Wednesday, May 27, 2009

Coaches Poll Confidential by 2010...

Beginning in 2010, ballots in the final regular-season USA Today coaches' college football poll will be kept confidential -- one of a handful of changes on tap for the poll that helps decide who plays in the BCS national championship game.

The American Football Coaches Association had asked Gallup to study its poll and recommend how to make it more accurate and credible. The AFCA's board heard the results in early May and announced them Wednesday in Waco, Texas.

Gallup recommendations being considered for the future include reducing the number of teams ranked from 25 to 10 or 15, and evaluating the merit of a preseason poll.

Starting this year, the poll will also eliminate bonus voters given to some conferences based on how their teams did the previous year.

The AFCA, and Executive Director Grant Teaff, spoke on the issue...
"...the trustees wanted to make it even better by reviewing all of the elements and methodology involved. We commend the Gallup Poll organization for its contributions and the professional way they approached this project. It resulted in providing the AFCA with unbiased and objective recommendations related to the way in which the poll is used to determine the nation’s top 25 collegiate football teams.”

Teaff indicated that some of the recommendations are being put in place immediately and one will be delayed for a year, while others will be taken under study.

The most radical change, however, will to be to return to the policy in place before 2005, when coaches didn't have to reveal their final ballots. Coaches will be allowed to release their own ballots if they choose.

Revealing the ballots has made for some awkward situations. Former Florida coaches Steve Spurrier (now at South Carolina) and Ron Zook (Illinois), for instance, took some heat last year when they ranked the Gators second behind Oklahoma in last year's final regular-season poll.

Zook, meanwhile, got lambasted two years previously when he picked the Gators as No. 1 over Ohio State, which is in the Big Ten with Illinois.

Zook said whether the poll is confidential or not, his method remains the same.

"To me, I was always going to vote how I felt," he said. "I think that's why you have a poll. That's why more than one person is involved. So what I try to do is rank the teams where I really feel they should be. I'm not real into the political stuff."

The AFCA also decided to continue allowing coaches to vote for their own teams and to select voters on a random basis beginning this year.

Here's the list of recommendations in full...

• Select voters on a random basis each fall.
• Eliminate a ‘bonus voter’ system based on ranking success of a conference’s teams in the previous year’s poll.
• Adopt the ‘round to even’ methodology for conferences with an odd number of institutions. The number of voters for each conference will continue to begin with 50 percent of the total of member institutions. The ‘round to even’ is the standard method used in many fields, including the business and scientific communities, and recognized by the American National Standards Institute. (Examples: Eleven teams: fifty percent = 5.5, rounded to 6. Nine teams: fifty percent = 4.5, rounded to 4).
• Continue the option of coaches being allowed to vote for their own teams.
• Do not release the individual coach’s ‘final end of the [regular] season’ ballot. Gallup recommended the voting process remain confidential. Historically, until four years ago, the AFCA kept the ballot confidential. (The AFCA does not restrict a coach from releasing his ballot). The AFCA Board decided to delay the implementation of the confidential ballot for one year, until the 2010 season, to coincide with the current BCS bowl cycle.

Among the Gallup recommendations that will be under study for the future:

• Reduce to 10 or 15 the number of teams ranked.
• Evaluate with other shareholders in college football the value of a preseason poll.
• Develop an online process for capturing coaches’ votes that would assist in improving the accuracy of the rankings and decrease USA TODAY’s review time for each vote.

Got all that...???

No comments: