Burnt Orange Nation: An SB Nation Community

Navigation: Jump to content areas:



Around SBN: Staturday: Does Jason Giambi help the A's? Bar-right-arrows



spread the word

In Defense of the BCS (Well, Sort Of...)

This article isn't going to win me any friends, I'm certain of that.  Why?  Because the other thing I'm certain of is that you hate the BCS.  This sentiment has never been as fever pitched in the great state of Texas as it is now.  Choruses of "we got screwed" ring out from every corner of Longhorn fandom.  I've had countless friends unsuccessfully try to console me by mentioning a playoff.  President Elect Obama went on Monday Night Football the day before the election and, while predictably behind the times McCain was rambling about steroids, made a brilliant political pitch to the South in advocating for an 8-team college football playoff.  He repeated this advocation on 60 Minutes after the election, showing that he's legitimately serious and wasn't just pandering to a constituency he coveted.

But I beseech you, Mr. Obama, if you are serious about this whole "Team of Rivals" business, bring someone into the fold who has thought endlessly about what the college football postseason should be and who, while agreeing with you on the fundamental issue that the BCS is flawed, does not agree with you on the best way to fix that flaw.  This is an incredibly nuanced issue and there should be an incredibly nuanced debate.  But to be honest, sir, for someone as enamored of nuance as yourself, you display a shocking lack of nuance in your college football playoff position. Bring someone into the fold who understands the nuance, knows the political angle of it all, and above all else who won't blindly agree with you. Someone like, say, me?

Kidding aside (NB to Obama transition team: not kidding; please call), this mess merits looking more closely at the BCS than most of us have been in our blind calls for a playoff.  So let's do that, shall we?

Star-divide

Let's start with the recitation of some facts and then go from there:

  • The BCS National Champion is decided on the field, not by a formula.  The BCS does not decide the national champion, it decides who plays in the national championship game.  Your beef is with how those 2 teams are selected, not how the national champion is selected.
  • The BCS is a playoff.  If a certain number of teams play each other on the field after the regular season and the last team remaining without a loss in that postseason is declared the champion of the sport, then it is a single elimination playoff.  Thus the BCS is a 2-team playoff.  Your beef is with how many teams are in that playoff, not with the fallacious "fact" that there is no playoff.
  • The BCS was meant to pit #1 vs. #2 at the end of the year to create a championship game in which the national champion was determined on the field rather than by a poll.  It was not meant to do anything else.  Your beef this year is not with the BCS, but rather with the Big 12 for foolishly tying its tiebreaker in with BCS rankings.

First of all, I want to address the third bullet point.  What happened to Texas this weekend is not the fault of the BCS and is absolutely no reason to abandon the BCS system.  Please stop blaming the BCS for this.  This is the fault of the Big 12 and the Big 12 alone.  The BCS is not a conference ranking system, it is a national ranking system.  A national ranking system takes into account factors that a conference ranking system should not, such as non-conference schedule.  The BCS is set up to do one thing and one thing only and the Big 12 decided to use it to do something completely different, and that's the fault of the Big 12, not the BCS.

Moving on to what the BCS was actually meant to do, it must be noted that the BCS was created to rectify the problem of the "mythical" national championship.  Teams like undefeated #1 Texas didn't play undefeated #2 Penn State in 1969 to determine the national champion so it was awarded to the team that was #1 in the polls (or, in some cases, by Richard Nixon in the locker room after the regular season), giving such national championship a "mythical" or somewhat illegitimate quality.

[Aside: In fairness, Penn State had the option to play Texas in the Cotton Bowl, but turned it down to go to their pre-assigned Sugar Bowl because, I don't know, they liked debauchery?  They were scared?  Nixon had already named Texas national champion on national television?  Who knows.  And for those of you who are curious, yes, Joe Paterno was the head coach in 1969.  In fact, JoePa's had 4 undefeated seasons in which Penn State was not named national champion, including BOTH 1968 and 1969 and most recently in 1994 with Kerry Collins, Ki-Jana Carter and Bobby Engram.  The BCS was created so things like this didn't happen.]

The national championship is no longer mythical, irrational claims by bloggers notwithstanding.  There are rules set forth before the season regarding who gets into a playoff to determine the national champion and whoever wins that playoff is declared national champion of division 1 college football.  This is not mythical.  This is exactly how every major sport does it.  Your beef is simply with how many teams make that playoff and how we decide which teams make it, not with the legitimacy of the national champion named.

No doubt that for most of us (though not all), this system is superior to the previous system.  But of course that doesn't make it perfect anymore than it makes Mack Brown perfect for just happening to succeed John Mackovic as head coach.  And most people will agree, even those that don't consider themselves "playoff proponents" (considering the connotation that rides shotgun with that term), that there are flaws in the BCS system.  As I stated in the facts above, though, those flaws are confined really to merely two categories: how many teams are in the playoff and how such teams are selected.  Let's look at each.

How Many Teams Make it to the College Football Playoff:  Currently the BCS restricts this to two teams.  Barack Obama wants eight.  I've heard calls for as many as 64 and I've heard calls that aren't patently ridiculous for as many as 16.  The number of teams that make it to a playoff depends on what you want that playoff to be, as I have discussed previously on this site.  Loyal and long-time readers know that I have been angling for a Flex Playoff system for over two years (see comments on this post and this post for the first primitive articulation of the system).  The general idea is: (a) the college football national champion should be the team that has had the best season overall, (b) the college football playoff should include only those teams that have a legitimate claim to have had the best season overall, (c) the number of teams with such a claim changes each year, and therefore (d) the number of teams in that playoff should change accordingly, under rules for determining who has a legitimate claim to have had the best season overall.

The ultimate articulation of this system was a lengthy two-part treatise I wrote in early 2007 outlining the theoretical basis for the Flex System and then the Flex System itself.  That's required reading for anyone who wants to fully understand what I'm talking about here, with the caveat that I will be slightly amending the rules of the Flex System itself this offseason as further thought has led me to slightly different conclusions about how to deal with certain situations.  Be advised that these revisions have nothing to do with Texas' exclusion from the Big 12 Championship game (and thus the revisions are not just sour grapes) because (a) that exclusion has nothing to do with the college football playoff, as I have mentioned previously, and (b) Texas would already be in the college football playoff under the current Flex System, no matter what happens this weekend.

At the very least, I think that just about everyone can agree that in at least some years, two teams are not enough to be let into the playoff, most notably in 2003 (no undefeated teams but three indistinguishable 1-loss teams: USC, OU, and LSU) and in 2004 (three undefeated BCS conference teams (USC, OU, and Auburn).  I personally believe that in some years, two teams is exactly the right number of teams that should be put into a playoff, most notably in 2005 when USC and Texas were the only two undefeated teams in the country and widely agreed to be the only two with a legitimate claim to have had the best year.  But, continuing with the Penn State theme, what if Chad Henne hadn't hit Mario Manningham for a game-winning touchdown at the last second to give the Nittany Lions their only loss of the season?  Penn State likely would have been an undefeated #3 and JoePa would have 5 undefeated seasons that didn't result in national championships rather than 4.  And that's not fair.  

So if there are some years in which two teams are simply not enough to put into a playoff, then you have two options: (1) adjust the number of teams who do get in depending on the circumstances of each year to only include teams with a legitimate claim to have had the best season overall (i.e. the Flex System) or (2) increase the number of teams that get in every year, and thereby let in teams that clearly do not have a claim to have had the best season and allow them the possibility of winning the national championship.  Option 2 is what just about every other sport does (though MLB held out against this for a long time and still tries to maintain some semblance of this idea) and it rewards teams who get hot at the end of the season but who may have lost a lot of games early over teams that have had better seasons.  Both are viable options and I have my obvious preference.  But something needs to be done.

How Teams that Make the College Football Playoff are Selected:  The determination of who gets into the playoff is a bone of contention as well.  This has to be done in some manner, whether it's by a committee at the end of the season (like college basketball) or it's by only letting in conference champions or it's by a ranking system of some sort.  Because very often some of the best teams in the country all come from the same conference, I think that you have to let in teams that did not win their conference if you really want to have a meaningful playoff.  So for me, it comes down to a committee or a ranking system.  Let's first look at a ranking system.

We're all familiar with the BCS ranking system after the massive amount of posts I've done over the past few weeks trying to figure out a possible way for Texas to make it to KC and Miami.  I actually think this is a fine system in theory, with three caveats: (1) there should be nothing that computers can't take into account except for perhaps margin of victory over a certain point (really, beating a team by 42 points isn't much different from beating them by 28 points), (2) there should be more computers to get a meaningful average, and (3) human voters shouldn't be idiots who don't follow college football or coaches who think a team they just voted ahead of Texas is undefeated when they actually lost to Texas or who just vote for the team that played better against their team rather than which had the better season.

Honestly, I think the computers plus humans system is a good ranking tool.  Humans are able to detect nuance in a team's performance that computers are not, and computers are able to give objective treatment to the accomplishments of each team whereas humans are not.  The BCS can easily fix problem (1) above by executive fiat, and can fix problem (2) by not eliminating the high and low rankings and adding a few more reputable computers.  Those are easy.  The hard part comes in fixing problem (3).  First, the coaches poll has to be eliminated from the system.  If the coaches want to have a poll, that's fine.  But it should have no bearing on who goes to the playoff.  The ideal human poll is something like what the Harris Poll is supposed to be: a group of intelligent college football fans who know the sport through and through and who legitimately respect the job of ranking teams.

But the main problem with the notion of human voters at all is the recent dawning of their enlightenment about the place they hold in the BCS system.  One thing that I'm not particularly comfortable with is the fact that human voters are learning how to manipulate their votes in their individual polls to get the result they want from the BCS.  Maybe they're not actually doing this and it's just us speculating that they are, but all this talk of "voters won't allow a 1-loss SEC champion to be left out of the national championship game so they'll vote Florida #1 over OU just to make sure Florida goes over Texas" is extremely scary.  Voters should be voting on who they think has had the best season and on no other basis, particularly not one with a specific agenda.  In a sense, voters should not be sentient about their place in the system.  They should, just as the computers do with their unique abilities, use their unique human abilities to rank teams and not worry about the ultimate effect of those rankings.  This is a corollary to the stated reason for why the AP pulled out of the BCS: they want to report news, not make it.

Let's not forget that after all our politicking last week, large swaths of human voters moved Texas ahead of OU after Texas blew out a terrible team and OU won fairly convincingly over a good team.  Did these two results warrant this change?  Absolutely not.  Voters were responding to what they saw as an injustice that was about to happen.  Voters probably should have had Texas above by a lot to begin with and then considered the possibility of moving OU ahead after they beat OSU, but that's tangential to my point here.  My point is that voters can be swayed by politicking. Sometimes it's politicking to get them to actually focus on the results of the season, but sometimes it's not.  Sometimes it's to say "Hey, make sure you put Florida ahead of OU and Alabama ahead of Texas on your final ballot so we can make sure that the computers don't send Texas to the national championship game ahead of the Gators."

How do you get rid of that?  You don't release any BCS rankings before the last one!  You don't release the polls that are included in the BCS formula until late in the season!  And you make every single person's vote public every single week that the poll is released along with a minimum 500-word explanation of why they ranked the top 5 they way that they did!  These aren't going to cure everything, obviously.  But the more transparency there is and the fewer opportunities there are for blatant manipulation of the BCS standings, the better the system will be.  If you have read the Flex System proposal, you know that my preference is for taking these BCS rankings and applying certain rules to them for determining who gets into a playoff (i.e. if there are only 2 undefeated teams in the top 5 of the BCS standings and they are ranked 1-2, then there is a 2-team playoff between those 2 teams--like in 2005 with Texas and USC).  But however the BCS standings determine who makes the playoffs, that ranking system needs some tweaks.

I owe you a brief word on the committee system, and here it is.  It only works if you have a giant playoff where the last few teams in have approximately 0% chance of winning the playoff and where the teams in the playoff have to win a lot of games to win the whole thing.  The fewer games there are in the playoff, the more illegitimate the committee will seem.  Now, you will say that the same is true of a ranking system which means we should just have a big playoff, but that's not true.  For instance, in my Flex System, the rules for who gets into the playoff are set before the season starts, not after the season is over.  You may disagree with who gets in and who doesn't but it's not illegitimate because it's not biased against any specific teams.  If you decide who gets into a playoff after the season, the decisions are suspect for bias, and if it isn't a huge playoff, there isn't a reason to dismiss any alleged claims of bias.

Conclusion

I know there appears to be a rift between the title of this post and the content, in which I have criticized the BCS throughout.  But if you look closely, the criticisms aren't of the premise of the system, but rather of its methods.  It's an affirmation of the BCS, which is a playoff system in which the winner of that playoff is named national champion, and a refutation of numerous arguments against it.  But it's also a recognition that the system would be better if sometimes we let in more teams than two, and maybe we slightly reformed the system for determining who gets invited to the playoff.  At the least, it's a plea for everyone to make nuanced arguments for what we believe rather than blanket statements condemning  the BCS (and that means you too, Mr. Obama).

REAL Conclusion: Seriously though, Mr. President Elect.....shoot me an e-mail on your blackberry while you still can.  Team of Rivals!

[Note to everyone: This post is not an invitation to talk about politics in the comments.  Talk about politicians and politics only in the context of what it means for college football.  Nothing else.]

0 recs | Comment 111 comments | Add comment | Digg!

Read Related

Story-email Email Printer Print

Comments

Display:

Advantage of the flex

Silver bullet to any and all worries about the regular season becoming less meaningful: If you don’t know whether you need to finish in the Top 2 or Top 6 to get in the playoff, you assume they’re all must wins and are grateful for any one- or two-loss opportunity the season happens to provide when it’s all said and done.

--PB--

by PB @ BON on Dec 3, 2008 4:22 PM CST reply reply   0 recs

Nice.

That’s one’s going in the rewrite.

by billyzane on Dec 3, 2008 4:46 PM CST to parent up reply reply   0 recs

Nice post BZ

I also like the BCS for the most part and though the lead up to figuring out just who those two teams will be at the end can get pretty messy, I think it’s hard to argue with the eventual MNC teams from the past 10 years.

USC in 2003 and Auburn in 2004 may have a legit beef, but I think the BCS still got it right. If OU goes on to win the title this year convincingly, I may change my mind, but again, that won’t be the fault of the BCS, it’ll be the fault of the Big XII.

And as a fan of college football in general, nothing was more frustrating than those years when teams like Nebraska and Michigan would share the title with no op to settle it on the field.

Be nobody but yourself in a world that desperately wants you to be like everybody else.

by 54b on Dec 3, 2008 6:40 PM CST to parent up reply reply   0 recs

"if OU goes on to win this year convincingly"

then, to me, the system worked. We’re all crying UNJUST because we legitimately think that we’re the better team. We think OU got some scheduling advantages, that their defense is suspect, etc. But if OU goes and stomps Florida (in Miami, no less), then I will have to concede that the BCS worked.

As long as the National Title goes to a top-3 team at worst, then I’m ok with the system. I think most sports allow for top-10 or top-12 teams to get a shot, which is far more unjust than any results the BCS will likely ever produce.

by BrooklynHorn on Dec 3, 2008 6:47 PM CST to parent up reply reply   0 recs

That's an interesting point

In this scenario, the BCS has affected the way the Big XII champion is chosen, which in turn then affects how the BCS champion is chosen. Thus the Big XII, through its own stupidity, has tarnished the entire BCS. Thanks, Beebe!

by Meekrob on Dec 4, 2008 8:43 AM CST to parent up reply reply   0 recs

What? How dare you defend the BCS!

