THE BCS BLOG = by K. D. Keep

While it is easy or popular to bash the Bowl Championship Series, we must agree on exactly
what it is.   It is essentially a Bowl Choice System run by the bowl committees and the 11
1-A conferences.   Under the current setup, there will always be one or more universities
to believe their football teams have been jilted.   Invariably, they will point to a certain
aspect of the BCS formula being the culprit.   Many changes have made to the formula in
response to these complaints, and the BCS should be applauded in addressing them.   However,
there seems to be a never-ending parade of incredible circumstances which bring forth a
new set of complaints every year.   As I wrote last year:  How many more tweaks and fixes
are in store for the BCS before they blow it up?   There are still some inherent flaws in
the system to be dealt with.   In that any kind of playoff is still nowhere in the future
(many thanks to Fox Network fer cripe sake), the BCS formula remains far from credible.

While I openly admit a slight prejudice where the computer raters are concerned (as I would
dearly love to have my system included), there are some obvious problems with their use.
In the beginning, the BCS included computer rankings to offset or balance the "human error"
aspect of the wire polls.   They thought that by removing the opportunity for bias or any
subjective process from the computers would result in a "fair" compilation in the end.
But, since the mandated removal of scoring-margin from the process, the computers have been
unfairly constrained to the use of won-lost record.   It was thought to be a deterent from
teams running up the score and skewing results, and further, the BCSters believed a fairer
judgement would result.   (Did they bother to tell the coaches or writers to ignore the
final score?  Nah.)   This restriction harms the formula in a couple of ways.   First,
the rankings produced by these systems greatly inflate the rating of non-automatic
conference teams (remember Miami-Ohio ranked 4th? really now...).   Or that teams who
suffer a crushing loss in their last game, yet are still locked into the championship game
(you remember who they were).  These outputs have obviously conflicted with those of the
human polls, yet only the computers were taken to task.   Well, of course their mission
for accuracy was compromised.   Second, by limiting the method for arriving at the rank,
the opportunity for a well-rounded and truly objective measurement is lost, by excluding
other methods and excluding a greater sampling of methods.   Just as 114 votes would be
more representative than 11, say 30 computer systems would represent better than six.

In 2003 there was a tremendous outcry through the land that both wire polls had a team
ranked Number One (USC), yet they were BCS ranked third BECAUSE OF THE EVIL COMPUTERS !
So the BCSters waved their magic wands, and behold, the polls were now twice as important
as the computers.   Then in 2004 we saw the hand-wringing and gripes from coaches and fans
alike over the poll voting policies, so the Associated Press said "can't touch this"!
In replacement we are subjected to the Harris Interactive Poll comprised of voters who are
allegedly in touch with the college football scene.   Yes, and my wife can pick-em just
as well by simply looking at the uniform colors.   The HIP may have replaced the AP in
the mechanics of the BCS, but the AP still has their poll, pre-season and all the way
through, and still gets as much attention as ever.   In the event that the Associated
Press votes a team number one and is not in the BCS Championship, will anyone exclude them
from a share of the national championship?   I don't think so.   Plus, from what we know
about the AP, how likely are they to vote a different team number one just to stick it to
the BCS?   In either case, the BCS champion will be strictly mythical having relied on the
polls to get there, as it has always been.   As long as voters have the major say in who
goes where, they should be verified for credibility and held accountable for their votes.

I said that last year the BCS made positive changes in disposing of the awkward sub-ratings
involving quality wins and strength-of-schedule, and refining the vote count by percentage.
They increased the ranking accuracy while also simplifying the process.   I also thought
that averaging the polls 2-to-1 over the computers was a good thing until 2004's voting
fiascos.   They are on the right track, but more needs to be done.    All the components
involved must be accountable with the maximum number of rankings possible.   I would like
to see all 119 Division 1-A coaches voting with the same number of voters in the Harris.
I would also like to see 24 computer ranking systems employed.   All voters must certify
their ballots with signatures.   All computers must submit an affadavit as to the sources
or reasoning behind their ratings (without giving the exact formula).   These computers
would be selected from the Massey Consensus (12 best from Top 25 ranking violation for
retrodictive ratings)  and Todd Beck's Prediction Tracker (12 best straight-up records).
As long as the computers are subjected to best-and-worst ranking removal, the polls
should have a similar percentage of their best-worst thrown out.   Increasing the ranking
sampling list will decrease the chance of skewed results.   Rules should be same for all.

As it stands now, there is only one meaningful relevant bowl game being played, the BCS
Championship game at whatever venue has it this year.   Schools, fans, bowl committees,
TV networks and sponsors are growing ever more impatient and dissatisfied with the bowl
matchups and interest in them.   To date the BCS has accomplished only two things:
(1) A system involving all the major conferences and bowls to avoid a split championship,
or so they thought.  (2) An incredible television package and dollar payout not nearly
possible under the old system.   Now is the time for the NCAA and school administrators
to advance into the 21st century and set up a post-bowl playoff to determine a true
national champion.   All of the bowl games need to become relevant.   This can be achieved
by taking just one simple and painless giant step.   A one-game playoff for the BCS
Championship on the third or fourth Saturday in January.   This game will be held at a
site to offer the highest bid for payouts.   (These payouts would be distributed equally
amongst the conferences per agreements.)   Participants are selected by a final BCS Rank
taken after the bowl games are all played.   Bowls would be open to invite or contract
with teams or conferences as they wish.  I would automatically qualify the six highest
ranked conference champions to four BCS bowls along with two highest ranked at-large teams
not winning a conference.  The fifth BCS bowl (presumably Cotton or Orlando) would host
the seventh and eight ranked conference champions.  The remaining bowls would be open to
a larger pool of teams from contracts with more than two conferences each.   This will
allow for fresher and more interesting matchups.   All of these bowls would be important
because the final BCS standings depend on their outcomes to finalize each team's strength-
of-schedule.   Actually, I believe the best scenario is a four-team playoff after the
bowls, but this one extra game would be a monumental step in the right direction.   If
the BCS continues to elicit more criticism and controversy instead of compliment, then
let's go the one step further instead of blowing it up.   Tell me what you all think.

The Bowl Champ Series bosses need to settle on a formula and go with it.   In my opinion,
they will continue to lose credibility by making major changes every year in response to
some team's wailing.   They will further gain credibility by fortifying the ranking
process and eliminating many of the automatic qualification rules.   They seem to revel
in the participation of coaches, media, and computers, and the interest they generate
throughout the media and fans.   If the BCS wishes to continue that participation, it
would serve them well to make these refinements.   Equalize the polls, certify the
coaches and Harris ballots (no voting before October).   Expand the computers, both in
number and computing methods.  Highest-and-lowest rankings tossed out of human polls by
similar percentage for computers, or (preferably) none at all.   Refresh the stale bowl
selection system, and add one more game to instantly raise the importance of many bowls.
(I would love to see a four-team playoff hosted by Orlando & Tampa each year.   Disney
could be the prime sponsor/host.  Held on the third and fourth Saturday in January there
would be ample opportunity to sell tickets, make plans, fill packages, host players and
fans.  Just think of it:  The Interstate 4 Final Four !)   To BCS or not to BCS, that 
is the question......Let your opinions be heard!