Famous NCAA Dynasties

[mountainman]

I am hopeful that someone who has written or is knowledgable about the computer programs that rank college football teams and a voter in both the Harris Poll or the Coaches Poll will become a member of this forum.

I am quite curious about how those things work. I have my own ideas, but would like to know with some certainty.

The fact that the BCS feels the need to use a couple of human polls with a lot of people voting in those polls and six computer polls where they throw out the highest and lowest results and then average the remaining four tells me something, but I'm not sure exactly what.

[ktffan]

I am hopeful that someone who has written or is knowledgable about the computer programs that rank college football teams and a voter in both the Harris Poll or the Coaches Poll will become a member of this forum. 8)

I am quite curious about how those things work. I have my own ideas, but would like to know with some certainty. :D

The fact that the BCS feels the need to use a couple of human polls with a lot of people voting in those polls and six computer polls where they throw out the highest and lowest results and then average the remaining four tells me something, but I'm not sure exactly what. :?

Most of the BCS computer programmers will not tell you what their criteria is. That's a problem in my book. If you're looking to win a beauty contest, it's nice to know what you should focus on. Not that I blame the programmers for not putting out their info, as it does earn them money, but it's not fair to team competing to not know the criteria.

If you have questions about computer rankings, I recommend you go to James Howell's site and send him your questions. He's been doing a computer ranking for quite a number of years and is a rather nice guy and would probably answer any questions you have.

[mountainman]

Thanks for the direction, ktffan.

I suspect there is a lot of subjectivity in both the human and computer polls and programs.

I would also be concerned that if the subjectivity were taken out and folks did know exactly what was going on, then college football would regress, if my opinion, back to the days when 'margin of victory' was being given more weight and teams were out there running up the score just for the sake of rankings at the expense of their opponent. That would be similar to teaching to the test and could possibly take the game in a direction it really should not go. For example, if yards rushing carried more weight than yards passing, coaches would probably run the ball more and thus reduce the value of having a 'balanced offense'.

[ktffan]

Thanks for the direction, ktffan. :D

I suspect there is a lot of subjectivity in both the human and computer polls and programs. :?

I would also be concerned that if the subjectivity were taken out and folks did know exactly what was going on, then college football would regress, if my opinion, back to the days when 'margin of victory' was being given more weight and teams were out there running up the score just for the sake of rankings at the expense of their opponent. That would be similar to teaching to the test and could possibly take the game in a direction it really should not go. For example, if yards rushing carried more weight than yards passing, coaches would probably run the ball more and thus reduce the value of having a 'balanced offense'. :(

Given that there is so little information out there to compare teams, margin of victory is a useful tool and when the BCS forced the computers to eliminate it, they made the computers less accurate. Having the margin of victory capped at 21 points was a good idea. It still allowed you to use that information, but gave less incentive for a coach to run it up.

[mountainman]

Maybe, but there's a flip side to it.

I saw incidences where the #1 offenses were playing against the #3 defenses and vica-versa in games that were already decided. Along with teams seemingly 'permitting' opponents to run up the score for what to me was blatant manipulation of the system for the sake of rankings. I realize that's a coach's integrity issue, but the system in place was the influence. A system that encourages or promotes that type of behavior is the problem, not the coaches.

[ktffan]

Maybe, but there's a flip side to it.

I saw incidences where the #1 offenses were playing against the #3 defenses and vica-versa in games that were already decided. Along with teams seemingly 'permitting' opponents to run up the score for what to me was blatant manipulation of the system for the sake of rankings. I realize that's a coach's integrity issue, but the system in place was the influence. A system that encourages or promotes that type of behavior is the problem, not the coaches.

You aren't going to get many third string defenses playing significant time when they score is less then 21 points. However, even after we've taken it out of the computers, coaches still run it up to influence the human voters.

[colorado_loves_football]

I am hopeful that someone who has written or is knowledgable about the computer programs that rank college football teams and a voter in both the Harris Poll or the Coaches Poll will become a member of this forum.

I am quite curious about how those things work. I have my own ideas, but would like to know with some certainty.

The fact that the BCS feels the need to use a couple of human polls with a lot of people voting in those polls and six computer polls where they throw out the highest and lowest results and then average the remaining four tells me something, but I'm not sure exactly what.

