Poll position

Excerpted with permission of the publisher John Wiley & Sons, Inc., from Bowls, Polls, and Tattered Souls: Tackling the Chaos and Controversy That Reign over College Football, by Stewart Mandel. Copyright � 2007 by Stewart Mandel

Sept. 5, 2006, 2:37 a.m., Room 2620, Biscayne Bay Marriott: Before leaving for tonight's Florida-State Miami game, I ranked 23 of the 25 teams on my AP ballot, leaving space to move up the winner and move down the loser of Monday night's contest. However, after watching the teams' utterly inept offensive performance, I'm not sure either team should be ranked any higher than, say, Prairie View. Florida State, the winner, rushed for a total of 1 yard. Miami, the loser, rushed for 2. The game went right down to the last minute, yet I couldn't have been any further from the edge of my seat. I take a quick look at my preseason ballot, drop FSU two spots, from 12th to 14th, and Miami six spots, from 17th to 23rd, then send the thing in so I can try to get a few hours' sleep before heading to the airport. The next day, it will be pointed out to me that as a result of dropping the 'Noles two spots for beating a ranked team on the road, Iowa moved up two spots, from 13th to 11th, for beating I-AA Montana on its own field. Whoops.

We are a society of people obsessed with rankings, and personally, I blame Casey Kasem. So many generations of pimply-faced American teenagers grew up listening to the host of America's Top 40 tell us where our favorite Madonna or Bee Gees song was on the charts that week (in addition those painful long-distance dedications) that now we crave rankings for nearly every aspect of our lives. U.S. News & World Report's Top 25 law schools. Sports Illustrated's Top 20 Hottest Female Athletes. E!'s Top 10 Celebrity Break-Ups of All-Time. Of course, those rankings are mostly for amusement. The issue of ranking college football teams is a decidedly more serious matter -- at least to fans of those teams being ranked.

Here's how it usually works. On Saturday, they play the games. Late Saturday night or early Sunday morning, most likely while you're sleeping, a panel of sportswriters and coaches evaluate the results of those games and compile their rankings of the 25 best teams in the country for that given week. Then on Sunday afternoons, fans of those teams scurry to their computers to find out the latest results -- and to fire off a nasty e-mail to whoever might listen (in this case me) detailing the grave injustice that has occurred as a result of their 7-2 team coming in 16th while a 6-2 team that their team beat by a field goal six weeks earlier is four spots higher. The e-mail will inevitably include some offshoot of the following phrase: "What were the voters thinking?"

Since 2005, I have been one of those voters, submitting a Top 25 ballot to the Associated Press at the completion of each Saturday's games. (This ballot also constitutes my weekly Power Rankings for SI.com, which I've been publishing since the 2002 season). Utilizing first-hand knowledge, I can assure you that the process of comparing two or more teams with identical records, that don't necessarily play the same opponents and whose most recent performance may or may not have been affected by injuries, venue, weather, kickoff time, a bad call by an official or all of the above is only slightly less subjective than ranking those Celebrity Break-Ups. While every voter I know in the media takes his responsibility very seriously, the reality is none of them, nor I, have any true way of knowing the exact order of the best teams in the country unless they were to all somehow play each other. In fact, for all the sport's other technological advancements, the only thing that's changed about voter polls in 70 years is that today you can e-mail your ballot to the AP rather than call it in. And yet, these archaic, largely arbitrary rankings still manage to create an indisputable craze among anyone with any attachment to the sport.

So who votes in these polls anyway, you ask? And what exactly qualifies them to evaluate football teams? The first question I can answer. The second one is open to debate.

The AP poll consists of 65 voters, 61 of which are "districted" state-by-state much like Congress or the Electoral College. States with one to three Division I schools get one vote; states with four to six such schools get two, etc., etc. In 2006, Texas and California had the most voters with four apiece; Florida, North Carolina and Ohio had three; everyone else had two or one. There were also four "national" voters, of which your humble author was one. Fortunately, the national scope of my beat allows me to get to different parts of the country and see most of the top teams in person at some point.

Because there is no playoff in major college football, the voter polls have become the sport's ultimate arbiter. As a result, it's also the only known sport where the teams are essentially seeded before they ever play a game. Preseason polls, long a subject for lively debate and conversation, have become even more controversial with the advent of the BCS, seeing as they effectively give the highest-ranked teams a leg up on everyone else in the chase for the national title. It's hard, though not impossible, for a team that starts the year ranked 20th to finish No. 1, even if it goes undefeated. The No. 4 preseason team, however, can lose a game and know it still has plenty of time to climb back up. Starting No. 1 in the preaseason is like having pole position at the Daytona 500 or the inside post in the Kentucky Derby. Fourteen times since 1936, the team that started the season No. 1 in the polls also finished there. Lately, the pollsters have served as pretty good prognosticators, too: For five straight years, from 2002 to '06, the AP and coaches' preseason No. 1 reached the BCS championship game. And either or both the preseason No. 1 or 2 teams have reached the title game every year since the BCS's 1998 inception. Three of those teams ( Florida State in '98 and 2000 and Oklahoma in '03) did so despite suffering a loss.

Every year, coaches and fans alike gripe about the existence of preseason polls, which, as any voter will tell you, are an exercise in educated guesswork. "If you're going to have this system, then [polls] should start around the first of October," said Auburn coach Tommy Tuberville, whose undefeated 2004 team failed to reach the title game in large part because it started the season ranked 18th. "People will have a little bit of an idea on how [teams] are doing rather than guessing." He's probably right, but reality dictates that preseason polls aren't going anywhere. Why? Because there's an insatiable public demand for them. Even before the AP and coaches issue their official preseason edicts in August, one can walk into a bookstore as early as May and see racks and racks of preseason magazines (Athlon, Sporting News, Street & Smith) with their own Top 25 lists. And thanks to the Internet, the poll conversation now begins almost as soon as the last season ends. SI.com published my first top 25 for the 2007 season on Jan. 16, a full seven months before teams report for fall camp -- and not surprisingly, it got hit hard. The organizers of the new Harris Poll tried a noble experiment in delaying the release of their first poll until the last week of September. Of course, when the poll did finally come out, it looked almost exactly the same as that week's AP and coaches polls. Even if the AP and coaches did delay their polls (which will never happen), it's na�ve to think the voters would sit down with a blank piece of paper and list 25 teams from scratch. Inevitably, they would use an Internet poll or other poll unofficial ranking as their starting point. Let's face it -- most pollsters don't have the time to be original.

So as long as there are going to be preseason polls, one might as well gain an understanding of how they work. I've been studying and conducting these things myself for the better part of a decade and in doing so have noticed four common criteria most voters emphasize: