-
Follow us on Twitter @buckeyeplanet and @bp_recruiting, like us on Facebook! Enjoy a post or article, recommend it to others! BP is only as strong as its community, and we only promote by word of mouth, so share away!
- Thread starter BB73
- Start date
CincyInterloper
Hall of Fame
The problem with head-to-head is that it is only a single sample point, which is a statistically insignificant sample. It is quite possible that if Oregon had played OSU ten times back in September, OSU would have won more than half of those games. So, if the objective is identifying the "best" teams, you have to start looking at other data to try make up for the lack of confidence inherent in small sample sizes. I don't mind the committee using statistics like margin of victory, offensive yards, etc., etc. if it is used constantly. It's basically a necessity. But the committee doesn't appear to be using any kind of consistent formula for factoring in all this stuff. They just cherry-pick statistics that support what they think the outcome ought to be.
With regard to Oregon, the best argument against them is their loss to Stanford. I mean, Stanford is bad. So losing to them makes it a lot more likely that Oregon's win over OSU was an outlier. In contrast, OSU's worst loss is to a ranked team. That ranked team happens to be Oregon, but if you compare OSU's record against a stronger schedule and their lack of bad losses, I think every handicapper in Vegas is going to favor OSU.
With regard to Oregon, the best argument against them is their loss to Stanford. I mean, Stanford is bad. So losing to them makes it a lot more likely that Oregon's win over OSU was an outlier. In contrast, OSU's worst loss is to a ranked team. That ranked team happens to be Oregon, but if you compare OSU's record against a stronger schedule and their lack of bad losses, I think every handicapper in Vegas is going to favor OSU.
Upvote
0
Zurp
I have misplaced my pants.
I agree that it is just one single data point, but I don't think it is "statistically insignificant". I think that it is very significant. Who wins and who loses should be very important. That's why the game is played. It isn't a beauty contest. It isn't a "who can put forward the better impression that they are a good team" contest. Score more points than the other team and get the win. Ohio State played Oregon and that game should count for something. Now, Oregon also played Stanford and that should count for something. So I'm not saying that Oregon should be in front of Ohio State, or that they should be behind Ohio State. But if we wanted to be in front of Oregon, we shouldn't have lost that game. As it is, it's up to the opinions of some toolbags who really don't have a reason to show any integrity. (I'm not saying that they're doing it wrong, but that the chance is there for them to do something wrong without any repercussions.)
The better team doesn't always win, but I'd prefer to go overboard on the importance of the game, rather than not enough.
Upvote
0
CincyInterloper
Hall of Fame
That is certainly a fair and objective way to do things. I think the fairest way to deal with the statistical uncertainties is to expand the field until there is something like a 95% chance the best team is included in the field. Not sure how to calculate that or how many teams that would be. I would be surprised if it took more than 8-12 teams to get to that level of confidence in most years. I would also dump the committee and go with some sort of open-source computer ranking system. I'd be just as happy dumping the CFP altogether and going back to the bowl system.
Last edited:
Upvote
0
It depends on a lot of things, including the experience of the starting QBs and to whom and when you lost (Ohio State to Oregon, second game; Oregon to Stanford, fifth game).
Upvote
0
As a fan of the beauty contest approach over an expanded playoff I would argue that it can be the better way to go about it.
Teams have lost many a game over the years by a blown call. Do the points recorded from that call make them the better team?
I watched the Bears lose a game on a last second 65 YD FG that hit the cross bar. There were many things one could look at in a game like that to evaluate who was the "better team" of more significance than whether that kick was one foot higher.
That the better team always wins is the "Big Lie" in all sports.
Upvote
0
ScriptOhio
Everybody is somebody else's weirdo.
This guy must sit around all day just making lists, he's predicting that the Juggalos get in the CFPs as #4:
Upvote
0
CincyInterloper
Hall of Fame
Some really rough numbers using Vegas Insider odds to win the CFP. I'm sure an expanded playoff would increase the odds of lower ranked teams winning the CFP, so these are kind of bullshit numbers right out of the gate. Once you get past Notre Dame, the current odds drop below 1%, so I'd say a 12 team playoff would have a pretty high probability of including the best team. Could probably drop to eight teams.
