Nate Silver's Final Prediction: Obama 332; Romney 206

http://fivethirtyeight.blogs.nytime...ama-leave-romney-with-longer-odds/#more-37295

At the very last second Obama has been predicted to win Florida by Nate's model. Even I'm a little skeptical about that final prediction. But I think everything else follows the polling pretty well. If Obama can take Florida is going to be a HUGE upset. An even bigger upset would be if Obama won North Carolina. It's the only battleground state left that Romney still has a hold of. But a 25% chance of victory in North Carolina is far from impossible!

However I think the odds are more likely to shift slightly towards Romney on stealing a few states out of this prediction. We shall see.

But Romney MUST WIN Florida, Ohio, Colorado and Virginia. If he loses any of those states it will be pretty tough for him to catch up in the math. Because no...he's not going to win PA, MI, WI or NH. Too many republicans on this forum have their heads in dream land.

 

Yeah, just like in 2010 huh?

....and then The Largest Political Ass Whooping in History took place.

332 for Obama.....Dreamland indeed....
.
.

 

You seem to forget that most people were predicting a republican victory in 2010. Including Nate Silver who was EXTREMELY accurate in predicting the outcome of the house and senate elections.

 

and Nate Silver predicted that too....

 

I trust Nate more than anyone on these election predictions. He's good. he's real good. It surprises me to see him predict this easy a victory for Obama as he is. Interesting.

 

My prediction is that Romney will lead until around 9 AM when Romney's supporters are all at work.
Then, as Obama supporters begin to get out of bed around 10AM Obama will catch up with Romney.
Finally Romney will defeat Obama as his supporters vote on their way home from work.

 

Yeah me too. It's Florida that I think is an extremely bold call on his part. I don't think it will happen. But if it does there simply is no way for Romney to come back from that. He could literally win every other swing state and still lose if Florida turns his way. But I don't think there is much of a scenario where Florida goes blue without enough of the rest of the swing states to go blue for an Obama victory. Hence why Nate Silver weighs Florida so little on the tipping scale. Even though Florida has a lot of votes there just isn't a way for Obama to win it without already having more than enough votes.

 

Plenty of Obama supporters have jobs. In fact most people who work high skill high pay jobs are Obama supporters.

 

This will definitely be a thread to bookmark because when this guy is wrong it'll be nice to remind people of it in future elections.

Obama may win but not by that much.

 

Yeah, hate to tell that jackass, but I got up at 5:45, woke the kids up at 6:00, got ready for work, dropped the kids off at school at 6:45, then my wife came home from working a 12 hour shift. We both went to the polls at 7am, cast our OBAMA votes, then I dropped her back off at the house and am presently at work. My plan after work is to help some elderly people I know who can't get around by driving them to the polls so they can also cast some more OBAMA votes.

Oh, and my oldest son got up at 7 this morning to vote, before he goes in to work this afternoon.

I really hate how cynical and one dimensional some people are getting. Like only Republicans work... what hogwash.

 

And what if, (though I do have a hard time believing some of the Florida stuff) he ends up calling it right? What if Obama not only wins the Electoral Vote, but also gets a clear majority vote as well? Even if it is only 1%, that would still be better than Bush's win that had him loose the popular vote but win through the electoral college.

What then, what does that say about the so called conservative mandate?

I'd personally take it to mean that the majority of people in this nation just want something that works, regardless of political affiliation. That the constant gridlock in Washington needs to be solved.

 

It would mean that the democratic praise of Nate Silver is unfounded.

People on the left are making him into some kind of political Nostradamus which he isn't.

If he does call it right then he is certainly the premier pollster in the world and will have to be given credit, but its not going to happen. If there were a fail safe way to predict elections we would have found it by now. Its like some guy going to Vegas and thinking he has a way to beat the odds of the house, not possible without cheating.

That was the point of my post. I really don't care if Obama or Romney wins as long as the republicans keep the House, which they will.

 

It's statistics. Not a psychic prediction. The only way he's wrong is if our polling is wrong. He's not saying anything that's far outlandish.

 

He isn't but you guys on the left are attributing some sort of divine code to his predictions as evidenced by you creating a thread about it.

He did some good calls in the past but to say he isn't going to miss some is irresponsible. The fruitcakes on the right who are predicting a Romney win are just as bad. Do you realize that if there was a sure method that everyone would be using it?

Nate Silver is no better than anyone else as we will see at the end of the day today.

 

LOL. Sure they do. Those who take tend to be at the high and the low ends. Don't they?

I took the day off to enjoy this. Locally record crowds are reported at all polling places. They are energized, they are enthusiastic and they are claiming they are voting for Romney. I am on my way to go vote. I wanted to wait for the "heading to work" crowd and for the school traffic to slacken.

 

Nate Silver is widely sourced because there are few people who have built a model and gone into as much detail explaining it as him. He's not only accurate but he provides excellent and detailed analysis. As someone who's into political science and political stats I find his articles fascinating. It's not really about whether he is right or wrong. And I don't trust him because he says Obama is going to win. I trust him because he backs his predictions up with solid polling evidence and facts about the demographics and various news events. It's a really fascinating read.

 

So you're criticizing Obama voters for not working but you took the entire day off to vote?

 

Well since you are into political science have you actually looked at his model? He not only averages polls but takes things such as economics into the question. Using that much data, without a pre-defined quantitative analysis of effects, is bound to give you a wrong outcome eventually.

Nate Silver is basically staking his reputation on this election. Hes made bets with people like Morning Joe that he will call it right and basically if he does, his reputation will be cemented. If he doesn't his reputation goes by-by. He has advertised himself too much as being correct and left little room for recovery if he blows it.

 

I think it's unlikely that Romney will only win one of the 11 toss up states. Possible but unlikely.

 

Yeah I can agree with that. For me though I understand how stats work. I understand that a 8% chance is still a nearly 1 times out of 10 chance. Meaning everything could possibly be completely different from what Nate is predicting. But in order for that to happen then all the polls would have to be wrong too. In which case... garbage in garbage out. Not really Nate's fault. The people we can really discredit are the people who are completely ignoring polls and saying Romney's going to win states like PA and Michigan.

 

I think you are wrong, the dems pushed their faithful to vote early. You will see a larger turnout of reps today than dems. Just my opinion.

 

Don't feed the trolls.

 

Well that's good to hear. Since you understand more than most you know that trying to predict polling results is at best, an educated guess. Polling in political science is problematic at best due the methods they are forced to used. It is not based off of empirical evidence but human perception, which is often undefined and subject to random changes.

You also have to conduct polling from a non-interventionist angle, which means you cannot rely on direct contact, for fear of running into the Hawthorne effect. This means that polling by mail, internet, or phone calls is not subject to verification of who your actual participants are. This means there is no guarantee that your actually getting the correct data your trying to obtain.

 

Yes. The difference is that I have a job.
I am one of those wealth makers who voted for Romney and for America.

I will also head out this morning to buy a large monitor to use with my work laptop and docking station since I tend to work from home on my "days off."

But do not be concerned. I will still find a way to listen to Rush to get my marching orders...

 

Nate has a probabilistic model. He's not necessarily right if Obama wins; he's not necessarily wrong if Romney wins. The EC vote total that his model shows is an *average*: if you look at the distribution, you'll see that the single-most likely outcome is 330 for Obama, but there are several other high-probability outcomes too.

The way we'll have to judge Nate's model is to see how close he got in each state and in the Senate races. That's a large enough set of races that his probabilities can be tested.