The People's Voice builds a better mousetrap

Dan Wetzel

With the BCS bowls finally upon us, we decided to give our readers the space to air suggestions on how to solve the great college football controversy.

There are about a million solutions that poured in, and most of them sound better than the current system. So below are some of them, ranging from the serious to the, well, you'll see.

We will return next week to your complaints and compliments from my regular columns.

Thanks again for all the input. I read everything, and continue to be humbled by the outpouring of passion. Remember, keep your submissions short and include your name and town if you want to get printed. My comments appear in italics.

Now on to the People's Voice . . .

BCS debate

I am amused by the situation with the BCS. I have always been of the opinion that the human polls were much more "accurate" in determining the best teams in the country than any computerized system.

The reason for this is very simple ... the human mind, without quantifying them per se, takes into account thousands of variables before coming up with a "ranking." The BCS computers can only take into account the limited set of parameters that were programmed into it.

Despite what the BCS people might tell you, the BCS is designed to identify the two teams with the two best SEASONS ... not the two BEST teams at the end of the season. These are two different objectives and have little correlation with each other.

Let me explain by example ... let's say two teams (A and B) have identical 6-5 records at the end of the season and the strength of schedule is identical. Taking out the human poll factor, the BCS computer will rank these two teams equally.

Let's say, however, that team A wins its first six games and loses its last five. On the other hand, let's say that team B loses its first five games, and wins its next six. While they both have identical records (seasons), which is the superior team at the END of the season? I would argue that the team that won its last 6 games is the better team.

While the BCS computers cannot differentiate between these teams, and hence would rank them equally, it would be the human polls, ironically, that would ultimately differentiate the two teams in this hypothetical scenario.

WHEN you lose and HOW you lose matter. An 11-game season is a set of 11 "events." The further out in time you go, correlation diminishes. In other words, game 1 and game 2 are more closely correlated than game 1 and game 11. In an extreme example, let's say that a "season" consisted of 1,000 games. I think you can see what I'm getting at. The human mind can take into account losses, the timing of those losses, and the circumstances of those losses.

The drawback with human polls is that no two humans weigh the factors equally. Human bias will also come into play. In spite of that, until a playoff format is adopted I would still rather rely on human polls.

Football is played in four dimensions (space-time). The BCS computer can only process information in two. It's missing the key variables of depth and time.

Luke Chun
San Francisco, Calif.

I can only say that this is as brilliant of an explanation of the shortcomings of computer formulas as I have ever read. I completely agree. And if I didn't, I still wouldn't dare argue since you are clearly a lot smarter than I am.

I agree that a playoff system is necessary to determine the true No. 1 team in the nation. However, I also believe that bowl games are entertaining and bring great matchups to various parts of the country.

I would propose a playoff system combining both, using bowl game sites (even names) for single games in a 16 or 32-team field. First round games would be hosted by lesser bowls such as GMAC and Gaylord Hotels, while the "Final Four" games, including a 3rd place game, be held in the original BCS Bowls (Sugar, Orange, Fiesta, Rose).

Gary Silverman

There are two simple reasons the BCS is not giving you the results you want: 1) margin of victory is not allowed as a consideration in the computer ratings, per a BCS committee decision two years ago and, 2) recent games are considered as no more important that earlier games.

If you have problems with the BCS, I suggest you advise those changes, not complain about how its goal is to hoard money and provide chaos.

Bapi Gupta
St. Louis, Mo.

There are two possible commonsense ways to reform the BCS system for determining the National Champion:

1) Keep the current system, but for determining the two teams that play in the National Championship game, use the BCS ranking ONLY as a tie-breaking procedure to decide between teams that are ranked either one or two in one poll (AP or coaches), but not the other. This would put the primary decision-making back in human hands, leave us resorting to the computers only when absolutely necessary, and avoid the fiasco that is likely to occur this year with clearly-deserving USC potentially being shut out of the Championship Game by the computers.

