FanPost

Continuing on some (Playoff) Thoughts...

First, let me say that the replies to this diary have been epic.  I've got a few things to say about whills', burnt in ny's, and Red Blooded's beasts.  If you don't want to read all of that first, the short of it is we're talking about the BCS vs. some kind of playoff system.

One, part of the problem with going to a big (or even mid-size) tournament, is that with each round you add to the post-season, you depreciate the value and thus lower the stakes of each individual regular season game.  I'm sure most of you are aware of most of the arguments of this nature, so I'll just leave that as a reminder.

Two, BZ (the almighty conqueror of statistical football and related things) is right, in his Flex System argument, about the varying need for a playoff.  Texas vs. USC was the BCS's raison d'être.  If so, then why not use the BCS to decide the champion in that case?  The BCS is perfect for choosing between two obvious, uniquely-undefeated contenders.  Why place more burden of proof than necessary on the champion?

If you're looking to determine a champion, you need to balance between choosing the "best team in the country" and choosing the team that "played the best season in the country."  By the first standard,  LSU probably would be #1 even after that Kentucky loss (The record doesn't count, all that matters is the team's potential, and this LSU team would probably beat any other team most of the time).  By the second, South Florida should probably be #1 (look at PB's BlogPoll ballot).  In my opinion, you should look first at who's played the best season, and then, if that doesn't decide it for you, you go to who is the better team.  That's mostly what the old "Polls 'n Bowls" system did, but it decided "who was the better team" by pollster opinion, instead of head-to-head competition.  

The BCS was an attempt to introduce head-to-head competition as the method of deciding between equal-resume teams, but it assumed that there would only be two teams with legitimate claims to having the top resume.  The problem people have with the BCS is that the floor is sometimes too high.  The third place team often has too much claim to the national championship for people to accept two other teams fighting over it alone.

Red Blooded pointed out, in a comment, that in a 16 team tournament world, if the 16th ranked team were to run the table in the tournament, no one could argue that the seventeenth ranked team had more right to the national championship.  True.  The problem is, fans of the seventeenth ranked team will not argue that they deserved to be national champs, they will argue that they deserved to have the same chance as the sixteenth, and they're basically right.  As you go deeper into the rankings, the uncertainty and dispute in a team's ranking rises as well.  You can easily see my point by glancing at The BlogPoll.  Notice the standard deviations (for anyone who isn't familiar, the standard deviation basically measures how spread out a team's ranking was on the individual ballots, higher StdDev means more dispute on where they belong).  They obviously increase as you move down to about number 15, then they decrease again because ranking someone 26th or 50th or 100th doesn't show up as a difference in opinion (if you had to rank all the teams, the StdDev would increase as you moved down until probably a little after halfway through the rankings).  This means that as you include more teams, you're really just increasing the probability that some fluke team will get lucky and win the national championship, (causing all the schools on the bubble to whine about not getting the same chance) more than you're increasing the probability that a national champion will be less disputed.  It probably won't get as much attention, because these lower schools obviously aren't as good as the higher ranked ones, (that's why they're down there in the first place) and the anger is more spread out, but overall, more people are pissed.

What you've got to do, then, is let teams into the tournament based on their claim to the top resume, then let head-to-head competition act as your Power-Poll.  If you think about it this way, the legitimacy of your claim to the national championship (at the end of the regular season) decreases as you move down the rankings, as well as the uncertainty in your claim's legitimacy.  That means that as you move the cutoff down the list, bubble teams have less and less of a case for being let into the tournament, and we (the general public) are more and more sure of that.  That is the only way to select tournament participants because it naturally suggests a cutoff.  Handing out "chances at the title" is not ideal for the same reason celebrities don't like to give out autographs:  For every one you give out, it becomes harder to justify not giving out the next one.  

If, then, we are to allot tournament berths based on claim to the best resume, rather than claim to being worthy of a  chance, this justifies cutting it off higher, rather than lower, and supports the use of the Flex System to determine a national champion.  At the end of the season, we should determine up to four national championship "candidates" (by something like BZ(TACoSFaRT)'s rubric from the Flex System), which will then play a tournament amongst themselves to decide the overall champion.

This system is not perfect.  It is not perfect in any way.  If it were perfect in one way, it would be horrible in the other ways.  If it were based purely on finding "the best team" it would be a power poll after we watched all of the teams practice in the fall (Hey, it would cut down on ticket prices, at least).  If it were based purely on finding "the best resume" it would be very similar to the "Polls 'n Bowls" system, but with 7021 games  (119 plus 118 plus 117...  We would be playing this season until the sun burned out and the universe collapsed...  No, but really, the season would be almost five times as long.)  If you think a tournament eliminates controversy, you could have a 120 team, 7 round tournament be half the season (keep in mind that lower-tier schools will have trouble paying for airfare on one week's notice for seven weeks, not to mention Hawaii).  This system concedes a small amount of controversy (whatever claim the 5th team has to the NC), in order to shorten the postseason and minimize the impact on the importance of the regular season games.  It's a good balance, and the best way to satisfy as many people as possible.

All comments, FanPosts, and FanShots are the views of the reader-authors who create them.