Here is what I would like to see happen for a college football tournament selection committee:

Form a committee of 25 potential voters from all over the country with varying professional backgrounds (former coaches, former players, admins, writers, TV guys, etc.). I'm a big tent kind of guy. Invite all of them to participate. I would even allow for a spot or two for fans to make the committee. Hell, you could do that via a contest that would be sponsored by Mastercard or whomever the title sponsor ultimately ends up being.

Each week, every single member of the committee forecasts all of the AP Top 25 games, irrespective of the point spreads. You can do it based on their confidence in each result. If anyone has ever played a confidence poll, you will know exactly what I mean. Their prognostications are then published for all the world to see, thus giving the entire process a greater level of transparency.

Starting on the first weekend of October, begin to publish the committee's Top 25 rankings. We will then start to see who knows what he's talking about and who is just guessing. We will also therefore see who is most qualified to pick the teams when it matters most.

Also, each Monday evening, perhaps prior to Monday Night Football, have a half hour or 1 hour television program in which the remaining members of the committee discuss and/or debate their findings and/or trends they are noticing. That would make for great television theater and at the same time it would give the average fan a much greater sense of the committee's mood.

Also in October, and this is key, begin to whittle down the committee based on their prognostication results. Make it a true meritocracy. By early December the committee will have been reduced to 10-15 members, each of whom will have earned his spot at the table, who meet for a weekend to discuss their thoughts. Following that meeting, they have a selection show similar to the college basketball selection show.

It would be phenomenal television.

At this point, people usually say things like, "That is ridiculous! How can you ignore 50 years of work based on five weeks of predicting football games?"

To that I say, tough schitt. Basically what we are doing to start with is forecasting which teams would fare better against the nation's best teams on a neutral field. That is, by its very nature, prognostication. Barry Switzer, for example, may know more about what makes for a good left guard but if that doesn't translate well to predicting how entire teams do then what good does that do anyone? Nobody doubts that Barry Switzer likely knows more about college football than say, Barry Tramel. However, if Tramel has forecast more games correctly up to that point in the season than Switzer, then he gets the nod and it's just that simple.