2) Modify the current system by adding a four-team mini-playoff to the BCS. Use the BCS rankings to choose the top four schools, with No. 1 playing No. 4 in one designated January 1 bowl and No. 2 playing No. 3 in another January 1 bowl. Then have the winners meet in a true Championship Game one or two weeks later. This allows the championship to be clearly decided on the field, while minimally changing the existing schedule and structure.

Jeff Staudinger

Time for the Supreme Court to step in, after all, this may be a national tragedy if the Men of Troy don't go to the Sugar Bowl.

Greg Schoenberg
Phoenix, Ariz.

Let's get that guy with the big eye from Florida who examined the butterfly ballots.

Amy Bloom
Chicago, Ill.

Wouldn't Katherine Harris be a fine arbitrator for this?

A.J. Gansey
Anaheim, Calif.

Interesting option, though I am not sure she could imagine the pressure – and inevitable fallout – of working on something as important as choosing the college football champion.

Why don't they just take the top six teams in the poll at the end of the year and let them play for the National Title? Then, let the rest of the teams fight for other bowl positions. That would take care of the BCS point spread to see who is No. 1, and has the right to play in the title game.

Brian DeBose

For what it's worth, I came up with my own NCAA tourney format that still keeps the BCS math in place, keeps the schools in the money, keeps the romance of the bowl games intact. A rough format, but something that might work.

First, all NCAA division one games must be completed by Thanksgiving Saturday, all teams will be required to honor an 10-game-max schedule that must be concluded by this time. BCS rankings at this time are final.

Take 2002 as an example. The Fiesta Bowl (this would rotate as is now) gets the title game between the two teams left in tourney, played New Year's week. Sugar, Orange and Rose Bowls would get the losers of tourney.

Tourney set up as follows.

1) Week One. First Saturday in December. The top eight teams in the country qualify for title playoffs – 1 vs 8, 2 vs 7, 3 vs 6, 4 vs 5. The games are played at the top-four schools in rankings. The four losers of week one will be pitted against each other in two of four big bowls (i.e. Sugar and Orange).

Week Two. Second Saturday in December. The remaining four schools are re-ranked according to where they were ranked at beginning of tourney. (In essence if #1 gets beat first week and #2 wins, #2 becomes #1...) 1 vs 4, 2 vs 3. The host teams are drawn out of hat this time and the two losers will play in Rose Bowl.

Week Three. New Year's Week. National Championship Game at Fiesta Bowl. The loser is guaranteed No. 2 status and both schools get a big purse financially. Number one is crowned.

The benefit to this in my opinion is that you are allowing the momentum from the regular season to continue for the top eight teams. Anyone who stubbed their toe can reconcile and be redeemed and no long layoffs. This is important because a Nebraska, a Penn State, USC, LSU, FSU, Michigan or a Tennessee etc. which might have a loss, still has the momentum with something to prove in tourney.

The revenue piece of a conference championship game is certainly a question. Not all conference championship games are guaranteed money, TV interest etc. Playoff games would no doubt have a bigger impact to the schools and networks.

I would like to have an academic slant to this argument, but in my opinion, the coaches, the athletic directors and the chancellors are all in the football business. They will make provisions for a kid to complete his finals, just a question of when.

I know you have been down this road several times and maybe I am missing something, but something has to change.

Mark Starr
St. Louis, Mo.

I'm a high school mathematics teacher and I'm ashamed of the BCS. I've devised my own system which is, in my opinion, 100 times more fair (and mathematically sensible) than the BCS.

It all started a couple of years ago when Nebraska got the nod instead of Colorado. I wrote an article and in it I show several ways in which the BCS is inherently flawed, and which teams should really have played who that year (CU vs. Miami).

