Quote:
Originally Posted by antneye
I got a nice buzz last night....took some shots at gobbomom (can't ever resist that) and awoken feeling quite calm and clear.
I don't know that I can offer much over the next few weeks. I expect Romney to win. I see Obama's bounce from the 3rd debate fading and I see this settling in at R 51 / O 47 nationally as Gallup and Ras seem to be converging at.
The swing states are the key though, and i think that Romney will carry them since the key ones are well within the margin of error. I suspect after the left has it's meltdown we will all engage in a nice post-mortem to try to analyze the various pollsters and who got it right/wrong and why.
Nate will be just fine because, hey, 30% shots hit and all that, but it will be interesting to see whether he tweaks his model and learns something from the process or just chalks it up to being on the wrong side of variance. (fwiw I see him as someone who is trying to get the model right and not as a partisan trying to help Obama). I wonder if there would be some way for him to have Romney win and actually explain that the model was right because x, y, and z happening which lead him to win were all part of the 30% likelyhood...I'm not sure if that makes any sense. I guess I am wondering how effective of a post-mortem one can do on Nates model.
antneye:
After reading your "interesting" logic and analysis of what you see as the probable election outcome, (along with your interesting interpretation of the flaws in Nate Silver's model), I've had a sudden epiphany: Regardless of how this election turns out, everybody is going to claim they were right - even if their side loses! (I was listening to Jason Lewis last night on the radio while I was driving around. He's already charging voter fraud claiming that a computer somewhere [Ohio?] is magically changing early votes cast for Romney into votes cast for Obama. Of course, he offerred absolutely no proof or a source to back up this allegation - he just threw it out there.)
There is no objective analysis or clear cut "logic" in this thread. Everybody is guessing - arguing their feelings and emotions. We're all (including me) cherry picking "facts" to support how we wish the election will turn out. This is not the sort of thing where you enter data into a computer, press some keys, execute a query, and out jumps the answer. You can't precisely "model" raw human emotion - despite the efforts of smart people like Nate Silver to do so. This is what makes the interaction between "numbers people" - who try to totally eliminate human emotion from decision making - and people who rely on "intuition" and "feeling" - so fascinating. This is also why this election is so captivating. People are really polarized - split right down the middle. If elections could be accurately predicted with models, there wouldn't be a need for elections - just press a button and see what the computer says. (If we did away with elections, then I suppose the battle would be over who programs the computer ...)
This squabble over Obama versus Romney - and who will be "butt hurt" on election night - is reminiscent of poker players arguing over the best way to play a hand. But maybe that's the whole point of politics - it's just another arena where people struggle to get the upper hand in a game that never ends.