Prohibiting computer rankings from considering margin of victory is a mistake that could produce "crazy'' results, claim two men whose calculations are part of the Bowl Championship Series formula.

The BCS is expected to announce a change that would require computer rankings to eliminate margin of victory from the calculations. Those who decline to conform to the new rule would be dropped.

Herman Matthews, whose ratings are syndicated by Scripps Howard News Service, said he replied to a recent request by a BCS representative to calculate ratings for college football teams if victory margin were eliminated from his formula.

"Look at the nonsense here,'' Matthews said over the weekend, disclosing the hypothetical ratings he sent the BCS office. "If you take out margin of victory, Tennessee would have been No. 2 before the bowls even though they got beat by LSU. That's really suspect. Then Tennessee slaughtered Michigan (in the Citrus Bowl) but the Vols would have dropped from No. 2 to No. 3 while Michigan increased from No. 25 to No. 20. That's crazy.''

David Rothman, a California statistician, says if the BCS goes through with the change, he won't be "brow-beaten'' into changing his calculations. As a result, Rothman's college football computer ranking is likely to be dropped by the BCS. Several more of the eight computer rankings could also be scuttled.

Rothman, Matthews, Jeff Sagarin (USA Today) and Peter Wolfe used margin as victory as a component of their calculations last season. Jeff Anderson & Chris Hester (Seattle Times), Wes Colley (Atlanta Journal-Constitution), Kenneth Massey and Richard Billingsley did not use margin of victory.

The issue came to a head because of the significant difference in Oregon's ranking by the computers. The Ducks were ranked third, third, third and second at the end of the regular season by those who did not use margin of victory. Oregon was sixth, seventh, seventh and eighth in the four computer systems that took into consideration margin of victory.

"I think they are making a terrible mistake if the BCS drops margin of victory, and I am unwilling to change'' Rothman said from his home in suburban Los Angeles. "If Team A beats Team B 99-0 and Team C beats Team B 7-6, they are saying Team A should be rated the same as Team C. That's absurd, flat-out absurd.

"My guess is that the three oldest guys who do the rankings -- me (age 66), Matthews (72) and Sagarin (54) -- will not cave in.

"You know the irony of the whole thing? After Nebraska got creamed by Colorado (Big 12 championship game), Nebraska dropped more in the polls (to No. 4 in both) than they did in the computers (2.17 average). So the voters in the polls put more emphasis on margin than any of the computers do.''

Nebraska finished No. 2 in the final pre-bowl BCS standings and Colorado was No. 3, but the difference was only five-hundredths of a point. If Colorado had moved up only one spot in any of the computer rankings, the Buffaloes would have jumped past the Cornhuskers to earn a spot in the Rose Bowl.

At the request of the Knoxville News-Sentinel, Rothman recalculated his 2001 rankings by changing only one score from last season: Colorado's game against Oklahoma State.

"Colorado won that game 22-19, but if Colorado had won 31-19 they would have moved a tiny smidgen ahead of Florida from No. 5 to No. 4 in my rankings,'' Rothman said. "That's the only change it would have taken for them to get enough BCS points to get in the Rose Bowl. That's even less of a difference than I had thought it would need to be to change everything.''

In Rothman's rankings, all games count equally regardless of when they are played and there are diminishing returns on scoring margin. Tennessee was ranked No. 1 by Rothman at the end of the 1998 regular season before the Vols went on to defeat Florida State for the national title. In 1999, then BCS coordinator Roy Kramer added Rothman to the mix when the number of computer ratings was expanded from three to eight. Now as the tweaking continues, it appears the number of computer rankings used by the BCS will be reduced for the upcoming season.

The computer component of the BCS formula is calculated by averaging six computer rankings. The highest and lowest ranking for each team is thrown out.

Rothman said if the BCS really wants meaningful reform it would eliminate the clandestine aspect of the system -- i.e. formulas to rate teams are kept secret by some of the computer directors.

"Right after all of the games are over on Saturday night, people should be able to test on their own home computers what will happen as a result of that week's games,'' Rothman said.

"But the BCS won't do that because they want to be able to issue a big press release on Monday making a big deal out of the standings.''

Rothman said he had sent Kramer a rough draft of a speech he will give to a professional statistical group in New York City on Aug. 12 citing reasons why margin of victory is essential to his college rankings.

"When they get in trouble in the BCS, they always go back to Kramer and he's always the one lurking in the background on this thing,'' Rothman said.

Matthews declined to say if he would drop out of the BCS process if he were told he must eliminate of margin of victory as part of the formula. He remains hopeful Kramer will step in at the 11th hour as a moderating influence. If Sagarin remains firm about sticking with margin of victory and is forced out of the process, the BCS computer rankings could lose credibility because of his large following in USA Today.

"I plainly told the BCS people it is a mistake to change it,'' Matthews said. "They are grasping at straws. These are all scholarly men who do the ratings and it doesn't make sense for the BCS people to tell them how to do things.

"I think you'll see a wholesale change of people getting out of the ratings. That's going to make the BCS worse, which will play right into the hands of the people who want a playoff system.''

Contact Gary Lundy of The Knoxville News-Sentinel in Tennessee at http://www.knoxnews.com.