(12-17-2020 04:03 PM)Attackcoog Wrote:  

(12-17-2020 03:09 PM)Captain Bearcat Wrote:  I said this about a decade ago when they introduced the committee: a committee of a few powerful people will be way more biased than 2 polls with hundreds of voters that are supplemented by a computer system.

I said the same. Hated the the idea. That said---there is an easy fix. Instead of a 13 member committee---knock it down to 10. Each conference appoints one member to that committee. Done.

I would agree with that. The current committee is way too loaded with people who have strong P5 ties. Very unbalanced.

I say go back to the BCS formula and do a 5-1-2 Playoff. The BCS formula just had to determine:

1. Which of the 5 G5 champs is the best

2. Outside of the 6 AQ participants who qualified by winning their conference, who the 2 best at large schools are.

3. How to seed the 8 participants

Seems pretty clean, easy, and simple to me.

How about a 5-team playoff. P5 champs only, bottom 2 ranked P5 champs play a play-in game on Christmas day.

(12-17-2020 04:49 PM)ChrisLords Wrote:  How about a 5-team playoff. P5 champs only, bottom 2 ranked P5 champs play a play-in game on Christmas day.

We need to be fair and give everyone a shot.

The best way to do this is a 32 team P5 only playoff. If you're in a P5 conference and can't make that you truly don't deserve to be in.

(12-17-2020 03:57 PM)bullet Wrote:  

(12-17-2020 03:25 PM)Wedge Wrote:  

(12-17-2020 03:09 PM)Captain Bearcat Wrote:  I said this about a decade ago when they introduced the committee: a committee of a few powerful people will be way more biased than 2 polls with hundreds of voters that are supplemented by a computer system.

It's not more biased, it's just a different bias. It was never unbiased or anywhere close to it. With the "coaches" poll, for example, it's anyone's guess which staff member or relative is filling out any particular coach's ballot.

This is just a matter of preferring that bias instead of this bias.

And, there's not much reason to think that a return to BCS rankings would do what Aresco's employers want, i.e., G5 teams in the playoff. It was easy for voters to rank "non BCS" teams #3, 4, or 5 when there were only 2 teams in the fake playoff. There was no impact in doing so. It was like a college giving out an honorary degree instead of a real one. If there had been 4 teams in the fake playoff during the BCS years, they would have just put the "non BCS" darling of the season at #5 or 6, max.

A real playoff would have far more teams, like the 24 in the FCS playoff. With 4 teams, no matter what ranking you use to pick those 4, it's a popularity contest pretending to be a playoff.

I thought the committee would be better, but I have been disappointed. The committee is more biased towards the name on the uniform and more impacted by the recency effect than the polls and certainly more so than computers.

It also seems to be more dominated by powerful personalities. That isn't uncommon in groups. There's a whole series of simulations about group dynamics (lost on an icy mountain, lost on the moon, etc.). What seems to happen is that the coaches have an outsized influence. They even talked about how closely they listened to Tom Osborne and Barry Alvarez. So there was a bias towards run oriented teams.

Since the start of the CFP, the committee's rankings have been pretty close to identical to the AP and Coaches' Polls, which usually come out before the committee announces their ranking. So, if the committee's decisions reflect the biases of a few dominant personalities, how do those personalities get into the heads of the 120+ poll voters? Do they use text messages, emails, or what?

I really don't think anybody wants to eliminate bias in the selection process. They just want the bias that is inevitable to be in favor of their team or conference.

(12-17-2020 09:50 PM)ken d Wrote:  

(12-17-2020 03:57 PM)bullet Wrote:  

(12-17-2020 03:25 PM)Wedge Wrote:  

(12-17-2020 03:09 PM)Captain Bearcat Wrote:  I said this about a decade ago when they introduced the committee: a committee of a few powerful people will be way more biased than 2 polls with hundreds of voters that are supplemented by a computer system.

It's not more biased, it's just a different bias. It was never unbiased or anywhere close to it. With the "coaches" poll, for example, it's anyone's guess which staff member or relative is filling out any particular coach's ballot.

This is just a matter of preferring that bias instead of this bias.

And, there's not much reason to think that a return to BCS rankings would do what Aresco's employers want, i.e., G5 teams in the playoff. It was easy for voters to rank "non BCS" teams #3, 4, or 5 when there were only 2 teams in the fake playoff. There was no impact in doing so. It was like a college giving out an honorary degree instead of a real one. If there had been 4 teams in the fake playoff during the BCS years, they would have just put the "non BCS" darling of the season at #5 or 6, max.

A real playoff would have far more teams, like the 24 in the FCS playoff. With 4 teams, no matter what ranking you use to pick those 4, it's a popularity contest pretending to be a playoff.

I thought the committee would be better, but I have been disappointed. The committee is more biased towards the name on the uniform and more impacted by the recency effect than the polls and certainly more so than computers.

It also seems to be more dominated by powerful personalities. That isn't uncommon in groups. There's a whole series of simulations about group dynamics (lost on an icy mountain, lost on the moon, etc.). What seems to happen is that the coaches have an outsized influence. They even talked about how closely they listened to Tom Osborne and Barry Alvarez. So there was a bias towards run oriented teams.

Since the start of the CFP, the committee's rankings have been pretty close to identical to the AP and Coaches' Polls, which usually come out before the committee announces their ranking. So, if the committee's decisions reflect the biases of a few dominant personalities, how do those personalities get into the heads of the 120+ poll voters? Do they use text messages, emails, or what?

I really don't think anybody wants to eliminate bias in the selection process. They just want the bias that is inevitable to be in favor of their team or conference.

Yes, if anything the polls influence the CFP, not vice-versa.

Also, while I am a big fan of computers, the bottom line is 20 years ago we did have a BCS formula that was heavily weighted towards the computers, but that was changed, and that's because in 2001 and 2003 the computer-driven formula produced what people regarded as outrageous results, namely two teams that got gob-smacked in their CCGs nevertheless getting in to the BCS title game.

To humans, that was absurd, but all the computers saw was a set of numbers. Computers can produce results that do not pass the giggle test to people and so despite protestations, people ultimately want people to have the final control.

The Massey Composite says that LSU is the 2011 national champ, despite being beaten in the BCS title game. It says Alabama is the 2016 national champ, despite losing the CFP title game.

(12-17-2020 04:06 PM)jdgaucho Wrote:  Maybe if Aresco drops that p6 schtick, it will happen.

Aresco is intolerable.

(12-18-2020 12:15 AM)Scoochpooch1 Wrote:  

(12-17-2020 04:06 PM)jdgaucho Wrote:  Maybe if Aresco drops that p6 schtick, it will happen.

Aresco is intolerable.

Worth every Penny

(12-17-2020 04:49 PM)ChrisLords Wrote:  How about a 5-team playoff. P5 champs only, bottom 2 ranked P5 champs play a play-in game on Christmas day.

Better than other proposals here. But how about breaking up the Big 12 and just having 4 conferences? Then technically the CFP can go to 8 with the 4 conf champ games (restructured of course).

(12-17-2020 03:50 PM)domer1978 Wrote:  Go to 6 team playoff with 5 auto bids to P5 and one at large which goes to highest rated G5 or independent. :)

I know where you are coming from but under this format, the lower level P5 teams could just join G5 conferences and dominate.

(12-18-2020 12:18 AM)Scoochpooch1 Wrote:  

(12-17-2020 03:50 PM)domer1978 Wrote:  Go to 6 team playoff with 5 auto bids to P5 and one at large which goes to highest rated G5 or independent. :)

I know where you are coming from but under this format, the lower level P5 teams could just join G5 conferences and dominate.

Vandy, Rutgers or Boston College dominate?


Sent from my iPhone using Tapatalk

(12-17-2020 09:50 PM)ken d Wrote:  

(12-17-2020 03:57 PM)bullet Wrote:  

(12-17-2020 03:25 PM)Wedge Wrote:  

(12-17-2020 03:09 PM)Captain Bearcat Wrote:  I said this about a decade ago when they introduced the committee: a committee of a few powerful people will be way more biased than 2 polls with hundreds of voters that are supplemented by a computer system.

It's not more biased, it's just a different bias. It was never unbiased or anywhere close to it. With the "coaches" poll, for example, it's anyone's guess which staff member or relative is filling out any particular coach's ballot.

This is just a matter of preferring that bias instead of this bias.

And, there's not much reason to think that a return to BCS rankings would do what Aresco's employers want, i.e., G5 teams in the playoff. It was easy for voters to rank "non BCS" teams #3, 4, or 5 when there were only 2 teams in the fake playoff. There was no impact in doing so. It was like a college giving out an honorary degree instead of a real one. If there had been 4 teams in the fake playoff during the BCS years, they would have just put the "non BCS" darling of the season at #5 or 6, max.

A real playoff would have far more teams, like the 24 in the FCS playoff. With 4 teams, no matter what ranking you use to pick those 4, it's a popularity contest pretending to be a playoff.

I thought the committee would be better, but I have been disappointed. The committee is more biased towards the name on the uniform and more impacted by the recency effect than the polls and certainly more so than computers.

It also seems to be more dominated by powerful personalities. That isn't uncommon in groups. There's a whole series of simulations about group dynamics (lost on an icy mountain, lost on the moon, etc.). What seems to happen is that the coaches have an outsized influence. They even talked about how closely they listened to Tom Osborne and Barry Alvarez. So there was a bias towards run oriented teams.

Since the start of the CFP, the committee's rankings have been pretty close to identical to the AP and Coaches' Polls, which usually come out before the committee announces their ranking. So, if the committee's decisions reflect the biases of a few dominant personalities, how do those personalities get into the heads of the 120+ poll voters? Do they use text messages, emails, or what?

I really don't think anybody wants to eliminate bias in the selection process. They just want the bias that is inevitable to be in favor of their team or conference.

It seems like the polls start to converge on the committee.

(12-17-2020 04:00 PM)bullet Wrote:  

(12-17-2020 03:52 PM)AllTideUp Wrote:  

(12-17-2020 03:45 PM)Wedge Wrote:  

(12-17-2020 03:37 PM)quo vadis Wrote:  

(12-17-2020 03:25 PM)Wedge Wrote:  It's not more biased, it's just a different bias. It was never unbiased or anywhere close to it. With the "coaches" poll, for example, it's anyone's guess which staff member or relative is filling out any particular coach's ballot.

This is just a matter of preferring that bias instead of this bias.

And, there's not much reason to think that a return to BCS rankings would do what Aresco's employers want, i.e., G5 teams in the playoff. It was easy for voters to rank "non BCS" teams #3, 4, or 5 when there were only 2 teams in the fake playoff. There was no impact in doing so. It was like a college giving out an honorary degree instead of a real one. If there had been 4 teams in the fake playoff during the BCS years, they would have just put the "non BCS" darling of the season at #5 or 6, max.

A real playoff would have far more teams, like the 24 in the FCS playoff. With 4 teams, no matter what ranking you use to pick those 4, it's a popularity contest pretending to be a playoff.

In fairness to the BCS poll voters and the CFP committee voters, the computers would have put basically the same teams in the playoffs as the CFP has, and to the extent there have been some disagreements, it has always been one P5 in instead of another, no cases of a G5 making it. In 2010, TCU, while a member of a non-AQ, did finish 3rd in the BCS computers so would have made a four-team playoff determined solely by computers.

But in the CFP era, the computers have never judged a G5 team to be among the four best.

It was easy to rank TCU #3 when #3 wasn't in the playoff. They would have been in a BCS bowl game but not in the BCS title game whether they were #3 or #7. If it had been a 4 team deal instead of just 2 that year, they would have found reasons to put TCU at #5 or 6.

I think you're correct.

I think the issue with the current playoff system is that none of the other games really matter. If you don't make the CFP then there's no real reward.

All bowl games outside of the national title race have always been exhibition games, but in the old days, at least you had more schools vying in games that had an impact.

Aresco wants Cincinnati or another G5 in the CFP and that won't happen with only 4 schools, but I wouldn't be opposed to a system that involved more schools in meaningful games.

I think the ratings of the NY6 outside the two playoff bowls has demonstrated that interest in the other bowls is declining. Only the Rose Bowl holds its own. You've also got a number of players opting out even for NY6 bowls that aren't playoff bowls. With a 5-1-2 the system isn't as big a deal as with the current 4 team "playoff" or the BCS 2 team game.

8 is better than 4 which is better than 2. Still not enough but clearly it's the only possible step that is not completely impossible to get the powers-that-be to sign onto. Whether or not any of us likes it, when the PTBs discuss the format of the next round of the CFP, the only real debate among them will be 4 or 8 teams.

(12-18-2020 12:31 AM)bullet Wrote:  

(12-17-2020 09:50 PM)ken d Wrote:  

(12-17-2020 03:57 PM)bullet Wrote:  

(12-17-2020 03:25 PM)Wedge Wrote:  

(12-17-2020 03:09 PM)Captain Bearcat Wrote:  I said this about a decade ago when they introduced the committee: a committee of a few powerful people will be way more biased than 2 polls with hundreds of voters that are supplemented by a computer system.

It's not more biased, it's just a different bias. It was never unbiased or anywhere close to it. With the "coaches" poll, for example, it's anyone's guess which staff member or relative is filling out any particular coach's ballot.

This is just a matter of preferring that bias instead of this bias.

And, there's not much reason to think that a return to BCS rankings would do what Aresco's employers want, i.e., G5 teams in the playoff. It was easy for voters to rank "non BCS" teams #3, 4, or 5 when there were only 2 teams in the fake playoff. There was no impact in doing so. It was like a college giving out an honorary degree instead of a real one. If there had been 4 teams in the fake playoff during the BCS years, they would have just put the "non BCS" darling of the season at #5 or 6, max.

A real playoff would have far more teams, like the 24 in the FCS playoff. With 4 teams, no matter what ranking you use to pick those 4, it's a popularity contest pretending to be a playoff.

I thought the committee would be better, but I have been disappointed. The committee is more biased towards the name on the uniform and more impacted by the recency effect than the polls and certainly more so than computers.

It also seems to be more dominated by powerful personalities. That isn't uncommon in groups. There's a whole series of simulations about group dynamics (lost on an icy mountain, lost on the moon, etc.). What seems to happen is that the coaches have an outsized influence. They even talked about how closely they listened to Tom Osborne and Barry Alvarez. So there was a bias towards run oriented teams.

Since the start of the CFP, the committee's rankings have been pretty close to identical to the AP and Coaches' Polls, which usually come out before the committee announces their ranking. So, if the committee's decisions reflect the biases of a few dominant personalities, how do those personalities get into the heads of the 120+ poll voters? Do they use text messages, emails, or what?

I really don't think anybody wants to eliminate bias in the selection process. They just want the bias that is inevitable to be in favor of their team or conference.

It seems like the polls start to converge on the committee.

Exactly. They didnt the first year. If you will remember, Marshall was undefeated and was showing up in the top human polls at #25 by week #7. Undefeated going into the very first week of the CFP Rankings---Marshall was not even ranked. After that, it appeared that many poll voters were more into trying to reflect what they thought the Committee would do on the following Tuesday (since thats the only poll that matters anymore) than really ranking the schools as they think they should be rated. Marshall kept winning, but would also not be included in any of the next three weekly CFP rankings. During that time they began to stop rising (peaked at #18)--and then began to slip back in the human polls to #19. Marshall finally showed up in the 5th week of the CFP poll of the season at #24 (week 13)---the same week they slipped to #19 in the human polls. Thats a good example of how the human polls began to track the CFP once they got a feel for how the CFP Committee operated.

Every year, the polls are sorted out and create an order and then the CFP rankings start. Once the CFP rankings start---the human polls start to fall in line with the Committee rankings. I remember that behavior was a point of speculation early on (because I absolutely hated the idea of a Selection Committee--especially after I saw its composition). I remember there being a question as to whether the polls would influence the Committee or would the Committee influence the poll voters. Well---it didnt take long to get our answer. We had a real good idea that the Committee would be the one doing the influencing by the end of year one.

What should happen: NCAA sanctioned 16-team playoff with 10 auto-bids and 6 at-large bids.

What could happen: current playoff expands to 8 with 5 P5 auto-bids and 3 at-larges (possible G5 auto-bid). I like the idea of giving the top 5 conference champs the auto-bid, P5 or G5, and then 3 at-larges.

Not likely: 12-team playoff with 10 auto-bids and 2 at-large bids.

It’s going to take a much expanded playoff with P5 and some G5 AQ component to get me back in at this point, but, even if that were to happen, there’s an attendance issue waiting to happen depending on when they start the playoffs. I will never forget the atrocious attendance issues Villanova had when it won the FCS title over a decade ago. And I think you will see that no matter the venue or press hype if you’re looking to play a first round toward finals and the winter break of schools. You can love college sports something fierce, but, it takes a different type to want to stay around a school activity when there’s a stressor like finals or a release like winter break.

Run it later, and good luck going up against the pro playoffs and other pro sports. Granted, January and February suck for sports anyway, outside the NY bowls and Super Bowl.

I don’t think the 4 teams hosting a FBS playoff game in December will need to worry about attendance issues. I would set it where the losers of the games have their bowl landing spot scheduled as well.

(12-18-2020 08:43 AM)esayem Wrote:  I don’t think the 4 teams hosting a FBS playoff game in December will need to worry about attendance issues. I would set it where the losers of the games have their bowl landing spot scheduled as well.

lol. teams that lose in a playoff game in December won't be bowling. the players would balk at that 100%.

(12-18-2020 08:47 AM)stever20 Wrote:  

(12-18-2020 08:43 AM)esayem Wrote:  I don’t think the 4 teams hosting a FBS playoff game in December will need to worry about attendance issues. I would set it where the losers of the games have their bowl landing spot scheduled as well.

lol. teams that lose in a playoff game in December won't be bowling. the players would balk at that 100%.

Right. They’ll play the QF on NYD (Orange, Sugar, Cotton, Rose) and play the SF Sat Jan 8-14 (Peach, Fiesta). You don’t want to lose fan bases before bowl season begins and the NFL’s schedule expansion opens up that second Saturday in January.

(12-18-2020 01:12 AM)Attackcoog Wrote:  Every year, the polls are sorted out and create an order and then the CFP rankings start. Once the CFP rankings start---the human polls start to fall in line with the Committee rankings.

It's amazing that this idea gets repeated to so often. There's no evidence for it, and the scant evidence of influence suggests the polls influence the CFP - which make sense, since they precede the CFP.