The American Football College Association (AFCA) decided that transparency in revealing coaches’ voting in its final poll was a mistake. Over the four years coaches’ votes have been publicized, fans have seen a pattern of coaches promoting their own teams, sometimes shamelessly, clear conference bias and some strange votes.
Rather than dealing with these issues, the AFCA, whose Board members make up those coaches who vote in the poll, jumped at the opportunity to hide its participants’ results. Like a rotting, barnacled hulk brought to the surface sinking out of sight again, the USA Today Coaches poll will once again become secret in 2010.
"Historically, we have never released the votes," AFCA executive director Grant Teaff said. "When it came up that, OK, it would be better if you did, I think there was acquiescing by the coaches. As to whether it's helped the poll or not, I don't think I can really say. Whether it's hurt it or not, I don't know.”
The Coaches Poll is one of the three components in the BCS and wants to be the only without accountability and visibility of votes. With millionaire coach’s livelihoods, contract bonuses with performance clauses for themselves and their assistants, and inherent biases, the poll’s credibility has been tested by revealing individual coach’s voting results.
"The perception is that there's a huge bias, and we've never really found that," Teaff said.
You’re not looking very hard, Grant, at your Board members’ voting.
Self-Promoting Their Teams
For the 2008 Coaches’ Poll, I found that twelve of the sixty-one AFCA Board members’ teams finished in the top 25. Of those twelve, nine (75%) voted their team one spot or higher. Four (33%) coaches voted their team two spots or higher.
Frank Beamer was the only coach who voted his team lower (-4 positions) than its final ranking.
Eliminating Beamer’s uncommon vote, the other eleven coaches voted their team higher at an average of 1.8 spots!!
Since Florida’s Urban Meyer and Texas’s Mack Brown could not have realistically increased their respective team’s spot by more than one—which they did, the other eight coaches inflated their team’s ranking by an average of 2.0 spots led by Mike Leach (+6) and Gary Pinkel (+4).
Three other AFCA Board members voted their teams into the top 25, though final voting totals left their teams out of the top 25. Using final total vote rankings, Greg Schiano (+11), Butch Davis (+7) and Bo Pellini (+4) significantly self-promoted their team’s rankings.
Schiano evidently did not consider Fresno State, who beat up Rutgers 24-7, a better team—nor Boston College, Iowa, Florida State or California, all of whom had better records and many more votes.
Including all fifteen AFCA coaches who voted for their teams in the final poll, the average Self-Promotion was 2.5 spots!!
Big 12 coaches (4) won the top spot in Self-Promotion by elevating their individual teams by an average of an overwhelming +4.0 positions, followed by the WAC (1) - +2.0, Mountain West (2) and ACC (2) – +1.5, Big 10 (2), Pac-10 (2) and SEC (2)– +1.0! (If I include Schiano’s voting, the Big East would have taken the top spot with 11.0.)
The AFCA coaches do not feel that coaches should not be able to vote for their own teams.
"By keeping things confidential, I think there will be less hidden agendas," said Michigan coach Rich Rodriguez, AFCA Board member. More secrecy, less hidden agendas, Rich?? Lewis Carroll would have loved that one.
Coaches also had a clear tendency to boost teams in their conference both for the Poll’s Top Spot and for the Top 25 rankings.
Conference Boost – Top Spot Voting
Six of the seven SEC coaches with a vote voted Florida to the top spot. Steve Spurrier had the Gators at number two to Big 12 Oklahoma. Urban Meyer had Florida and Alabama 1-2 after the SEC Championship.
Four of seven Big 12 coaches (Chizek, Hawlins, Pellini, Pinkel) had Oklahoma and Texas 1-2. Mike Leach had Oklahoma and his Texas Tech 1-2 with Florida No. 3. Art Briles (Baylor) had Oklahoma and Florida 1-2. Mack Brown was the Spurrier of the Big 12, picking out of conference SEC Florida No. 1 and his Texas Longhorns No. 2.
Of the seven Big Ten coaches voting, only one (Lynch) had Florida No. 1, four (Bielema, Dantonio, Rodriguez and Zook) had the Gators No. 2, and the other two (Tiller and Tressel) had Florida No. 3.
With 26 of the 61 first place votes for the Gators, an average percentage for a block of seven coaches would have netted Florida two or three top spot votes.
Vanderbilt coach Bobby Johnson said of the change: "My personal feeling is that a person is more free to vote his conscience instead of worrying about what the public is going to think about his vote."
Bobby, you mean in secret balloting more SEC coaches would vote for a Big 12 champion instead of the SEC champ? How will we know?
Ranking Boost by Conference – Top 25 Teams
Urban Meyer: “I think at some point you have to trust the people who vote.” Urban wants fans to take a “leap of faith” after observing four years of clear bias in voting for conference teams.
For 2008, conference coaches tended to vote for each other’s teams. Pac-10 – 2.3 spots, WAC – 2.0 spots, Big 10 – 1.6 Big East -1.6, Big 12 – 1.6, ACC – 1.5, Conference USA – 1.5, SEC – 1.3, MWC – 0.1, MAC – -0.6.
In other words, if you have a team competing for the top 25 in any conference except the MWC and MAC last year, you could expect an average boost by conference coaches of 1.7 spots!!
Six of the seven Big 10 coaches included Iowa in their top 25. Boston College was named on four of the six ACC coaches ballots. SEC coaches were not as uniform with regards to Mississippi. Five of the seven coaches boosted the Rebels higher on their ballots. Mississippi was bumped up 15 spots by Les Miles, 10 by Urban Meyer, 5 by Phil Fulmer, and 4 by Mark Richt.
Tommy Tuberville listed them one spot higher, Sylvester Croom one spot lower, and Steve Spurrier three spots lower. Miles and Meyer lost to Mississippi, while Spurrier beat them. Does that mean these coaches are more impacted by head-to-head conference matches and less able to evaluate other teams nationally over the course of a season?
Tommy Bowden was king of Conference Boosting. Though he lost his job at Clemson in mid-October, he kept his vote and voted five ACC teams in the top 25—Georgia Tech 10th (7 spot boost), Virginia Tech 13th (+6), Boston College 16th (+10), Florida State 21st (+7), North Carolina 25th (+6).
His ACC advocacy resulted in his voting five conference teams 34 total spots higher than their results, an average of 6.8!!
Regional Voting – Florida vs. Texas
In addition to the six SEC coaches voting Florida No. 1, the five other coaches in Florida ranked the Gators either No. 1 or No. 2. Mario Cristobel, Florida International, and Howard Schellenberg, Florida Atlantic, ranked the Gators as No. 1 - as did Tommy Bowden.
FSU’s Bobby Bowden, USF’s Jim Leavitt and UCF’s George O’Leary voted Florida No. 2, and former Florida coaches Steve Spurrier and Ron Zook.
Of the nine votes for SEC Champion and eventual National Champion Florida as No. 3, five came from Big 12 coaches, two from non-Big 12 Texas coaches and two from Big 10 coaches.
Mack Brown did not get the kind of solid support in Texas that Meyer got in Florida. His Longhorns were not conference champions, though they had only one loss. Three coaches from the state of Texas voted UT No. 1 or 2 and the other three voted them No. 3 or 5. In all, four coaches—Mike Price of UTEP, Todd Dodge of North Texas, Gene Chizek of Iowa State (former Defensive Coordinator at Texas), as well as Rick Neuheisel of UCLA (whose assistant coach at Washington was Texas’s Bobby Kennedy) voted the Longhorns No. 1.
Brown voted his Longhorns No. 2. Mike Patterson of TCU chose them No. 3. Art Briles of Baylor and Mike Leach of Texas Tech voted Texas No. 5.
Also, no significant regional boosts came from the six MAC coaches to the Big 10 teams (+0.6 spots), nor from the six Conference USA coaches to the SEC teams (+0.6), nor from the four Mountain West coaches to the Pac-10 teams (+0.15).
I picked these conferences since the MAC had 13 games against the Big Ten (4-9), Conference USA had 9 games against the SEC (0-9), and the Mountain West had eight games (6-2) against the Pac-10.
Kentucky coach Rich Brooks feels keeping the Coaches Poll’s final voting public "maybe has a little more validity if it's not protected, if it's open."
Steve Spurrier feels the same: "I thought that we would stay public on that last vote. I sort of think we ought to stay public, you know. It keeps everybody pretty honest."
Coaches Poll 2010
Mack Brown remarked, after looking at the AFCA decision: “I appreciate Grant Teaff and the AFCA taking such a serious and thoughtful look at improving the coaches poll.The Gallup Poll provided valuable insight and the AFCA put a great deal of thought into it. Since it is such a key factor in our national championship game, the goal is to continue to make the poll as fair and accurate as possible. I like the changes that are taking place this year and am excited about the recommendations for the future."
For the Coaches Poll results for 2005, 2006, 2007, The Blue-Gray Sky, a Notre Dame football blog, made a thorough analysis. Their blogger, Jeff, concluded in his 2005 analysis “it’s still somewhat shocking to see such blatant gamesmanship laid bare”.
Each year he showed worsening patterns of Self-Promotion and Conference Boosting among the coaches. This year Jeff took the four years of Coaches polls rankings,and crunched the numbers to see how accurate coaches ranking of teams were, assuming the higher ranked teams were considered better by the coaches. He compared those rankings to bowl outcomes. Coaches were less than 52% accurate in predicting bowl winners over those four years and, in BCS games, were worse - 49.3% correct. He concluded "You would be better off flipping a coin to figure out who was going to win a BCS game than looking at how the coaches ranked the teams." (May 14th article)
Whether the Coaches Poll remains part of the BCS formula is not up to them and their participation may be in doubt.
After declaring their final vote would no longer be publicized, BCS Coordinator, John Swofford: "In the past, the commissioners have favored transparency in voting by the people who participate in the two polls that are used to compile the BCS standings. The commissioners review all aspects of the BCS arrangement—including the BCS standings—at the conclusion of each season, and I know the AFCA's decision will be on the agenda for that review after the January 2010 games."
After viewing such blatant partisanship and obvious bias, the relic of the Coaches Poll should be allowed to slip quietly beneath the surface of college football.