CFP Discussion

[bobongo]

I understand your position. My point was I thought the teams should have been Michigan, Washington, FSU and Alabama. If UGA had beaten Alabama, i think the four teams would have been UGA, Michigan, Washington, FSU. The committee marginalized FSU because they lost their QB. If anything was going to be marginalized, i think it should have been Texas' win over Bama at the beginning of the season.

How about not marginalizing any of the games or teams? Then we can get the four teams that actually deserve to get in. Just count the season as it stands, period.
There should be no consideration of anything other than the games played and their results. Otherwise, what's the point of playing them?
Texas goes to Tuscaloosa, gets a hard fought ten-point win, and it means...nothing, because it's too early in the season? WTF?
Leaving Texas out in favor of Alabama would have been at least as unjust, if not moreso, than leaving FSU out IMO.

 

[stinger78]

It’s a problem of the NCAA being too weak. The NFL, NBA, and others don’t have this problem.

ESPN is a business. Protesting is valid. Not watching is valid. Making fun of them is valid. Suing is valid, for those that have standing. If ESPN is abusing a semi-monopoly position, then they should be forced to be fair actors. And they’ve cornered the market on college football (and the “playoffs”), so I think a smart set of lawyers could do something.

Ironically, ESPN is edging towards a monopoly, or was. The NCAA can’t dictate terms because they don’t have monopoly power.

Someone in the food industry said that there’s a lot of money to be made in agriculture if you’re not the farmer. Clayton Christensen wrote about what businesses made a lot of money in the supply chain (Microsoft) and which didn’t (hard drive manufacturers), based on where they were in the chain. ESPN is in that nice place in the chain where they dictate terms.

Clay Christiansen was one of my favorite business/leadership thinkers. His book, "The Innovator's Dilemma," is one of the best I've ever read. And yes, there needs to be disruptive innovation in college football. What we are seeing is not it - this is merely a power grab due to weak leadership and intrusive court decisions.

 

[Vespidae]

College Football Playoff - Rankings

collegefootballplayoff.com

Go to the college football playoff website rankings. Scroll past all of the team logos in the rankings, the conference logos below that. Then you will see the ESPN logo,much bigger than any logo of any team or conference. It’s literally five times bigger than any other image. That says it all right there.

This one? Looks pretty normal.

College Football Playoff - Rankings

collegefootballplayoff.com

 

[Vespidae]

How about not marginalizing any of the games or teams? Then we can get the four teams that actually deserve to get in. Just count the season as it stands, period.
There should be no consideration of anything other than the games played and their results. Otherwise, what's the point of playing them?
Texas goes to Tuscaloosa, gets a hard fought ten-point win, and it means...nothing, because it's too early in the season? WTF?
Leaving Texas out in favor of Alabama would have been at least as unjust, if not moreso, than leaving FSU out IMO.

The regular season is not a single elimination format. It's a series of conference and non-conference games that are then ranked. Alabama dropped from 3rd to 10th in one week, so it did mean something. Alabama paid the price for not figuring out its offense under Tommy Rees and what quarterback to use. It then won ten in a row.

The problem is, and always has been, the ranking system. The AP uses 63 writers and broadcasters to vote each week. A system like ultimately becomes a herd mentality ... "Lets' vote UGA No. 1 because heck, they were No. 1 last week, won and got a lot of votes." Confirmation bias in action.

It would be far better to simply get rid of the ranking system entirely ... but its hugely useful as a marketing ploy. Everyone hangs on the latest ranking ..."Did we go up or go down?" with all sorts of implications from that.

I think one thing to agree on is that the system we have is no system at all. It's a patchwork of polls, bowls, and media packages that have cobbled together over time with different stakeholders. And the shakeout will continue ...

 

[stinger78]

The regular season is not a single elimination format. It's a series of conference and non-conference games that are then ranked. Alabama dropped from 3rd to 10th in one week, so it did mean something. Alabama paid the price for not figuring out its offense under Tommy Rees and what quarterback to use. It then won ten in a row.

The problem is, and always has been, the ranking system. The AP uses 63 writers and broadcasters to vote each week. A system like ultimately becomes a herd mentality ... "Lets' vote UGA No. 1 because heck, they were No. 1 last week, won and got a lot of votes." Confirmation bias in action.

It would be far better to simply get rid of the ranking system entirely ... but its hugely useful as a marketing ploy. Everyone hangs on the latest ranking ..."Did we go up or go down?" with all sorts of implications from that.

I think one thing to agree on is that the system we have is no system at all. It's a patchwork of polls, bowls, and media packages that have cobbled together over time with different stakeholders. And the shakeout will continue ...

This is correct, iMPO. The ranking system, especially preseason polls, are an exercise in uselessness. No one can know what a team has prior to suiting up and playing. But the system, from beginning to end, is a bunch of knuckleheads engaged in confirmation bias. Now we see the reality clearly: they presume to know how teams will perform beforehand at both the beginning and the end of the season.