What it tells me is that the BCS ranking is an 'ad hock' arrangement.
I provided a reasonable 'alternative' to a computer ranking, although it does utilize a formula to fairly rank teams, it's the power ranking, slightly modified to allow for a more 'representative' overall ranking of teams.
Kttfan dismissed it, but unless I'm mistaken, Spence mostly appreciated it. See, I'm not an antagonist. I look for reasonable solutions, then apply them where appropriate.
TCU, incidentally, didn't finish any higher in that poll than they did in the Harris or coaches' poll, in fact they finished lower.
Averaging the two together would have given TCU a 'respectable' #10 ranking overall, and that's one reason why I support it, it's fair.

TCU wasn't 'rewarded' disproportionately for beating Oklahoma early in the year, neither were they 'punished' disproportionately for losing to SMU, although that did affect their overall ranking.

A #11 ranking was likely a 'fair' ranking. Since ten teams are represented to the BCS, TCU likely wouldn't qualify under those standards. Even so, I think they are likely represented, as an 'at large' to the BCS, so it would be a 'fair' way to either select or reject a team.

An 'average' between the coaches' and the ranking I mention would likely give a team a 'reasonable' chance at a BCS bid, even if they 'stumble' as TCU did. And, it woudl'nt require a multitude of formulas, averaged together. I think anytime you do that, you are playing with fire. Even if they don't throw out a 'high' or a 'low' it's likely not a representative number, unless they all agree.

[mountainman]

Hey, CLF. How would you measure the results of this system you've been advocating to replace the BCS system. What assurance can you give that it would produce the correct results, not the ones one might think are correct, but the results that are indeed correct?

[ktffan]

Hey, CLF. How would you measure the results of this system you've been advocating to replace the BCS system. What assurance can you give that it would produce the correct results, not the ones one might think are correct, but the results that are indeed correct?

Simply put, he likes the results. That's all the qualification he needs.

[billybud]

A ranking system purports to display the relative strength of one team versus others....If the system predicts "winners" of games at a very high rate, I'd rate the algorithim as a "good" one. The best algorithmns are picking winners in 3/4 of the games....now, when you factor in how many mismatches like Illinois/Mich, FSU/Duke there are....you'd think that 3 of four should be doable.

Wolfe and Massey were wrong last year on one third of their picks...I might have some doubts about their algorithims.

[Spence]

I don't know. To me, it's a healthier sign of a program, if they are planning ahead and meeting their needs. I've always been perplexed that Nebraska fans are so proud about their sell out streak. Essentially, it's next to impossible to get a ticket, and they think this is a good thing. At the same time, they have a stadium that is a good deal smaller than the largest stadiums out there. It seems to be that if they had planned for such a thing in the 70s, they could easily have a stadium that would be meeting their needs now instead of a string of sell outs, which really mean nothing except people can't get in.

There are two sides to that, though. Nebraska would be better served with a bigger stadium and I agree that it isn't worth a "sellout streak" not to do it. On the other hand boosters and fans have a tough time with "progress", they want their traditions(including stadiums) to remain.

Ohio State has wrestled with this for years. They do not want to change the flavor of Ohio Stadium, on the other hand they know they can put another 10 or 12 thousand people in the 'Shoe if they can figure out a way to get more seats in there. Boosters and supporters are a very significant part of the revenue at the large D-1 schools. You can't go and lose their support for 2 or 3 million a year in revenue.

I do agree that 65,000 seat stadiums that sellout regularly should find a way to add 20,000 seats. Not doing so is leaving much needed money on the table.

A ranking system purports to display the relative strength of one team versus others....If the system predicts "winners" of games at a very high rate, I'd rate the algorithim as a "good" one. The best algorithmns are picking winners in 3/4 of the games....now, when you factor in how many mismatches like Illinois/Mich, FSU/Duke there are....you'd think that 3 of four should be doable.


Wolfe and Massey were wrong last year on one third of their picks...I might have some doubts about their algorithims.

I agree. To be taken seriously, you would need to be right 70-75% of the time.

[colorado_loves_football]

Hey, CLF. How would you measure the results of this system you've been advocating to replace the BCS system. What assurance can you give that it would produce the correct results, not the ones one might think are correct, but the results that are indeed correct?

If you are asking me if it covers every possiblity, I guess the answer is 'no' but that depends on how the BCS is represented, by poll or by conference.