I did the same figuring for OU, USC, LSU (the top three one-loss teams) and threw UM in just for ha-has. The end result was that OU (3-1 against ranked opponents) and USC (2-0 against ranked opponents) both played too weak of schedules compared to LSU (4-0 vs ranked opponents) and UM (5-2 vs. ranked opponents) to earn a berth into the Sugar Bowl. Nick Saban's LSU Tigers vs. the University of Michigan?

Sounds crazy, but it makes the most sense to me.

Daniel J. VandeBunte
Hobart, Ind.

LSU vs. Michigan in the Sugar Bowl? And you thought we had a controversy now?

Call me old-fashioned but I liked the old system. There was nothing wrong with a "mythical national champion." More bowl games really mattered and most year's there was no debate, one team stood out.

Sam Vetter
West Hartford, Conn.

I have an idea for the "wonderful" BCS system that actually includes a playoff format: Create a 16-team playoff involving all 10 conference champions and 6 at-large teams as derived from the BCS formula. Any team not selected for the playoff and having the minimum number of wins would be eligible for some of the bowl games. Other bowl games would be reserved for teams that get systematically eliminated from the playoffs (these would be higher tier bowls like the Citrus, CapitalOne, etc.).

The further a team goes in the playoffs, the better the bowl game. The Rose, Sugar, Fiesta, and Orange Bowls would, on a rotating basis, get the quarterfinal losers, the semifinal losers, and the National Championship. This way, everyone should be happy: The mid-majors get a chance at the title, the people who want a true champion will get a real title game, and the colleges will make an even greater sum of money due to the increased revenue of playoff games on top of the Bowl Game payouts. Only in a perfect world... .

Matthew Duff
New Richmond, Wisc.

Keep the BCS format, just make it a playoff.

You have the four major bowls rotate the final, the semis and one quarterfinal; and you have the other six "big" bowls rotate the other three quarterfinals. I'm sure the Gator would rather have a game like Michigan-FSU with national title implications than West Virginia-Maryland. You could keep the same amount of bowl games, and yet the lower bowls would have more quality teams, since teams that are now playing in the big bowls that would be quarterfinal games would be available for the other bowls.

This year's games:

Cotton/Holiday Bowl: Oklahoma-Miami
Outback/Gator Bowl: Michigan-FSU
Peach/Capital One Bowl: LSU-KSU

Orange Bowl: Winner Rose-Cotton/Holiday
Fiesta Bowl: Winner Outback/Gator-Peach/Capital One

Sugar Bowl: Winner of Orange and Fiesta

Alex Sosa
San Pedro Sula, Honduras

I just like that someone in Honduras is worried about this.

I say keep the BCS. Let the computers determine who the top eight teams in the country are ... (there is some validity to the system) then institute an EIGHT team playoff of the best EIGHT teams in the country. It might leave out the TCUs and the Boise States, but it is better than a one-game championship where the contestants are picked on tenths of points, rather than "first and tens" and "touchdowns."

Jonathan Swenson
Logan, Utah

Steps to crowning a college football national champion:

1) Have each D-I team play five games.
2) Pair up all 5-0 teams.
3) Have the ADs of those schools meet, and, on ESPN, flip a coin. The winner of the coin flip moves on.
4) Repeat until just two schools remain.
5) Have those two schools play in a "lucrative national championship game."

Hey, it may not be perfect, but it sure beats the BCS.

Joshua Carey
Candor, N.Y.

Solution for the BCS mess:

1. Remember there will always be controversy. (The school left out in 5th, 9th, 17th, or whatever place will always be slighted.)
2. Eliminate the BCS.
3. Eliminate all polls.
4. Eliminate all sportswriter's opinions since they aren't worth more than that of the guy next door.
5. Watch the bowl parades, regattas, etc., and games! (Party time in midwinter, anyone?)
6. Then we can all enjoy the entertainment along with the controversy over who is the "true" Number 1 team. Or did I already say that above?

Oh yeah,

7. Watch the sun come up the morning after the last bowl game in January.

Ron Columbo
Chandler, Ariz.