World & Nation
AP The Wire
Comics & Games
Home & Garden
Advertise with the Times
Polls a crapshoot; truth lies in playoffs
By GREG AUMAN
© St. Petersburg Times, published May 7, 2000
Every week for the past two months a new state softball poll has been released, and every week for the past two months I've wondered what the margin of error on something like that must be.
This week there is no new softball poll, but at the same time, for the first time all season, you'll have a good idea of just how accurate the state rankings are. The results so far confirm what I've feared all season: These state polls, while helpful and well-intended, are ultimately a crapshoot.
Here's the problem: With hundreds of high schools across the state, even the most intrepid sports reporter will only see a fraction of the teams he is asked to vote for. Ballots are handed in each week by a dozen or so writers, and each week they'll rank 10 schools in six different classes. At best, they have seen 15 of those 60 play, and the actual number is probably less than 10.
This isn't the voters' fault. They aren't paid anything extra for the time it takes them to vote, and before that, to collect results for their local teams, which are sent out to the rest of the voters to better inform them so their selections can be accurate as possible.
Unfortunately, it's hard enough to keep up with teams in your own readership let alone a school 400 miles away, based on one line of information. Here's a typical debriefing:
No. 6. Kings Academy (West Palm Beach) 16-7: Beat Royal Palm Beach 4-1, lost to American Heritage 2-0. Pretty sure the best hitter was sick or something.
Even when fully informed on a team, all a voter knows is its record, a few scores, and very little context. Voters are asked to compare two teams that have never played, probably won't play and likely won't have so much as a common opponent.
What does a Miami writer know about the Panhandle's perennial district champions? Does anyone in Tallahassee know anything about the dozen Miami schools in the poll except their names? It's like standing in the frozen foods section, squinting over into produce and announcing with authority which cantaloupes are the ripest.
What's more, each voter is torn between two equally compelling insecurities: the fear that he has grossly overestimated the local talent, and a growing paranoia that all season long, he had a state champ across town and was putting it at No. 10.
There's a natural tendency to vote schools from big cities higher, simply based on the fact you do not nor will you ever know where Wewahitchka is, let alone how deep its bullpen is.
And if sheer ignorance isn't powerful enough to scuttle a poll's dreams of accuracy, a little sabotage can do the trick too. The final softball poll had nine voters -- each ballot gives 10 points to its No. 1 team, nine for No. 2 and so on. You need look no further than the first two teams on the poll to see how one voter can mess things up.
In the Class 6A poll, Orange Park received eight of nine first-place votes but finished second behind Tallahassee Leon because that ninth ballot had Orange Park ranked 10th. Leon gets 82 points, Orange Park gets 81. One vote can outweigh eight others in otherwise unanimous agreement.
This doesn't happen often, but it's another good reason why these championships are decided on the field and not on somebody's calculator. The 6A discrepancy hopefully will be cleared up by the two teams this week, as both schools were still alive entering Saturday's regional finals.
Likewise, the Class 3A poll -- which has Hernando ranked third and has Central just missing the top 10 by a single point -- seems remarkably prescient right now. Of the eight teams that advanced to Saturday's regional playoffs, six were ranked in the top nine, with No. 11 Central also in the mix. Of the top seven ranked teams, the only two that were knocked out lost to other ranked teams.
"Pure genius!" you think, marveling at the way the poll has canvassed the state and gleaned from the masses only the very strongest of teams.
But not all the softball polls are quite that accurate. Of the other five polls, none has more than four ranked teams among the eight still alive. The Class 4A bracket, for instance, has just one of the top eight teams in the final poll still alive in the round of eight -- and that one team is defending state champion Palm Harbor, a consensus No. 1 pick all year.
But what could remedy this? Perhaps the Dish Network will start a full line of live softball telecasts. With dozens of ranked teams playing on the same nights, better get that VCR ready too.
A lineup of a dozen teenagers is going to surprise its own coach a couple times a year, good and bad, so how can you expect a reporter six hours away to have their future pinpointed?
The only thing more preposterous than attempting a state poll is to take the collective guesses of all 50 states and assemble a national top 25. Anytime a team's motivation to win is to leapfrog a team from Oregon and banish another from Arizona into "also receiving votes" land, there are bigger issues to be dealing with.
© St. Petersburg Times. All rights reserved.