 

[slugboy]

Moved some of the posts to the opt-in politics forum. Knock yourselves out

 

[Richard7125]

How about not marginalizing any of the games or teams? Then we can get the four teams that actually deserve to get in. Just count the season as it stands, period.
There should be no consideration of anything other than the games played and their results. Otherwise, what's the point of playing them?
Texas goes to Tuscaloosa, gets a hard fought ten-point win, and it means...nothing, because it's too early in the season? WTF?
Leaving Texas out in favor of Alabama would have been at least as unjust, if not moreso, than leaving FSU out IMO.

There will typically be some level of discretion every year. The question is what will be important. Devil's advocate. There are 133 teams in FBS. How would you treat 13-0 Liberty or do we only care about the 64 P5 schools? Liberty is playing in one of the New Year's 6 bowls so people are acknowledging they are good, but I guess they are not perceived to be good enough. How do we know they are inferior? They won all of their games. Next year there is potential for a G5 school to make the 12 team playoff so it's kind of acknowledged that some of these schools are pretty good (and deserving), but I guess they aren't deserving right now even though they won all of their games.

Again, I think FSU got shafted, but I also have very little problem with Alabama being in the playoffs.

 

[bobongo]

There will typically be some level of discretion every year. The question is what will be important. Devil's advocate. There are 133 teams in FBS. How would you treat 13-0 Liberty...

No, Liberty's schedule is near the bottom of the FBS. SOS has to part of the calculation, because it's an intrinsic part of your record. Obviously, a win over Alabama is more important than a win over Akron or somebody. Few would argue otherwise. But not when you played, injuries, weather, referees, or any other consideration.

 

[Oldgoldandwhite]

I wonder what the results would be if FSU had been ranked #1 preseason and the rest not in the top 25? I bet they would have stayed number 1 all year with the same results.

 

[forensicbuzz]

There will typically be some level of discretion every year. The question is what will be important. Devil's advocate. There are 133 teams in FBS. How would you treat 13-0 Liberty or do we only care about the 64 P5 schools? Liberty is playing in one of the New Year's 6 bowls so people are acknowledging they are good, but I guess they are not perceived to be good enough. How do we know they are inferior? They won all of their games. Next year there is potential for a G5 school to make the 12 team playoff so it's kind of acknowledged that some of these schools are pretty good (and deserving), but I guess they aren't deserving right now even though they won all of their games.

Again, I think FSU got shafted, but I also have very little problem with Alabama being in the playoffs.

Well, now that Texas's starting QB went into the Portal, how good are they going to be with their back-up QB? Are they still "better" than FSU?

 

[stinger78]

No, Liberty's schedule is near the bottom of the FBS. SOS has to part of the calculation, because it's an intrinsic part of your record. Obviously, a win over Alabama is more important than a win over Akron or somebody. Few would argue otherwise. But not when you played, injuries, weather, referees, or any other consideration.

People have criticized FSU's SOS, I have responded that #55, while not great, is still not bad. I would submit that there is very little difference between #30 and about #75.

 

[orientalnc]

Well, now that Texas's starting QB went into the Portal, how good are they going to be with their back-up QB? Are they still "better" than FSU?

Their backup, who started games when Quinn Ewers was injured, is in the portal. Ewers is staying and will start in the Sugar Bowl game. Arch Manning, their 3rd stringer, is now the backup.

 

[roadkill]

People have criticized FSU's SOS, I have responded that #55, while not great, is still not bad. I would submit that there is very little difference between #30 and about #75.

There are multiple problems with using SoS in the rankings, but at least it’s a bit more objective than the “eye test.”

For one thing, since teams play a very limited number of games and most are confined to a conference, you have a small data set to use in computing SoS, and several methods are used. The various computer SoS rankings seldom agree with any precision.

Also, the number of games is so small that a single win can noticeably decrease your SoS. This penalizes teams like FSU who won all their games. Had they lost them all, their SoS would have been considerably stronger even though they played the same teams.

Teams with the same record can be somewhat objectively compared using SoS. But what about teams with different records? Bama may have a tougher SoS than FSU, but they lost a game. It's possible to compare the SoS just using the games a team wins, but I don’t think that is what is being used in the rankings. To put it in perspective, Arizona St. had arguably the hardest SoS in the country, but at 3-9 no one is saying they should be in the playoff, because they lost too many games. Another way of looking at it is that Bama’s SoS is tough because they played great teams like Texas. Texas is a great team because, among other things, they beat Bama. It's circular logic.

 

[forensicbuzz]

Their backup, who started games when Quinn Ewers was injured, is in the portal. Ewers is staying and will start in the Sugar Bowl game. Arch Manning, their 3rd stringer, is now the backup.

Ah. That makes more sense. The article title I saw on my feed said Texas' starting QB into the portal, leaving backup QB Archie Manning...

What you say makes more sense. However, what if the starting QB for one of the 4 teams in the CFP is a guaranteed 1st round pick and decided to sit out the post-season? That is easily foreseeable. Would they still be "better" than FSU?

 

[forensicbuzz]

People have criticized FSU's SOS, I have responded that #55, while not great, is still not bad. I would submit that there is very little difference between #30 and about #75.

SOS is highly biased by the circle jerk better known as the polls.

 

[takethepoints]

The regular season is not a single elimination format. It's a series of conference and non-conference games that are then ranked. Alabama dropped from 3rd to 10th in one week, so it did mean something. Alabama paid the price for not figuring out its offense under Tommy Rees and what quarterback to use. It then won ten in a row.

The problem is, and always has been, the ranking system. The AP uses 63 writers and broadcasters to vote each week. A system like ultimately becomes a herd mentality ... "Lets' vote UGA No. 1 because heck, they were No. 1 last week, won and got a lot of votes." Confirmation bias in action.

It would be far better to simply get rid of the ranking system entirely ... but its hugely useful as a marketing ploy. Everyone hangs on the latest ranking ..."Did we go up or go down?" with all sorts of implications from that.

I think one thing to agree on is that the system we have is no system at all. It's a patchwork of polls, bowls, and media packages that have cobbled together over time with different stakeholders. And the shakeout will continue ...

I've had considerable experience with subjective rating systems and in general they tell us only two things: who or what is underperforming badly and who or what is overperforming significantly, usually with about a 5 - 10% limit for both. The under average, average, and above average whos or whats are extremely hard to distinguish. So you make up some jive numerical scale that puts the whos and whats in a rank order and anchor it with the top and bottom 5 - 10% and try to work with it. The best scales I helped with had specific rubrics for performance in categories that were relevant to the purpose of the evaluation system. I think the biggest problem with the college football rankings is that such rubrics either don't exist or produce results that appear unrealistic. This in turn is caused by very few organizations taking this seriously. You couldn't get the coaches, for instance, to go through a required set of rankings according to established and validated rubrics. They would say, "Shoot, everybody knows who has the best team in the country." and they might very well be right.

So yes, and with a bullet, we don't have a system to do rankings correctly and we might not be able to validate one if we tried. That's what a 12, then 18 or 24 (it's coming) team playoff is such a great idea. Settle all this stuff on the field, not on-line.

 

[Root4GT]

A few I could find.





https://www.espn.com/college-footba...-college-football-2023-conference-realignment

https://www.tallahassee.com/story/s...t-conference-revenue-must-change/69943371007/

https://www.tomahawknation.com/flor...p-5-revenue-distribution-money-acc-sec-big-10

https://theathletic.com/5020559/2023/11/01/florida-state-conference-independence-acc/

In fact, the entire concept of them pushing for unequal revenue distribution earlier this year is based on the shaky/unprovable premise that "our BRAND is worth so much more than the rest of the schools...so we should get more money." This was constantly postured, over and over, this fall. Since then, after extorting the policy change from the other members, they initially backed off the rhetoric...then continued on as the season wore on. The first tweet I posted is a case in point.

That Brand got them left out of the CFP for 2 different 1 loss teams! Good luck finding a conference FSU!

 

[RonJohn]

There are multiple problems with using SoS in the rankings, but at least it’s a bit more objective than the “eye test.”

For one thing, since teams play a very limited number of games and most are confined to a conference, you have a small data set to use in computing SoS, and several methods are used. The various computer SoS rankings seldom agree with any precision.

Also, the number of games is so small that a single win can noticeably decrease your SoS. This penalizes teams like FSU who won all their games. Had they lost them all, their SoS would have been considerably stronger even though they played the same teams.

Teams with the same record can be somewhat objectively compared using SoS. But what about teams with different records? Bama may have a tougher SoS than FSU, but they lost a game. It's possible to compare the SoS just using the games a team wins, but I don’t think that is what is being used in the rankings. To put it in perspective, Arizona St. had arguably the hardest SoS in the country, but at 3-9 no one is saying they should be in the playoff, because they lost too many games. Another way of looking at it is that Bama’s SoS is tough because they played great teams like Texas. Texas is a great team because, among other things, they beat Bama. It's circular logic.

The big problem with ANY metrics or statistics in sports is that they are never used to actually analyze. In science, you try to gather data and decide what the result it. In sports, people decide what the result is and then cherry pick metrics to fit that decision.

Of course it is circular logic. Alabama had a 'killer SEC week-in-week-out' schedule. The SEC gets credit for knock-down-drag-outs every single week. Except this year, the SEC was under .500 OOC against P5s. FSU played two SEC teams and won. FSU played three ACC teams who beat SEC teams. It is circular logic, but the circle only goes in one direction.

 

[Root4GT]

This is correct, iMPO. The ranking system, especially preseason polls, are an exercise in uselessness. No one can know what a team has prior to suiting up and playing. But the system, from beginning to end, is a bunch of knuckleheads engaged in confirmation bias. Now we see the reality clearly: they presume to know how teams will perform beforehand at both the beginning and the end of the season.

Preseason polls are to generate interest and viewers. They do that well. They will always be here. As an aside this season's preseason polls were fairly accurate. Michigan and Alabama were in the top 4 and Washington and Texas were 12 and 10 respectively.

 

[stinger78]

You made two points:
1. Polls should not be to create interest. That’s a byproduct. They should be to generate a true picture. That cannot be done a priori.
2. Accurate for a couple of teams doesn’t make them accurate. You will not win any wager espousing the accuracy of preseason polls. (Of course, the devil is in the standard of accuracy).