Transparency needed in new BCS

CHICAGO -- It's a bit ironic to see college football's power brokers walk out of historically significant meetings held behind closed doors and then talk up transparency.

While the negotiations about the sport's future postseason structure remain very much a classified event -- confined to hotel meeting rooms in cities like Chicago, Dallas and Hollywood, Fla. -- conference commissioners are in agreement that whatever model is chosen must be more open to the public and easier to understand. These men might not agree on much right now, but they stand together on one item.

"Any improvements we can make in terms of transparency will be welcomed by everybody," ACC commissioner John Swofford said following Wednesday's meeting here.

"Transparency is something we would like to see happen down the road," SEC commissioner Mike Slive said.

Notre Dame athletic director Jack Swarbrick added that the need for greater transparency has been "uniformly embraced." Pac-12 commissioner Larry Scott said the topic frequently surfaced in the most recent BCS meeting.

"Whatever we do," Scott said, "that's what we'll aim for."

Since its inception in 1998, the BCS formula has sparked as much confusion as outrage. Most understand the basic makeup of the BCS standings -- one-third coaches' poll, one-third Harris Interactive poll, one-third computer average -- but the specifics are murky.

The computer rating systems are particularly mysterious, which makes their varying results all the more head-scratching. Five of the six proprietors of the systems don't make their formulas public (not even BCS officials know them), in an effort to protect their patents. Consequently, there's no explanation when Oklahoma, buoyed by a seemingly indestructible computer average, remains No. 1 in the final BCS standings in 2003 despite a 28-point loss in the Big 12 championship game.

In 2010, the final BCS standings had to be revised after independent analyst Jerry Palm spotted an error in Wes Colley's computer ranking, the lone metric made public. Colley had failed to include the score of an FCS game in his data, which resulted in teams incorrectly listed at both No. 10 and No. 17. If similar mistakes occur in the other five computer systems -- ones that could impact teams much higher in the BCS standings -- they likely would never come to light.

The coaches' poll, meanwhile, debuts before the season actually kicks off, includes voters with fairly obvious biases and doesn't publicly reveal its ballots until after the regular season.

According to a research paper by professors from Yale and UC-Santa Barbara published last fall in the National Bureau of Economic Research, the coaches' votes are not only biased toward their own teams and teams in their leagues, but the biases are accentuated when ballots aren't disclosed and lessened when ballots are made public. As one of the paper's authors, Yale professor Matthew Kotchen, told the Yale Daily News, "It is hard to come up with a reason for why these ballots should continue to be confidential."

Before 2005, the coaches never revealed any of their ballots. They tried to stop making the final ballots public in 2010 before reversing course.

"On a scale of 1-10, the transparency we have now would have a hard time getting above 3, 4, 5," Big Ten commissioner Jim Delany said. "We're not sure what's in the guts of the computer programs. We don't know why people rank teams where they rank them before they've ever played a game. It'd be nice to know how you can come up with a ranking without respect to a game. Maybe that's not an issue of transparency, but an issue of rationality."

Or an issue of insanity.

Delany has been a prominent BCS supporter over the years, and he represents a league whose presidents recently stated they prefer keeping the status quo in the postseason. But with change on the way and a four-team playoff generating the most discussion, Delany has been pointed in his criticisms of both the polls and the computers.

He also has been the most vocal commissioner about the need for transparency in whatever new selection system is used, whether it's a selection committee, a computer system or some combination.

"Transparency comes in the form of knowing what the standards and instructions are with the committee," Delany said. "It comes in the form of knowing what's inside the guts of the computer program. And it comes in the form of a chairman coming forward and saying this is why we did what we did. You might disagree with it, but it's the human face and the human voice to explain.

"There's no one to explain why Oregon would be out and Stanford would be in if you used those polls as a proxy for the top four [in 2011]. There's no one to explain how Oklahoma could lose in the [Big 12] championship game and still be the No. 1 team in the country [in 2003]."

Delany's preference is for a selection committee with specific guidelines, known by the public, for choosing the playoff participants. Committee members could use "any number of helpful kinds of analysis," Delany said, as long as those tools can be explained to the public.

The NCAA men's basketball tournament selection committee, for example, uses the Ratings Percentage Index to help choose at-large teams every year. The RPI formula is no mystery: 25 percent winning percentage, 50 percent opponent schedule strength, 25 percent opponents' opponent schedule strength.

"You know what's there," Delany said. "You can argue whether or not it's properly weighted, but there's no misunderstanding of what's there.

"It's OK to look at an RPI," he continued. "It's OK to look at an AP [poll]. It's OK to look at a coaches' poll. But none of them, by themselves, or even in some sort of combination, provide an appropriate proxy without a human judgment."

The selection committee concept seems to be gaining traction in the BCS meetings.

"We've got several commissioners that have served on the basketball committee, and there have been some positive lessons learned about the need for transparency in this day and age," Scott said. "It's been a positive discussion."

The basketball selection committee deliberates outside the public eye in a heavily secured section of an Indianapolis hotel each March. Would a football committee take the same closed-door approach, especially after so many have cited the BCS for a lack of transparency?

Judging by the passion of college football fans and the interest in the decisions, a committee might want to ask President Barack Obama (a playoff proponent, by the way) if Camp David is available.

"You can't have an absolute in this area," Delany said. "You're going to have to have your deliberations, and they're going to be had in a place and in a way that allows people to be candid and not held captive to every word that's said. There are levels of transparency, but when you look at what we have now, we can compare it to what could be."

What it could be undoubtedly will be a significant improvement on the current structure.

"Any time you have transparency, you have simplicity in terms of fans' capability of understanding it, and that's a plus to the system," Swofford said. "Any time it's a system people don't understand and don't trust, that's problematic."

There are more closed-door meetings on tap this week in Chicago and next week in Washington. Eventually, a postseason model will emerge, and when it does, college football's kingpins are prepared to pull back the curtain.

"Lessons," Scott said, "have been learned."