know for sure, but that doesn't mean you shouldn't try. But I would rather have the old bowl system of exhibition games. If we have to have a playoff, then I think it should be a tournament of champions and take the winner of every conference regardless of strength. I think that would at least be fair. If you want to be the best, win your conference and prove it. That wouldn't guarantee the two best teams, but at least the way you got to the tournament ould be fair and what you do from there is up to you.

- old bowl system of exhibition games: agreed (resoundingly)
- If playoff, then conference champions: agreed (resoundingly)
- cannot know, but should try to determine the best: agreed (resoundingly), and it is noted that CCR represents right now the best we can do for the "try" part.

If you want to be the best, win your conference and prove it.

Totally agree again. Remember, it's all about earning it on the field, because that's how we actually define "being the best." If you do not win your conference in a given year, you did not earn it that year and that's the end of it, even if by some objective metric you are the best team in the country. To consider any other option is tantamount to suggesting that a team could lose the NCG and still be considered #1. Yeah, they lost, but they are clearly the best team (b/c of SoS, or _____ insert any argument), so they are national champs. Silly, and the fact that it almost happened in 2011 season is evidence of BCS failure. The better team does not always win, but the team that wins has EARNED it. That is, in fact, the core of why any game is played at all, and the very reason why NIU should be given at least a marginal amount of street cred.

[Spence]

I think we mostly agree. As do most here, even Billybud and Donovan to the main point about being fair about the process. We just disagree sometimes about the process. The problem I have with a weaker schedule - and I don't believe Florida State's schedule is weaker than Ohio State - is that you really don't know. Especially when the strength of a team is their offense. A good defense can and will most always beat a good offense. Good offenses can score lots of points on bad teams. Good defenses make squeak by every game. The good defense can keep it's team in the game if the offense is average. A good offense that can't play good defense will get exposed if their offense is slowed down. It is hard to even tell if a team has a great offense if the teams it plays do not have good defenses. That is the problem in the B10 right now. There is only one good defense in the whole conference - Michigan State. We won't know until that game if Ohio State is more of a complete team than we thought they were. I is my hope that they do, but I have so little to compare at this point, I don't know how good or average they are this year.

[Spence]

I did have a fair way of comparing teams worked out a couple years ago that would require national scheduling and Notre Dame to join a conference, but it would require a computer to schedule teams and it would require the big boys to give up a home game or two - meaning they would never go for it.

[Eric]

NIU beats up on nobodies and is undefeated and then when they ran into a Florida State (in the bowl last year)...oops.

NIU's best win was by three points over four loss (and counting) Iowa.

NIU has lost just one game in the last 20 (to FSU) but has played nobody at all in the last 20 except 4-9 Iowa last year and a bad Iowa this year.

NIU will finish this season having not beaten a BCS team with a winning record...as they did last year.

Why are they considered good?

Lynch carried 23 times for 44 yards against FSU and was 15 of 41 passing. Yet they were all world against their bum schedule.

It is all schedule.

Florida State was only beating that head-coach-less team 17-10 after three quarters with an NFL QB at the helm.

I don't think anybody is pretending that NIU is an elite team. The polls aren't entirely a power poll and everybody knows all wins are not created equally. In 2006, Boise State's best victory was over Oregon State who wound up losing 3 conference games in the Pac-12 that year; a good, but not great team. Would you have denied them an opportunity to play in a spotlight game? Keep in mind nobody is saying they should have played for the national title in 2006 and nobody ever suggested NIU or Fresno should play for one this year. All they're asking for is a shot on primetime television against another team who had a great season. Non-BCSers are 4-2 against BCSers in these games. If NIU makes it through, I say give them a shot. If they get Baylor though, they might as well stay in DeKalb for the Winter . Texas or Oklahoma or maybe Oklahoma State? That might be a worthwhile game. Exact same goes for Fresno.

Now last year, even I argued against NIU's conclusion because they did lose to Iowa in the first game of the season. They probably should have won that game because they gave up 9 points late and self-destructed. Lynch was still getting ready for his role, too. But that game alone is one they needed to have if they wanted to have any argument, yet I was surprised that the pollsters gave them enough respect to slide in (part of that happening because a string of 7 teams in front of them lost on the same weekend). That Orange Bowl should not have happened in my opinion and the fact that Kent State would have gotten there with a bad loss to Kentucky on their resume would have been 100x worse.

[donovan]

One of the unknown factors is how the owners of these sites determine SOS. It is all over the place and most of then seem to hold those card close to their vest.
Most say that information is proprietary. I think SOS is very subjective in the making and then we treat it as objective data.

[billybud]

NIU beats up on nobodies and is undefeated and then when they ran into a Florida State (in the bowl last year)...oops.

NIU's best win was by three points over four loss (and counting) Iowa.

NIU has lost just one game in the last 20 (to FSU) but has played nobody at all in the last 20 except 4-9 Iowa last year and a bad Iowa this year.

NIU will finish this season having not beaten a BCS team with a winning record...as they did last year.

Why are they considered good?

Lynch carried 23 times for 44 yards against FSU and was 15 of 41 passing. Yet they were all world against their bum schedule.

It is all schedule.

Florida State was only beating that head-coach-less team 17-10 after three quarters with an NFL QB at the helm.

I don't think anybody is pretending that NIU is an elite team. The polls aren't entirely a power poll and everybody knows all wins are not created equally. In 2006, Boise State's best victory was over Oregon State who wound up losing 3 conference games in the Pac-12 that year; a good, but not great team. Would you have denied them an opportunity to play in a spotlight game? Keep in mind nobody is saying they should have played for the national title in 2006 and nobody ever suggested NIU or Fresno should play for one this year. All they're asking for is a shot on primetime television against another team who had a great season. Non-BCSers are 4-2 against BCSers in these games. If NIU makes it through, I say give them a shot. If they get Baylor though, they might as well stay in DeKalb for the Winter . Texas or Oklahoma or maybe Oklahoma State? That might be a worthwhile game. Exact same goes for Fresno.

Now last year, even I argued against NIU's conclusion because they did lose to Iowa in the first game of the season. They probably should have won that game because they gave up 9 points late and self-destructed. Lynch was still getting ready for his role, too. But that game alone is one they needed to have if they wanted to have any argument, yet I was surprised that the pollsters gave them enough respect to slide in (part of that happening because a string of 7 teams in front of them lost on the same weekend). That Orange Bowl should not have happened in my opinion and the fact that Kent State would have gotten there with a bad loss to Kentucky on their resume would have been 100x worse.

Still...a 21 point win for FSU.....in a poorly played game for the Noles. It was an ultimate trap game....Playing a disappointing opponent in a BCS Bowl while the other team sees it as a chance to be heroes. The game was actually never in doubt and NIU's stats told the story.

[billybud]

For those of you who enjoy math and formula related comparisons...an Oregon site has compared an Oregon vs FSU match up using FEI and S&P methodology...

http://www.addictedtoquack.com/2013/10/ ... fei-and-sp