(10-29-2014 10:31 AM)stever20 Wrote:  

(10-29-2014 10:26 AM)quo vadis Wrote:  

(10-29-2014 09:57 AM)stever20 Wrote:  

(10-29-2014 09:51 AM)quo vadis Wrote:  

(10-29-2014 09:25 AM)b0ndsj0ns Wrote:  And of course you cut out the part of my post about why ND's schedule is rated higher than TCU's. Care to comment on that or just gonna ignore it?

I didn't ignore it, I read it and decided it wasn't worth commenting on, because it boiled down to you disagreeing with how SOS is calculated, and I didn't find that convincing.

ND's schedule is rated higher by just about everybody. That's pretty strong evidence that they have in fact played a tougher schedule than TCU. And that's before we consider that ND has a better loss than them or Michigan State. No way should ND be ranked behind those teams.

I think what is far more worthy of nit-picking is why is FSU ranked 8 spots ahead of Notre Dame? Outside of their game versus each other, their schedules are almost identical, and FSU "won" thanks to being bailed out by the refs despite trailing almost the whole game and being badly out-gained, out-first downed, and out- time of possessioned. Factor in that the game was at Tallahassee, worth at least 5-6 points, and objectively, ND 'won' that game. And yet FSU is 8 spots higher thanks to a single ref's call.

Go figure.

Clemson is better than anyone that ND has played.

THe problem is ND hasn't beaten ANYONE. Everyone else in the top 10 has beaten a top 20 team. The committee is clearly valuing who you have beat- and that kills ND.

Clemson sucks. They couldn't beat FSU without Jameis Winston. Stanford is better than they are.

Beyond that, though, I've explained why the committee is wrong to "value who you have beat" more than things like SOS and who you lost to, especially when the teams you beat are marginal - like FSU beating Clemson and Michigan State beating Nebraska.

Stanford has 3 losses and could very easily finish up with 5 or 6 losses(3 of their last 4 games are against ranked teams). They are not Stanford of the past few years. Clemson will finish up 9-3 or 10-2.

Like it or not but Clemson and Nebraska are top 25 teams. Stanford isn't.

If only Nebraska could play a team close to the caliber of Clemson. Can't controle conference play however.

An interesting one for me is this:

Ole Miss has played two ranked teams. They beat #6 Alabama at home and lost to #19 LSU on the road.
Baylor has played two ranked teams. They beat #7 TCU at home and lost to #20 WVU on the road.

Clearly SOS is killing Baylor here as good wins/bad losses seems to be equal for these two.

I'd say the fact that Ole Miss other wins- Boise, Memphis, Texas A&M
Baylor's other wins- Texas, Iowa St. Next best one is either Buffalo or SMU.

(10-29-2014 10:57 AM)stever20 Wrote:  I'd say the fact that Ole Miss other wins- Boise, Memphis, Texas A&M
Baylor's other wins- Texas, Iowa St. Next best one is either Buffalo or SMU.

I agree. And, to be honest, Ole Miss has looked better - or, at least, more consistent - than my Bears.

My point was not to say Baylor was better.

It just seems like, if the committee is looking at best wins/worst loss then they stack up pretty closely. Clearly, overall SOS is a big factor here and justifies- in the committee's mind- an 8 spot difference.

(10-29-2014 10:07 AM)b0ndsj0ns Wrote:  

(10-29-2014 09:59 AM)Zombiewoof Wrote:  

(10-29-2014 09:44 AM)adcorbett Wrote:  

(10-29-2014 08:32 AM)b0ndsj0ns Wrote:  

(10-29-2014 08:26 AM)quo vadis Wrote:  Since when is "who you beat?" more important than overall SOS or "who you lost to"?

That is ignorant thinking. ND should clearly be ranked ahead of Michigan State.

In my world who you beat and who you lost to always should have been the standard we are judging teams by. Hell ND's second best win at this point is UNC, the same UNC you have been giving ECU hell about for how much they suck. Now if you'd like to change your mind and say beating UNC by 7 at home is impressive then beating them by 29 at home is super impressive.

I assumed Quo was being sarcastic because it "used" to be who you beat. But in 2011, the SEC (And media) clearly changed the qualifications, and said it was all about "who you lost to, more than who you beat." Anyone remember that?