Actually, good post.

However, let me focus on this one point:


The BCS National Champion is decided on the field, not by a formula. The BCS does not decide the national champion, it decides who plays in the national championship game. Your beef is with how those 2 teams are selected, not how the national champion is selected.

This is true in theory, but the BCS arguably leaves out teams that had a much better chance of winning. For instance, virtually everyone knew that Ohio State had no shot against LSU if the Tigers showed up half-awake (which they did, allowing a quick 10 points but coolly taking control from there). In that regard, some may argue that LSU was “given” the national title, not having to play superior teams to Ohio State like Georgia or USC.

by TheElusiveShadow on Dec 3, 2008 4:48 PM CST reply reply   0 recs

This furthers my point, though...

The BCS formula did not crown the national champion. LSU won that by beating Ohio State. Your beef is with how the teams are selected or how many teams get selected.

by billyzane on Dec 3, 2008 4:52 PM CST to parent up reply reply   0 recs

Eliminates Contenders

The BCS formula does not crown the champion but it sure makes it impossible for several legitimate teams to be champion.

As you state we don’t have a beef with whoever wins the BCS championship, but with who gets to play in the BCS championship game.

by drycreek on Dec 3, 2008 10:27 PM CST to parent up reply reply   0 recs

BCS and Margin of Victory
There should be nothing that computers can’t take into account except for perhaps margin of victory over a certain point (really, beating a team by 42 points isn’t much different from beating them by 28 points)

This is what I thought as well, however when I was reading the Colley Matrix methodology today (work? Whats that?) which was written in 2002, and apparently not updated since I came across this:

However, even with considerable mathematical skulduggery, reliance on scores generates
some dependence on score margin that surfaces in the rankings at some level. Rightly or wrongly,
this dependence has induced teams to curry favor in computer rankings by running up the score
against lesser opponents. The situation had degraded to the point in 2001 that the BCS committee
instructed its computer rankers either to eliminate score dependence altogether or limit score
margins to 21 in their codes.

Which seems to indicate that what you suggest currently exists. Does anyone know for certain that MOV has since been completely removed?

Colley Matrix, by the way, concluded to not include MOV at all in its calculations.

by BoddickerIsClutch on Dec 3, 2008 4:49 PM CST reply reply   0 recs

It was completely removed after 2003

When OU’s massive lead in the computers (due to things such as it’s 77-0 romp over A&M and it’s 65-13 romp over Texas) allowed it to overcome AP and Coaches’ #1 USC in the BCS.

by billyzane on Dec 3, 2008 4:51 PM CST to parent up reply reply   0 recs

So..

The history of the BCS and MOV in computers is:

Inception – No rules, take whatever data you want.
2001 – Running up the score is rampant, limit MOV considerations to only 21 points
2003 – Ah screw it, remove it altogether, OU Sucks

If thats true, what argument would there be to add some MOV back in, much less at 28 points? The only way it could happen if the timeline above is correct, is if it was at an even smaller margin than 21. (My personal suggestion is a binary boon based on 10.5 points, but I digress)

by BoddickerIsClutch on Dec 3, 2008 4:58 PM CST to parent up reply reply   0 recs

The rationale is that it was stupid to take it out to begin with.

I actually had no idea that it was reduced to 21 for a while; that’s news to me. I don’t think there’s anything wrong with admitting you made a mistake. The BCS used to change the formula every damn year, which in effect was admitting that they were doing it wrong the previous year. I don’t know. But it needs to happen. 21 is a fine number, as is 28. I’m sure someone more mathematically inclined than I am can plug in some data and figure out where the valid “hump” is.

by billyzane on Dec 3, 2008 5:03 PM CST to parent up reply reply   0 recs

When MOV was used...

…did it take under consideration the relative MOV?

By that, I mean the following:

Team A defeats Team X 30-0 in a truly dominant defense performance.
Team B defeats Team X 62-28.

Although most observers would agree that Team A posted the “better” win, a pure MOV method would not recognize it as such. But if the MOV is calculated relative to the total points in a game, then maybe not.

by kjm017 on Dec 3, 2008 5:24 PM CST to parent up reply reply   0 recs

I have thought the same

Rather than margin of victory, I think some kind of relative victory score should be included. Some combination of the two might be appropriate.

by Longhorn in Canada on Dec 3, 2008 6:27 PM CST to parent up reply reply   0 recs

Problems with the "flex"

While I agree that the system should ultimately try to reward a team that was both (a) good in the regular season, and (b) good in the post season, I don’t believe that a flex system would acheive such a thing.

First – Having any sort of selective postseason playoff process ensures that every game matters. You don’t need to make it a moving target in order for teams to care about the regular season. I don’t think it is a fair assumption to say that teams will stop trying and therefore make regular season games less meaningful. For one, the sheer size of Div-1 forces you to care. College football has almost 120 teams in it vying for 8, 12, or 16 playoff spots and you are almost always assured someone is going to have an undefeated, one loss, or two loss season. Secondly, there is no motivation to “tank” games by teams who have no shot of making the playoff. Similar to what we see in professional basketball, football, and baseball because in the end the motivation is different. In professional sports, teams that have no chance of getting to the post season are more likely to be motivated to lose in order to secure a better draft position. In CFB, programs don’t have an incentive to lose — its not going to help recruiting.