If it's by a poll, then the BCS would be assured of a 'competitive' BCS.
They would also likely be assured of a 'concensus' national champion because only those teams viewed as sufficiently 'capable' are included.

But, that's not what I would want, in principle. I prefer a 'representative' BCS, one where every conference is given an opportunity to participate.

This would also be more in the 'tradition' of a playoff bracket, although it would also require that teams be sufficiently 'capable' of playing competitive BCS football.

Last year, for example there were teams, represented, that likely wouldn't have been in the BCS, regardless. I refer to the Liberty Bowl, where Tulsa, and Fresno St, were paired together. A preferable arrangement, might have allowed TCU and Oregon to play, either in the Holiday Bowl, or the Liberty Bowl.

I likely would have 'kept' the Fiesta Bowl pairing of Notre Dame and OSU, but in a 'pre-BCS' arrangement, thereby assuring the BCS of the best possible pairing in the 'traditional' BCS. The Fiesta Bowl, likely would have been the 'host' bowl for the eventual winners (OSU vs. TCU/Oregon).

Now, would the BCS serve to select a 'concensus' national champion?
I am inclined to think it would. Remember, only the 'best' teams would be represented in the BCS. So, the 'remaining' bids would still be occupied by confernece champions, but likely in more 'traditional' venues.

For example, I believe the Rose Bowl would likely want to 'preserve' tradition and pair USC and either Ohio St, or Penn St.

Similarly, the Sugar Bowl, would likely prefer a Texas vs. Georgia pairing.

The Orange Bowl, in my opinion, would 'accept' a Big East vs. ACC pairing, although they might 'prefer' Penn St (as they did).

Regardless, of how the final pairings are determined, only the 'best' four teams are left standing. They are then paired together, competitively.
The probability that USC plays Texas in the national championship is high.

The winner of that game is crowned 'national champions'. So, yes, I think it covers most, if not all, possibilities. Could an upset happen?
Certainly. Would that 'ruin' the BCS? I do'nt think so.

[Spence]

But, that's not what I would want, in principle. I prefer a 'representative' BCS, one where every conference is given an opportunity to participate.

It may not ruin the BCS, but it would be unfair to the teams that work their butts off to be the best to watch a team who competitively couldn't keep up with them. that kind of system promotes mediocrity. I also makes the regular season a joke.

[mountainman]

Well, I've read over the course of the past several months how your system would work.

My questions were about measuring results and assurances of correctness. Those are the tough ones, but need to be answered, demonstrated and proven, with at least some certainty, before any system should be implemented or any system should be changed. There is much at stake, as far as college football is concerned, and they do not have the luxury of experimenting with things they 'think' or 'hope' might work.

The reason I support the BCS arrangement is not because it is fair or unfair. Any system I've seen so far, including yours, can be said to be fair or unfair and will have its shortcomings .... polls are bias, computers lack perception, playoffs ask too much of the student athletes and of college football tradition ..... I support it because, even with its shortcomings, it demands excellence while supporting tradition, provides a national championship game and is representative of every program that has chosen to be classified as D-1A.

When all things are considered and someone comes up with a better system then I'll support it.

[colorado_loves_football]

Well, I've read over the course of the past several months how your system would work.
The reason I support the BCS arrangement is not because it is fair or unfair. Any system I've seen so far, including yours, can be said to be fair or unfair and will have its shortcomings .... polls are bias, computers lack perception, playoffs ask too much of the student athletes and of college football tradition ..... I support it because, even with its shortcomings, it demands excellence while supporting tradition, provides a national championship game and is representative of every program that has chosen to be classified as D-1A.

When all things are considered and someone comes up with a better system then I'll support it.

I never said my proposal was 'perfect' but I think it's the best solution for the BCS, barring an all-out playoff, that would destroy whatever bowl tradtion you favor.

We can argue about whether the BCS is 'fair'. I don't think it is, but that's an argument for another day. It is mostly a 'representative' arrangment, so if nothing else it does allow for some equality in that respect. I would prefer that every conference were given a representative, but that wouldn't appear to be on the horizon.

Now, you said something about favoring an arrangement that gives every team 'fair' opportunity. I don't know of any such arrangement.

The BCS, really isn't supposed to be a 'fair' arrangement. What it is, is a competitive arrangement of teams, equally 'qualified'. I simply would arrange them in a fashion that allows one team to win the whole thing.
It really isn't a 'playoff' as much as it is a competitive way to select a national champion.