Of course his later posts indicate he actually believes that, but nonetheless that was precisely how Alabama got their rematch vs. LSU in 2011. They did not have near the quality wins Oklahoma State did, but they had a "better loss."

that said, when you have a loss like ND does, on the road at number 1/2, losing on essentially the last play of the game at the goal line... you can't count them as an unbeaten team, but it does give you room to be at the top of the one loss teams (similar to Oklahoma in 2006 years ago when they lost on a very controversial call vs. Oregon St). Note I don't think it was really a "controversial" call against Notre Dame, but I give credit for having one play at your own 2 to win the game (and yes that might possibly be because I am a Louisville fan and our second loss came the exact same way).

In a vacuum, losing to the #1 or #2 team on the last play is a quality loss. Are we to ignore the rest of the schedule? Who have they beaten that is a quality win? Stanford I will give you, then who? Do you rank them above one loss Oregon, which has top 25 wins over Michigan State and UCLA? Do you rank them ahead of TCU, which has wins over 6-2 Minnesota, 5-2 Oklahoma and 5-3 Oklahoma State? To me, Notre Dame's loss is no more impressive than Kansas State's loss to Auburn and the Irish don't have a win as good as the Wildcats's win over Oklahoma.

Make the case for Notre Dame to be higher without using a computer's SOS ranking.

Notre Dame has arguably the best loss of anyone in the country, which is why they are the highest rated 1 loss team without a top 25 win. I think that's pretty fair to them. Every 1 loss team rated above them has at least 1 win over a current top 25 team, and all have losses that are also against teams rated in the top 25. Having a great loss doesn't give you more than having a great win.

You forgot about Kansas State.

(10-29-2014 10:27 AM)quo vadis Wrote:  

(10-29-2014 09:59 AM)Zombiewoof Wrote:  To me, Notre Dame's loss is no more impressive than Kansas State's loss to Auburn

What is so funny? Notre Dame?

(10-29-2014 11:14 AM)FloridaJag Wrote:  

(10-29-2014 10:07 AM)b0ndsj0ns Wrote:  

(10-29-2014 09:59 AM)Zombiewoof Wrote:  

(10-29-2014 09:44 AM)adcorbett Wrote:  

(10-29-2014 08:32 AM)b0ndsj0ns Wrote:  In my world who you beat and who you lost to always should have been the standard we are judging teams by. Hell ND's second best win at this point is UNC, the same UNC you have been giving ECU hell about for how much they suck. Now if you'd like to change your mind and say beating UNC by 7 at home is impressive then beating them by 29 at home is super impressive.

I assumed Quo was being sarcastic because it "used" to be who you beat. But in 2011, the SEC (And media) clearly changed the qualifications, and said it was all about "who you lost to, more than who you beat." Anyone remember that?

Of course his later posts indicate he actually believes that, but nonetheless that was precisely how Alabama got their rematch vs. LSU in 2011. They did not have near the quality wins Oklahoma State did, but they had a "better loss."

that said, when you have a loss like ND does, on the road at number 1/2, losing on essentially the last play of the game at the goal line... you can't count them as an unbeaten team, but it does give you room to be at the top of the one loss teams (similar to Oklahoma in 2006 years ago when they lost on a very controversial call vs. Oregon St). Note I don't think it was really a "controversial" call against Notre Dame, but I give credit for having one play at your own 2 to win the game (and yes that might possibly be because I am a Louisville fan and our second loss came the exact same way).

In a vacuum, losing to the #1 or #2 team on the last play is a quality loss. Are we to ignore the rest of the schedule? Who have they beaten that is a quality win? Stanford I will give you, then who? Do you rank them above one loss Oregon, which has top 25 wins over Michigan State and UCLA? Do you rank them ahead of TCU, which has wins over 6-2 Minnesota, 5-2 Oklahoma and 5-3 Oklahoma State? To me, Notre Dame's loss is no more impressive than Kansas State's loss to Auburn and the Irish don't have a win as good as the Wildcats's win over Oklahoma.

Make the case for Notre Dame to be higher without using a computer's SOS ranking.

Notre Dame has arguably the best loss of anyone in the country, which is why they are the highest rated 1 loss team without a top 25 win. I think that's pretty fair to them. Every 1 loss team rated above them has at least 1 win over a current top 25 team, and all have losses that are also against teams rated in the top 25. Having a great loss doesn't give you more than having a great win.

You forgot about Kansas State.

No I didn't, they beat Oklahoma on the road and their only loss was to Auburn. So they have a top 25 win and a top 25 (top 3) loss.

(10-29-2014 09:50 AM)stever20 Wrote:  actually Alabama in 2011 DID have close to the quality wins that Oklahoma St did...
Alabama beat #6 Arkansas, #22 Penn St, and #25 Auburn
Okla St beat #8 Kan St, #14 Oklahoma, and #24 Texas

that's pretty damn close if you ask me.

Depends on how you look at it. There is a reason OSU was number 2 almost across the board in the computer rankings:

Their overall SOS was somewhat equal.


OK ST's -opponents' record was 81-63 (56.25%)
Alabama's was 74-59 (55.6%)

However if you only include wins, it looks like this:

OK ST's -opponents' record was 75-57 (56.81%)
Alabama's was 61-59 (50.83%)


OK State played nine conference games, a P5 opponent OOC, and no 1AA team. Alabama played 8 conference games, one P5 opponent, and a 1AA team.

Alabama beat 2 good teams in Penn St (9-3) and Arkansas (10-2), and a BCS ranked Auburn who was 7-5, and not ranked in any human poll. The only other team with a winning FBS record they played was LSU. They also had a weak conference schedule as their SEC opponent's in conference record was 26-38 (40.6). It drops to a very ugly 18-38 of the seven teams they beat (32.14%)

OKS St beat two good teams in Oklahoma St and Baylor, a and a BCS ranked Texas who was 7-5, and not ranked in any human poll. They also played 3 other teams with a winning record. Their opponent's Big 12 Conference record was 37-44 (45.68%), and of the teams they beat it was 34-38 (47.2%).

And finally, the final SOS rank for each:


So by any reasonable method, Oklahoma State had a better SOS, especially if you account for who you beat (Alabama did not beat LSU in the regular season). Thus the only reason they were given the spot, is because of who they lost to.

I don't see the CFP poll (with a sample of one) so far being patently unreasonable.

To me unreasonable has been when say the #5 team in the country travels to play #1 and the game is determined on the final possession with #1 winning and you see #5 fall. Or #21 hosts #3 and loses by 10 and falls out of the poll.

The AP and Coaches polls have historically punished teams for losing even when the loss was about what a reasonable person would expect the outcome to be or even better than expectation.

If CFP stays out of that trap, it will be easy for me accept their results.

(10-29-2014 09:57 AM)stever20 Wrote:  Clemson is better than anyone that ND has played.

I assume you mean that Clemson is better than anyone ND has beaten. Seeing as how ND "played" FSU, who beat Clemson already.

(10-29-2014 11:21 AM)adcorbett Wrote:  

(10-29-2014 09:50 AM)stever20 Wrote:  actually Alabama in 2011 DID have close to the quality wins that Oklahoma St did...
Alabama beat #6 Arkansas, #22 Penn St, and #25 Auburn
Okla St beat #8 Kan St, #14 Oklahoma, and #24 Texas

that's pretty damn close if you ask me.

Depends on how you look at it. There is a reason OSU was number 2 almost across the board in the computer rankings:

Their overall SOS was somewhat equal.


OK ST's -opponents' record was 81-63 (56.25%)
Alabama's was 74-59 (55.6%)

However if you only include wins, it looks like this:

OK ST's -opponents' record was 75-57 (56.81%)
Alabama's was 61-59 (50.83%)


OK State played nine conference games, a P5 opponent OOC, and no 1AA team. Alabama played 8 conference games, one P5 opponent, and a 1AA team.

Alabama beat 2 good teams in Penn St (9-3) and Arkansas (10-2), and a BCS ranked Auburn who was 7-5, and not ranked in any human poll. The only other team with a winning FBS record they played was LSU. They also had a weak conference schedule as their SEC opponent's in conference record was 26-38 (40.6). It drops to a very ugly 18-38 of the seven teams they beat (32.14%)

OKS St beat two good teams in Oklahoma St and Baylor, a and a BCS ranked Texas who was 7-5, and not ranked in any human poll. They also played 3 other teams with a winning record. Their opponent's Big 12 Conference record was 37-44 (45.68%), and of the teams they beat it was 34-38 (47.2%).

