Polls don't matter (part 255)

I've written several essays on why I object to the media's overreliance on polls, particularly as a substitute for actual reporting. I don't think an article based on a poll should ever be on the front page, certainly not a poll concerning an election for office.

Using polls as news tells nothing about the candidates' positions on the issues. It tells nothing about their character. It tells nothing of their record. It merely illustrates the perceptions of a tiny handfuls of individuals. Would Sports Illustrated ever waste space with weekly articles on what percentage of baseball fans like the Yankees vs the Red Sox vs the Cubs?

Polls as a substitute for news is an attempt to make news rather than report it, something most in the news media swear up and down journalists never do.

The USA Today's front page top headline yesterday was Poll: Bush leads Kerry by 8 points, 52-44%. The Washington Post has Bush up by 3, TIME has Bush by 2. Who's right?

They all are. The reason these polls are often widely divergent is that they each sample different tiny handfuls of individuals.

The real question is why voters should pay attention to them in the first place. If you need a poll to decide who to vote for, should you really be voting at all? Should the commander-in-chief be chosen by voters who stick their finger in the wind? Sadly, this will probably be the case again this year.

But what makes the intense media focus on these national polls even more inane is their utter irrelevance.

If you accept that Bush is really ahead nationally by 2 or 8 points, it still doesn't matter in the least. Because of the absurd Electoral College, the real question is how his 52% [or whatever mythical figure you accept] is distributed.

The total number of AMERICANS who vote for a presidential candidate doesn't matter at all (and polling and voting aren't even the same thing, as many people forget). In the most extreme case, a candidate could theoretically receive 19 popular votes for president and be elected.

Even if people were asleep in high school social studies class, you'd think 2000 would've reminded the press and everyone else that the national popular vote doesn't matter.

But I guess you'd be wrong.