Every league defines a minimum threshold for an acceptable regular season performance either through RPI and a committee, or a holistically through a wild card, or what have you. In college football, that should be the top 12 or 16 teams as ranked by a new ranking system which I will describe in another post below.

by BMG on Dec 3, 2008 4:52 PM CST reply reply   0 recs

No one said anyone was tanking anything.

The regular season does in fact count less with a larger playoff because you can still win the national championship with 3 losses. More games matter, but the outcome of those games doesn’t matter as much.

by billyzane on Dec 3, 2008 4:54 PM CST to parent up reply reply   0 recs

my point is simply

that if you take the top 8 or at most the top 16 out of 120 teams there is a very, very, small likelihood that their will be a three loss team in the mix. More often than not, before bowl season, the top 16 is comprised of teams with 2 losses at most.

by BMG on Dec 3, 2008 5:16 PM CST to parent up reply reply   0 recs

Realistically

Flex system, BCS, or full on playoff, you are still only talking about 16 teams a year with a realistic shot of making it to the championship game. Bump that to 25 if you want to take the current break point for “good” teams. Its not 120 vying for the top spots.

I don’t think its about teams tanking or not trying their hardest, but for me, as a fan anyway, its about the importance not only of my team’s regular season, but other teams as well. Like how I wanted Baylor to win badly last weekend, or Florida state, or insert-name-here. It makes me be a fan of the entire sport on a larger level than in any other sport.

That is what I usually think of when people talk of the importance of the regular season.

by BoddickerIsClutch on Dec 3, 2008 5:03 PM CST to parent up reply reply   0 recs

If we aren't worried about teams giving less than 100% effort

what difference does it make?

The measure of meaningfulness in sports is if a team gives 100% effort in the competition.

My argument is that as long as the playoff field is extremely selective (i.e. 16 out of 120 teams) every team will continue to give 100% effort in every game. If that is the case, then every team believes every game in meaningful.

by BMG on Dec 3, 2008 5:22 PM CST to parent up reply reply   0 recs

What?
The measure of meaningfulness in sports is if a team gives 100% effort in the competition.

I don’t understand this. This is the only measure you have of whether sport is meaningful? The only thing that matters to you in sport is what goes into it, not what comes out?

by billyzane on Dec 4, 2008 9:03 AM CST to parent up reply reply   0 recs

meaningfulness

BZ – correct me if I’m wrong, but what you want to continue see is that a game like Florida/Ole Miss is “meaningful” so that the Florida loss has consequences in a playoff system. Better yet, you like the idea that Nebraska/Colorado was “meaningful” because it impacted Texas and Oklahoma’s rankings hence their post season aspirations. In that sense, the games have meaning, correct? What you and others want to avoid is that a team with 3 or 4 losses is able to make it into a playoff and compete against undefeated, one, or 2 loss teams for the same mythical championship. This would somehow undermine the legitimacy of the championship because it would give lesser teams the ability to win it all and would make their additional regular season losses “meaningless” versus the undefeated or 1-loss teams.

I have a few problems with this view. First, its too simplistic. The last team into any playoff system has to beat out at least one other team in order to make it. There will always be a cutoff, regardless of whether you have a flex system or a standard top 8, 12, or 16 system. The best team left out of the playoff can always make the argument that they system left them out because somone’s win or loss was considered less “meaningful” in retrospect. The last team in will likely have some arbitrary reason for being in while another team is left out and depending on your bias that reason could be meaningful while another possible deciding factor could be meaningless. Furthermore, in any playoff system teams are seeded, so less qualified teams have a tougher road to win and are inherently at a disadvantage. Thus teams will always be motivated to win (and give their fullest effort) to secure a higher seed – ensuring that the wins and losses of each team have significance.

Second, you approach is a hindsight view of assigning “meaningfulness” in which a person can cherry pick a few games whose outcome that person feels should have had greater consequence. You will always be able to find games that are not “meaningful” if you assess meaning this way. Does this mean that some teams shouldn’t bother playing or that we should have just handed the national championship trophy to Georgia or USC without playing the season? This disregards the fact that there were dozens of other “meaningful” games that had to occur in order for the “meaningless” game to have no postseason significance. Regardless of whether or not the outcome of the game had or did not have consequences for the post season does not make a game “meaningless”.

Lastly, because you are determining meaning in hindsight, no one can predict which future game’s outcome will or will not be meaningful. It is only once the season is finished that you cherry pick a game here or there that was or was not meaningful to the outcome of the season. Therefore, to the players on the field and everyone involved every game at the time is meaningful or at least has the potential to be meaningful. Thus, there is 100% effort and the chips fall where they may.

by BMG on Dec 4, 2008 2:13 PM CST to parent up reply reply   0 recs

Very deontological, BMG.

That’s like the 4th formulation of the Categorical Imperative, right there. Now, I’m no utilitarian, but the meaning of sport has to be about more than just what’s put into it. If that’s all that mattered, then the value of the 1980 US hockey victory over the Soviet Union would have the same “meaning” as last week’s Texas game against A&M simply because all four teams involved gave 100% effort. There are numerous ways that games have different meanings attached to them, from the rational consequence of each game in the sport to the irrational emotion evoked by such games.

Now, you are misconstruing what I mean by meaningfulness, and I have absolutely no idea how you got this “retroactive” business and assigned it to me because I believe no such thing. You somehow manage to conflate my position (which you more or less state correctly in the 1st paragraph) with the idea that I would consider one team’s win or loss less meaningful than another’s and by doing so let one of the teams into the playoff instead of the other. This is absolutely not correct. The meaningfulness of each game is the same within a given system. Florida’s loss to Ole Miss has the same meaning as Texas’ loss to Tech. I’m not saying the meaning changes, only that the quality of the results (with regards to determining who had the better season) diminishes more with Florida’s loss than with Texas’ for obvious reasons. Perhaps this is “retroactive” as you say (and I’m not conceding that point, only saying that this is probably what you meant when you said my definition of meaning was retroactive).

What you’re doing is confusing my use of the world “meaningful” with “quality” or some other word that describes how the results of a certain game affect a team’s overall resume. I mean “meaningful” in the sense of how much the results matter to the crowning of a national champion. That is, when Florida loses to Ole Miss, Florida’s chances to win the national championship under the BCS system decrease a certain percentage. The more teams that are let into a playoff, the less of a decrease that is to Florida’s chances to win a national championship. The greater the decrease, the more the regular season matters. If Florida’s loss to Ole Miss decreases its chances of winning the national championship by exactly 1, then that game is a whole lot less meaningful towards deciding the national champion than if that loss had decreased its chances by 50.

The meaningfulness of the results of the regular season with regards to the national championship is inversely proportional to the number of teams involved in a postseason playoff that determines the national champion.

by billyzane on Dec 4, 2008 3:35 PM CST to parent up reply reply   0 recs

OUCH!

No need for philosphical name calling here. In my defense I was only answering your call for more nuanced debate.

We’ve got slightly different definitions of “meaningfulness” but are both using “quality” or “consequences” or “outcome” synonymously. Our disagreement is on (1) having a moving target for determining what a legitimate claim to the best regular season is and (2) the flex concept subjectively deciding how many teams can make that claim.

Let’s forget the concept of meaning (which we’ve completely beaten to a pulp) and stick to the notion of “quality” (or lack thereof) such as the one you define (i.e. quality of a regular season loss should undermine one’s championship chances). A fixed playoff system would rank teams based on the quality of their seasons via selection process. Inherently, teams with better quality face inferior teams. If the regular season is truly meaningful (%$^@, there is that word again) then the top ranked team should rarely if ever lose to the last ranked team. The results of the matchups should all approach their expected outcomes similar to how the first round of the NCAA tournament plays itself out over time (little to no upsets of teams ranked #1 or #2). The flex system just shields top ranked teams from having to play opponents we don’t want them to lose to, not teams they could potentially lose to. If a playoff system was in place in 2005 and USC had somehow lost prior to playing Texas you might have called it a travesty. Maybe USC just wasn’t as good as their regular season resume implied. At the heart of our disagreement is that if a team is worth its ranking, it ought to be able to beat inferior opponents on the field and not just on paper.

Logically speaking, your last point is conflict with the purpose of the flex system you propose. If we compared a scenario with eight 1-loss teams comprising a playoff field, or 10 2-loss teams comprising the playoff field you would say that the regular season in the first scenario was more meaningful (8 teams) than the second scenario (10 teams). If you agree that year-over-year, regular seasons are always meaningful, then there should always be a fixed number of teams in a playoff field.

by BMG on Dec 4, 2008 5:20 PM CST to parent up reply reply   0 recs

2005 USC playoff scenario...

…Sure they ought to be able to beat inferior teams, and that’s exactly what they did during the regular season. Why should they have to prove it again in a 4, 8, or however many team playoff? Nobody (except Texas) had a comparable resume, and making them play any playoff games before facing Texas in the finals would have devalued (or made less meaningful, if you prefer ;) their regular season.

by Sweed4Heisman on Dec 4, 2008 6:26 PM CST to parent up reply reply   0 recs

The problem with the 16 out of 119 rationale

is that it assumes that all 119 teams had a legitimate shot to begin with. If college football were somehow an egalitarian, “stock car” type sport, in which no team had significant advantages coming into the season, then you would be correct in your assessment that 16 /119 is a relatively low percentage that should increase the urgency of competition. But college football is a sport based on elite competetive advatages. In some respects, National Championships are won off the field by way of recruiting, facility up-grades, television contracts, etc.

More reasonably, we must concede that there are really only 6-8 legitimate title contenders entering a given season. In an 8-team playoff, all would make the cut, and in my eyes this kills the urgency of the regular season. The 6-8 teams we thought had a shot at the beginning, are the very same teams we’ve got at the end – and all each team needs to accomplish this is a 1-loss or possibly 2-loss season, which they all almost certainly would. So why play the regular season at all? An 8-team playoff would essentially take the preseason top-8 (which we’re correct about something like 95% of time) and build a tournament around them, rendering all the games in September and October as virtual pre-season games.

16 teams, to me is unthinkable. That would stretch even beyond the number of teams that legitimately have a shot, and would alter the regular season irreparably.

by BrooklynHorn on Dec 3, 2008 5:21 PM CST to parent up reply reply   0 recs

16 and 12 would always be too many, 8 almost always would be, and sometimes 4 as well...