And finally, the final SOS rank for each:

Code:

Team_ NCAA_ A&H_ CM_ Sag*_ KMass_  
OKS___ 7___ 9____ 2___ 3___ 5
Ala____ 3___ 22___ 24__ 15__ (over 10)

* Sagarin ratings are post bowl, as I could only find top ten for pre-bowl rankings for Sagarin which had OK St 6, and Alabama outside of the top ten.

So by any reasonable method, Oklahoma State had a better SOS, especially if you account for who you beat (Alabama did not beat LSU in the regular season). Thus the only reason they were given the spot, is because of who they lost to.

The thing was- the wins were close enough- 3 BCS top 25 wins for both teams. 2 of the winning teams that OSU played was La-Lafayette and Tulsa.

Also you don't just look at conference records. That's crazy. Alabama's conference wins were 42-42 overall. 24-4 OOC record that you can not deny. OSU's conference wins were 55-41 overall. 21-3 OOC record.

OOC while yes Oklahoma St had 2/3 winning teams- they were La-Lafayette and Tulsa. Alabama had 9-3 Penn St better than either of those by a wide margin.

So it was pretty close on the wins and no contest on the losses.

(10-29-2014 09:03 AM)ken d Wrote:  The fact that this selection committee's poll differs slightly from the AP poll is to be expected. The fact is that if you formed a second committee of equally knowledgeable and unbiased "experts", their poll would be different from both of them, and by about the same margin.

In fact, you could put together a committee of 12 members of this forum with the same results. There is a good reason for this. It is indisputably true that there is no one "correct" ranking of football teams. That bears repeating.

THERE IS NO ONE "CORRECT" RANKING OF FOOTBALL TEAMS.

Let me explain. Team A is better than Team B. Team B is better than Team C. Team C is better than Team A. All of these statements can be objectively true simultaneously. In what order should they be ranked?

Just like in basketball, some teams match up better against some teams than they do against others. In the basketball tournament, all you can hope for in the selection and seeding process is that no team with a realistic chance of winning the tournament is excluded. To that end, we know we are including many teams who do not have a realistic chance of winning a tournament that spans three weeks.

In football, we have no such luxury. No matter how hard we try, if we only include two teams, or four teams, we will fail to include some team that is capable of winning it all. That might also be true sometimes if we expand the field to 8. But the more teams in the field, the less likely that is. And in any case, there is no guarantee that the eventual winner is, in objective fact, the "best" team. They are just the Champion.

But the point is that there is a VAST amount more of difference between the playoff poll and the AP than there is between the AP and Coaches. Those two track each other very, very closely. That shows that the committee has been doing their job and looking at things independently. 12 of the top 25 in AP and coaches are in the same spot. 9 are only 1 spot different. WVU, ECU and Ole Miss are two spots different. Michigan St. is 3. This poll only has 6 the same as the AP (2 of which are pretty indisputable) and 5 the same as the coaches.

I still would like to know more about the stats they are using. Check out this picture from CBSSports:



What stats do they have at their disposal in this "comparative analysis tool"? Should the general public know what stats they are looking at?

(10-29-2014 12:16 PM)Frog in the Kitchen Sink Wrote:  I still would like to know more about the stats they are using. Check out this picture from CBSSports:



What stats do they have at their disposal in this "comparative analysis tool"? Should the general public know what stats they are looking at?

from the CFP website.
"Data

Selection committee members will have a wealth of information including review of video, statistics and their own expertise to guide them in their deliberations. They will emphasize obvious factors like win-loss records, strength of schedule, conference championships won, head-to-head results and results against common opponents. The playoff group has retained SportSource Analytics to provide the data platform for the committee’s use. While the details of the platform have not been finalized, it is anticipated that it will include countless pieces of statistical information for every Football Bowl Subdivision team. It will also include general information such as each team’s opponents’ record and opponents’ opponents’ records. The platform will allow the committee members to compare and contrast every team on every level possible.

