|
I think we've all seen versions of this, either "as is" or broken down into paper-trail and no-paper-trail states, over the past few days. Most of the time, you seem to have been the one that posted it.
Look, no offense, but I'm a little suspicious of your statistical analysis. You seem to have had Kerry as a 90%+ lock to win the election throughout the campaign, even in early autumn when Bush had a big lead. At the time, I couldn't see any reason why such optimism was justified, no matter how many "Monte Carlo Simulations" you might run.
As the race went on, even though the polls showed a tight race with Bush slightly ahead more likely than not, you kept to your 90%+ probabilty. As I looked at your figures, I noticed that you had a built-in assumption that 60% of the undecideds would vote for Kerry. As the election wrapped up, still a "dead heat" or "too close to call" in most opinions, your Kerry victory probability kept going up -- and it took reading your site to realize that you were getting that figure only by raising your percentage of the undecideds that would break for Kerry from 60% to 75% (!).
Look, we all wanted Bush out -- and I think, in retrospect, that this is why we looked at the polls and only saw what we wanted to see. Even though the race was mostly dead-even within the MOE, and could have gone to either candidate by a few percent, we used various mental dodges (the "under 50% rule," the "approval/disapproval" comparison, the "Kerry better in the battleground states" reasoning) to convince us that what in fact was a virtually-tied race -- in which there was every bit as much chance that Bush would win as Kerry -- was really something where the odds were overwhelmingly in our favor.
And I'm not saying the election might not have been stolen. As a matter of fact, I would tend to think it was. But the fact remains that it had to be close enough in the first place to be credibly stolen, and the notion that Kerry was 90%, or 95%, or 99% sure of victory was more wishful thinking than anything "reality-based." Not to deny that some electoral chicanery likely took place, and that it may even have provided the decisive margin, but isn't it possible that your statistical analysis might simply have been dead wrong?
|