… 99 times out of 100, there won’t be 8 equally deserving teams at the end of a season. This year, it would be very tough to argue that USC, Penn State, and Utah deserve the same shot at a National Title as Florida, Alabama, Texas, OU and Tech. Like BZ mentioned, in 2005 (and 2002 – Ohio State/Miami) you simply didn’t need any more teams.

On top of usually not having enough deserving teams to fill an 8, 12, or 16 team bracket, the likelihood of regular season rematches increases as you add more teams. When I worry about the meaning of the regular season being taken away because of a playoff, I don’t worry so much about teams not playing hard as I do about getting rematches. I realize that this may sound dumb since Texas just got snubbed in favor of a team it beat in the regular season, but I blame that on the Big XII tie breaker, the voters, and the computers.

by Sweed4Heisman on Dec 3, 2008 6:20 PM CST to parent up reply reply   0 recs

Big 12 Tiebreaker Issue Not News

Many of us have forgotten that last season could have also been a 3-way tie between Texas, Oklahoma and Okie State at 5-3 if the Cowboys had won. The Big 12 was going to ignore the language of Rule b. 3., which states: " The records of the three teams will be compared against the next highest placed teams in their division in order of finish (4, 5 and 6)", (which would have favored Texas) and instead go to Rule b. 5. awarding the South Division to the highest BCS team (which would have been Oklahoma).

This was a small controversy in the week before the 2007 Texas/aTm game. So small I guess that everyone forgot it and ignored the rule about the BCS deciding our conference champion until now.

by RMHorn on Dec 3, 2008 5:07 PM CST reply reply   0 recs

the reason they threw that out

A better record against teams 4, 5, and 6 means that you only had QUALITY losses, whereas if you lost more games against 4,5,6, you have some good wins but more bad losses… It doesn’t really decide anything, except one team performed as expected (tough losses to good teams, undefeated against bad teams) and one team was multiple personality (beat the good teams, crap your pants against the bad ones)

I think I just confused myself.

"I have CDO. It's like OCD, but the letters are in alphabetical order. Like they should be."

by BigMOman on Dec 4, 2008 2:44 PM CST to parent up reply reply   0 recs

Actually, using it makes sense

If you’re looking for the best team and everything else is equal, you should pick one that’s proven it can beat better teams, even if it’s slipped up against worse competition. So the teams that beat #4 and lost to #6 should probably be considered “better”, or at least “higher ceiling” (again, all else equal) than ones that lose to #4 and beat #6.

by SpartanDan on Dec 5, 2008 2:19 PM CST to parent up reply reply   0 recs

AP, etc.
This is a corollary to the stated reason for why the AP pulled out of the BCS: they want to report news, not make it.

That may have been their stated reason, but the real reason was exactly the opposite. A champion got crowned that didn’t follow their collective opinion so they took their ball and went home.

Your flex idea is interesting. I have had a similar though regarding the Big 12 CCG – it should only be played in years when the winner of the North and the winner of the South haven’t already played each other. If 2 teams play each other head-to-head twice, why should 1 win decide a champion?

by Horncasting on Dec 3, 2008 5:10 PM CST reply reply   0 recs

I think you have the years wrong

As I recall, the AP withdrew not after 2003 (the split championship year) but rather after 2004, when there was a lot of consternation not only about which two of three unbeatens should have been in the championship game but also who between Cal and Texas should have been going to the Rose Bowl. Didn’t that idiot in Alabama who had been voting us low (#9?) most of the season finally say “no mas” and rank us higher just to stop the flood of hate mail he was getting from Texas each week?

I think the AP not only didn’t want to make news but also didn’t want to subject its voters to the increased lobbying in the era of easy email communication.

by kjm017 on Dec 3, 2008 5:37 PM CST to parent up reply reply   0 recs

a better ranking system

You make a good point about the strengths and shortcomings of the human and computer elements in the polls. Humans are short sighted and tend to take the “what have you done for me lately approach” and some Harris and Coaches’ poll voters sound completely oblivious to what is going on in reality. On the flip side, computers don’t pick up on nuances, particularly when injuries occur or when teams start to look flat out dominant.

The problem with the ranking system as it currently stands is that these two things net against each other, instead of building on each other. We need a system that uses computers to reinforce the opinions of voters, and not the other way around. In other words, we need an RPI type system for college football. My suggestion would be as follows

1. Computers are solely reponsible for ranking Strength of Schedule.
2. SOS computed for all Div 1 teams and teams are ranked according to SOS score weekly from hardest to easiest. Raw scores would be converted into %, so that the team with the toughest schedule gets a 1.000 score and on down the list.
3. Harris voters rank the teams independant of the computer results weekly and point totals are assigned just as they are now.
4. Finally, each team’s Harris Poll point total is multiplied by its SOS computer score to create a composite score. Composite scores are then ranked weekly.

A ranking system like this would then allow the human polls to be complimented by the computers. Humans would consider what happens on the field and the relative impression of a team week-to-week, while the computers would be responsible for putting that human score into the context of the 12 week regular season. Addtionally, because the polls and the computers are not averaged against each other as they are now, you would not have the additional bias created by humans who elevate a team in the polls in order to overcome lower computer rankings.