It should be noted that the committee will not use a single data point such as the Ratings Percentage Index (RPI) that is used for NCAA championships.

(10-29-2014 12:21 PM)MWC Tex Wrote:  

(10-29-2014 12:16 PM)Frog in the Kitchen Sink Wrote:  I still would like to know more about the stats they are using. Check out this picture from CBSSports:



What stats do they have at their disposal in this "comparative analysis tool"? Should the general public know what stats they are looking at?

from the CFP website.
"Data

Selection committee members will have a wealth of information including review of video, statistics and their own expertise to guide them in their deliberations. They will emphasize obvious factors like win-loss records, strength of schedule, conference championships won, head-to-head results and results against common opponents. The playoff group has retained SportSource Analytics to provide the data platform for the committee’s use. While the details of the platform have not been finalized, it is anticipated that it will include countless pieces of statistical information for every Football Bowl Subdivision team. It will also include general information such as each team’s opponents’ record and opponents’ opponents’ records. The platform will allow the committee members to compare and contrast every team on every level possible.

It should be noted that the committee will not use a single data point such as the Ratings Percentage Index (RPI) that is used for NCAA championships.

Yeah, but I'd like to know what they are emphasizing. No way they are using "countless pieces of statistical information" to compare teams. There must be some pieces of statistical information they are emphasizing from team to team to team, and I think the public should know what those are.

(10-29-2014 11:35 AM)stever20 Wrote:  The thing was- the wins were close enough- 3 BCS top 25 wins for both teams. 2 of the winning teams that OSU played was La-Lafayette and Tulsa.

Also you don't just look at conference records. That's crazy. Alabama's conference wins were 42-42 overall. 24-4 OOC record that you can not deny. OSU's conference wins were 55-41 overall. 21-3 OOC record.

OOC while yes Oklahoma St had 2/3 winning teams- they were La-Lafayette and Tulsa. Alabama had 9-3 Penn St better than either of those by a wide margin.

So it was pretty close on the wins and no contest on the losses.

Let's see. so you point out that wins versus UL-Lafayette and Tulsa don't count because they are not P5 (remember that when talking about your ECU team), then ignore that one team played one additional P5 team, and the other replaced that with a 1AA team? Then you say that Bama's conference opponents conference record doesn't matter, even though you more or less just discarded non-P5 wins, when those conference wins represent nearly all of those teams P5 wins? Just to note, those seven teams combined to win two P5 games OOC, so yes their conference record most certainly matters as they are their only quality wins (according to you). And you think that doesn't matter when it comes to measuring SOS?

Oh, and then you ignored all of the ACTUAL SOS rankings? Or the fact that Alabama played 8 teams with non-winning records, and act as though that doesn't matter?

You should probably realize that when you make an argument on one side, it must be consistent. You have a problem with that. If your barometer is only power conference wins apply when it comes ot establishing OK St's relative strength, then it also only applies when establishing their opponent's strength. You can't then disregard that, when trying to establish the teams on the other side.

Frog in the Kitchen Sink Wrote:  Yeah, but I'd like to know what they are emphasizing. No way they are using "countless pieces of statistical information" to compare teams. There must be some pieces of statistical information they are emphasizing from team to team to team, and I think the public should know what those are.

I bet if we saw the data they were looking at, it would actually be a game of Asteroids on the computer.

(10-29-2014 12:53 PM)adcorbett Wrote:  [quote='Frog in the Kitchen Sink' pid='11304869'
Yeah, but I'd like to know what they are emphasizing. No way they are using "countless pieces of statistical information" to compare teams. There must be some pieces of statistical information they are emphasizing from team to team to team, and I think the public should know what those are.

I bet if we saw the data they were looking at, it would actually be a game of Asteroids on the computer.
[/quote]

I was thinking Pac-Man.

Looking at that picture I have only one question......
What can brown do for you?

Penn St was a quality win for Bama as well that year.

And I'm sorry but when a person looks at a team they don't look at just the conference record. they look at the entire record. For instance Vanderbilt was 2-6 in conference but 6-6 overall. HUGE difference there. Also especially with the Big 12 it's a round robin so their conference opponent record will be close to .500. it has to.