Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

longship

(40,416 posts)
8. I do not like the use of the word calculus here.
Wed Sep 5, 2012, 01:56 AM
Sep 2012

But then again, I am a mathematic pedant.

I appreciate people's respect for Nate, simply from the number of citations here on DU.

However, as a 64+ year old pol, and a mathematician, I am skeptical of his methodology. I cannot help but to see a bit of Confirmation Bias in the methodology. One wants to see a specific outcome so one looks for that outcome and rejects all others. One that is pissing me off is the Electoral Map that rejects Ras polling because it is biased. Bull shit! All polling is biased. One does not throw out a poll because of a perception that it is biased. It may be an outlier, but it will be balanced by an outlier biased the other way. No respectable scientist would throw out such data. They would just assign an appropriate margin of error. You shalt not arbitrarily throw out data just because you do not agree with it. Rasmussen is data. It stays. It may be outside a standard deviation, but that's part of the dirtiness of the data. Too bad.

But, don't get me wrong. I am not saying that Nate's techniques are wrong. However, this may also be called the Lottery Fallacy.

Of all the electoral map predictions, somebody's gotta get it closer than the others. In the lottery, somebody's going to eventually win. I am sure that a gazillion dollar lottery winner could make multiple gazillion dollars by claiming that they had the secret to success. But the Lottery Fallacy says that somebody has to win. It is a statistical necessity. So the winner of the the Lottery Fallacy really had no more power than any other player in the game, no matter how much effort they put into it.

These are the statistical and mathematical realities.

When somebody claims historic polling accuracy, or ability to predict stock prices, or ability to predict Three Card Monty, it is all the same thing. The Lottery Fallacy.

I look at these electoral college maps, often every day. But I always see them as confirmation bias. Obama's ahead this week again. That spurns me on. If Obama's behind? That spurns me on, too.

That's how we should all see these things.

Latest Discussions»Retired Forums»2016 Postmortem»A little electoral colleg...»Reply #8