This concept is similar to what “judged” sports have with a level of difficultly. This system would be intended to (a) create a hollistic approach to SOS ranking, and (b) rank teams based on their performance against the difficultly of their schedule as assessed by voters.

by BMG on Dec 3, 2008 5:13 PM CST reply reply   0 recs

Rejoinder

Some points I have made elsewhere that probably belong here:

1. For a playoff winner to be seen as a legitimate champion, then a certain percentage of the field needs to qualify. Look at every other sport — somewhere between a quarter and a third of the teams go into the playoff. Even with that silly ‘play-in’ game, there is no doubt that the NCAA basketball champion is somewhere in the top 65 teams. The problem for NCAA football is there are too many teams and too few games. For a comparable ratio, you would need to hold a six round playoff and there is no way for that to be scheduled.

2. The Big 12 blew it big time in not following the lead of the other major conferences and instead opted to continue to rely on the BCS rank as a tie-breaker, even after it became apparent what might result. I figure this was pure politics — they did not want to be seen as repudiating the BCS. It’s a bit self-serving for us to look back and say the Big 12 ought to use the head-to-head result between the top two of the three teams as the tie breaker, since it still leaves open how the top two are to be determined. And it would be absurd to bring the BCS back in again at that point. The top four tiebreakers pretty much exhaust all the common-sense methods for resolving the matter. You are left with things like score differentials and who went most recently. And the Big 12 lacks an impartial arbitrator. So what’s left? Coin flip?

3. It does not matter how many computer systems you employ, because you are assuming that the biases will all even out and there is no reason to think that is the case. You just have more rating factors with smaller coefficients. More importantly, none of these have ever been validated by comparing their results against an objectively correct set of rankings, because those do not exist. The best they can do is show they can match the human rankings, which kinda defeats the whole point of using computers.

4. This is not to say that the human rankings are not also a disaster. Ignorance, bias, and simple errors are enough to shade results enough to undermine their accuracy. And we have this arbitrary way of translating individual rankings into a composite rating that would make a statistician scream.

Before you criticize someone, walk a mile in their shoes. If they get mad, you're a mile away AND you have their shoes.

by Caradoc on Dec 3, 2008 5:14 PM CST reply reply   0 recs

Disagree about # of teams for a legitimate champion

I think 4 the majority of the time and 8 at most would be plenty. Getting beyond that and you are filling it with teams that quite frankly don’t deserve at shot at it.

Look at 2005. UT navigated an very tough schedule. At the end should they really have to play a #15, a #7, etc. Would a #15 or a #7 seed team have really deserved a chance at them? If so, doesn’t that just completely cheapen the regular season?

by Horncasting on Dec 3, 2008 5:21 PM CST to parent up reply reply   0 recs

So why doesn't basketball just start with the Final Four?

After all the season is longer and nobody loses out because of a single game.

Before you criticize someone, walk a mile in their shoes. If they get mad, you're a mile away AND you have their shoes.

by Caradoc on Dec 3, 2008 10:17 PM CST to parent up reply reply   0 recs

College football and college basketball are different animals

And nobody cares about the regular season in basketball nearly as much as they do football.

by Horncasting on Dec 4, 2008 9:15 AM CST to parent up reply reply   0 recs

Well, they actually used to.

The reason they don’t anymore is that long ago the NCAA decided to give up on the postseason basketball tournament as a way to determine the team who had the best season and went instead for a completely different season focused exclusively on entertainment. And they sure did a good job of making it entertaining, I’ll give them that.

Think about the big 12 basketball season. There are two conference champions, the regular season and the conference tournament champion. Which one do you think better measures how good your team is? The one that accumulates wins and losses over months and months and ranks teams accordingly? Or the one where you have to win 3 or 4 games in a week? If you said the latter, then we’re just going to have to agree to disagree on the nature of sports championships generally, but if you chose the former, then you can understand what I’m getting at here. A postseason tournament can be incredibly entertaining, but it rewards different things than a regular season championship does and if the postseason tournament champion is viewed as more important than the regular season, then the regular season really doesn’t matter a whole lot other than for seeding purposes, does it?

The point of the Flex System is to have an exciting conclusion to the season in the form of a playoff rather than a poll, but to also restrict entry into that playoff to teams that have a legitimate claim to have had the best season, which honors the regular season.

by billyzane on Dec 4, 2008 9:16 AM CST to parent up reply reply   0 recs

Flex appeal

Don’t get me wrong. I think the flex is a great idea. But isn’t it inconsistent to complain about the playoff system and then reward the teams having the best season with a playoff? Maybe what you really want is the NIT like back in the olden days (post season, top teams). And how about a separate honor for “best team” awarded after the tourney?

Before you criticize someone, walk a mile in their shoes. If they get mad, you're a mile away AND you have their shoes.

by Caradoc on Dec 4, 2008 1:19 PM CST to parent up reply reply   0 recs

This is all addressed in the Flex System articles.

But briefly, the goal of a large playoff in which lots of teams who clearly don’t have a claim to have had the best season overall are involved is very different than the goal of a smaller playoff which includes only teams which have a claim to have had the best season. The latter type (which includes the inflexible BCS and the more nimble Flex System) aims to say, “This number of teams has a claim to have had the best season and each could thus potentially be considered the national champion based on the regular season. Because we want to crown an individual national champion though, we need some way to differentiate between these teams. The most exciting way to do this is by allowing them to play each other and we’ll declare the winner the national champion.” This is very different from saying, “we’re going