Team Odds Implied Prob. Cumulative Prob.
Georgia -120 54.55% 54.55%
Alabama 350 22.22% 76.77%
Ohio State 400 20.00% 96.77%
Oregon 2000 4.76% 101.53%
Michigan 3000 3.23% 104.76%
Cincinnati 4000 2.44% 107.19%
Oklahoma 5000 1.96% 109.16%
MSU 5000 1.96% 111.12%
OKST 8000 1.23% 112.35%
ND 10000 0.99% 113.34%
Wis 20000 0.50% 113.84%
Iowa 20000 0.50% 114.34%
Team Odds Implied Prob. Cumulative Prob.
Georgia -120 54.55% 54.55%
Alabama 350 22.22% 76.77%
Ohio State 400 20.00% 96.77%
Oregon 2000 4.76% 101.53%
Michigan 3000 3.23% 104.76%
Cincinnati 4000 2.44% 107.19%
Oklahoma 5000 1.96% 109.16%
MSU 5000 1.96% 111.12%
OKST 8000 1.23% 112.35%
ND 10000 0.99% 113.34%
Wis 20000 0.50% 113.84%
Iowa 20000 0.50% 114.34%
Upvote
0
Zurp
I have misplaced my pants.
Someone on the internet disagrees with my opinion?!?
There's no way to get everyone to agree with this kind of thing. And I respect your opinion that there's other things involved in determining the mythical "better team". Ohio State had a freshman quarterback against Oregon. But Ohio State lost. And to be fair, I'm not buying the "CJ Stroud is a freshman and that's why Ohio State lost" garbage. He threw for almost 500 yards. Maybe it should have been more, but the defense was atrocious. If you say that the defense has since changed, making Ohio State a better team, I'm on board.
8-12 teams to get a 95% confidence that the best team is in the field? Yikes. I'd say that most years, 4 teams gets me a 95% confidence.
From 1998 through 2013, was there really any year where the general college football fanbase thought that the "best team" wasn't in the top 4? What was Ohio State ranked in 1998? Top 4, from what I remember. 2003, USC was screwed, more or less, but they were #3 and would have been in the 4-team playoff. It seems most years, #1 wasn't debated much, and there was controversy over #2 and #3. But I don't remember any year where someone said, "#5 is actually the best team!!!"
2014 - there was some fun controversy, but I think Ohio State shut the Big 12 fans up.
2015 - Ohio State fans complained because they thought the game against Michigan State shouldn't count. That goes back to the beauty contest vs. on-field results debate.
2016 - Penn State fans whined. Ha. I don't think anyone feels that Penn State had a claim at "best team".
2017 - Ohio State again kept out, but they had 2 losses. Did anyone else have any claim to a "best team"?
2018 - Ohio State again kept out. Tough loss to Purdue, and a garbage win against Maryland. Tough to put any faith in the "beauty contest" argument that year. Maybe there's a case that year for Ohio State over... was that the year Alabama got in without playing the conference championship game?
2019 - I think the only controversy was between who is #1 and who is #2. No one cried about who #5 was.
2020 - I think the only controversy was that Dabo voted Ohio State #11 in a worthless poll. No crying about who's #5.
So in 28 selections, I think only 2 (2015 and 2018) have a legitimate complaint. You can have a different opinion. (Not that you need my permission.) That's 7%. So 4 teams, by that math, is giving me a 93% confidence. Pretty close to 95%.
Upvote
0
CincyInterloper
Hall of Fame
I do agree that in most years, the odds of the fifth team being the best team will be pretty low. They would be even lower if the committee based their "analysis" on things more statistically rigorous than eyeball tests and style points. The situations that would generate the most uncertainty for me is when we end the season with a ranked undefeated team that is outside the top four. That happened to UCF in 2017, and they went on to beat Auburn in the Peach Bowl.
Last edited:
Upvote
0
ORD_Buckeye
Wrong glass, Sir.
I really want UGA in the first round. I think our OL is good enough to control their front seven enough that CJ and receivers can do their thing against a good but not great secondary. I also think the UGA defense against The Ohio State offense is the matchup that 'merica and the committee desperately want, so they'll do everything possible to arrange it either as 1-4 or 2-3.
Upvote
0