If I recall correctly, it went *great *for the incumbent, not so great for the bland, uninspiring blueblood from Massachusetts.
I imagine you are.
What data?
(A rhetorical question, of course, unless you want to go ahead and answer it.)
Oh, you’re so smart.
I have never seen Nate be anything less than transparent in his methods and being interested in being right rather than by being partisan. Look at his 2010 record. He overestimated the Republican senate pick-up by one (getting 34 of 36 elections right in the process) and he also missed one of 37 in the state governor races, missing only the one in Illinois, in which his model predicted the Republican, and the Democrat won. He, like all of us, may have his political biases, but when it comes to his 538 model, it seems to me he’s mostly interested in being right, not having his “team” win. He slightly overestimated the Republican victory in the Senate and state governorships. As far as the house goes, his model showed +55 median gain for the Republicans. The actual result was +63. He also warned that the 95% confidence interval was wide: anywhere from +23 to +81, and to keep that in mind with the house races.
It’s possible that Nate’s biases are involved, but it’s also possible that his model predicts the undecideds to break out in a way that results in ~50% for Obama- based on historical trends, or economic data, or “leaners”, etc. Alternately, the model might disregard the undecideds in a poll, such that a 47-47 poll is modeled in as a 50-50 poll (for example). Or some combination.
Yes, I know Nate’s not a pollster- I never said he was. But there’s tons of polls, and they show a very wide variation. Nate has a model (not the only one- I look at others too, like the Princeton Election Commission) that uses all the polls (and other data like market and economic trends) to make a prediction. His model has a good track record, so I trust that it’s close to the “real story”.
You don’t have to trust it, but you’ve presented no good reasons why I shouldn’t.
Also, you’re focused on the national prediction for some reason- the state polls show much more consistent, steady, and solid leads for Obama in the electoral college. I’m much more interested and focused on the electoral vote count prediction.
You could actually spend six and a half minutes reading something about Nate’s model. How it factors in economic figures, etc. etc. Or not, your choice.
It’s not like the old days, when you had to trudge over to your local library for facts like this.
Does not the comparison of approval ratings of Obama and Congress, currently at 48.7 and 15.4 respectively at RCP, suggest that people blame Congress to a larger degree than the president? I understand that since these are general levels of (dis)satisfaction, you can’t determine how people arrived at their approval level and whom they hold responsible for what, but surely the results are telling to a pretty important degree, no?
Just curious. Is there a good reason you feel this way, or are you just hoping it comes true? 
I suspect you may be right, though.
Facts have a well-known liberal bias. Omg will therefore eschew them.
Here is a good article about Nate Silver from The Telegraph (UK).
Also interesting is this bit, which seems relevant to what we’re hearing lately from our conservative Dopers:
I don’t want cause you any more despair, but I think it is quite likely that if Romney loses, the right wing will conclude it was because he was not conservative enough, and move even further to the right.
If Silver’s prediction once again turns out to be very close to the mark, I predict that there will be whining from the other side that it’s a self-fulfilling prophecy, because the model’s prediction depresses the losing side and reduces their turnout.
Or something.
I think we can probably say that Romney has opened up a narrow lead in the popular vote, but that Obama maintains a stubborn lead in enough states that matter and that would push him over 270.
The issue, then, is whether Romney can erase that electoral college lead in the span of a week. Can he do it?
Hell, anything is possible, but I just struggle to see how on earth Romney could pull something together like that in such a short amount of time. I mean, if he can’t take away Obama’s lead in Ohio after having gone through what was unarguably his best month of this campaign, I just don’t know what kind of magic trick he could pull out of his ass that would make that state more competitive for him.
By this time next week, we’ll have a pretty good idea of who is going to win. As it stands today, Obama would win narrowly, but he’ll still probably win narrowly on election day.
What I’m really curious to learn, though, is whether the Romney lead in the PV is driven by mass consolidation of support in reliably red states as opposed to independents breaking disproportionately his way.
Romney is going to win the red states in a big way. Really big. Obama just can’t get as many votes in blue states to make up for that fact. Thankfully, for this Obama supporter anyway, Obama leads where it matters and that’s electorally.
2% is just an OK number. I doubt it will have much of an impact on 538 where the economic component is falling steadily day by day. And a single month’s jobs number is largely noise so we really don’t know what will happen there. Anyway it will only have a small impact on the election in all likelihood.
Nobody knows for sure, especially because during the debates, he showed a move toward the center, but his stump speeches have been tossing red meat to the ultra-conservative GOP base.
In other words, Malleable Mitt is in overtime mode.
A CNN poll has Obama +4 in Ohio. I think this could move 538 up by a point since it’s so sensitive to Ohio. While a healthy Ohio lead is already priced into the model any high-quality poll with +4 or above will move the needle at this point, I am guessing. The other polling hasn’t been terribly informative today but my rough sense is Obama has slightly had the better of it.
New numbers up for 538.
Nov 6 Forecast: 74.4% for Obama, 295.4 EVs
Nowcast: 78.6% / 296.8
He now has Ohio 80.3% for Obama.
but but but unskewedpolls is showing a landslide for Romney!