Skip to comments.My apologies to Nate Silver
Posted on 11/07/2012 9:08:42 AM PST by Raycpa
My apology to Nate Silver. I doubted your statistics. I assumed you were a smug know-it-all type numbers guy. You were right and you had a right to be smug.
I hear ya. ; )
I said all along that Nate Silver is the ONLY poll analyst worth listening to. Ignore EVERYONE else.
Ditto. I thought he had to be full of sh*t, but damn, his accuracy was uncanny. He was dead-on.
Instead, curse Barone.......
agreed.. he nailed it..
I am so surprised he was right. I expected more of our citizens.
If he knew obama was losing 10 million votes from what he had in 2008 why didn't he speak up?
RCP final poll average of each & every state also predicted each and every state correctly including Florida.
RCP ran a very simple algorithm - called average.
Nate Silver added a lot of weightings and secret sauce and what not.
Both of them ended up with the same result.
Wow, I’m surprised to see so many naive people on Free republic of all places. Nate Silver did not predict the outcome, he used his position to influence it. It came out to be about the same, because that’s what they’ve been working towards this whole time, by discouraging Republicans, harassing them at the polls and rigging voting machines.
In 2008...optimism among many (not me) reigned supreme. All we heard about was oversampling of dems...and anecdotal evidence about record turnout at their individual polling places. Republican turnout as described as mammoth.
Forward to 2012. SAME story...except this time I expected a landslide. All we heard about was "That poll is crap. D+8? YEAH RIGHT!!!" All we heard yesterday was story after story of record turnouts.
It all turned out to be crap. Any FReeper in ANY future elections who talks about a poll being garbage because of their D sample...I will simply remind them of yesterday.
RCP's average in Florida was Romney +1.5, two percentage points away from the final result.
Two surprises for me that threw my analysis off.
1. I never guess D turnout would be almost as high this time as last time. WRONG.
2. I assumed there would be movement towards the challenger as the race ended. WRONG.
I’m still not exactly sure how Silver knew both of those things, especially the second one, or how Axelrod and Plouffe were so confident the election was totally in the bag.
I give them my grudging respect though. Those guys nailed it. I hate them all, but they nailed it.
Shows that, if anything, polls are biased towards republicans. A GOP candidate needs to be up 2-3 points in order cover union and democarat turnout machine.
Let’s not go overboard. Likely he just gets Obama’s internal numbers handed to him.
You are right. But it’s still great. No fancy algorithm, no secret sauce (whatever that means). Nate got 1 state better than them. And even Nate gave Obama just 50.3% chance of winning Florida.
I actually doubt all the polls because I understand a little bit about sampling because I use it for auditing. I’m trained to be skeptical of my own results. I also have a great deal of respect for the polls done and am very hesitant to challenge a poll because the internals seem wrong.
My issues with Nate’s analysis was twofold. I thought he was cherry picking which polls to use and was adding bias instead of decreasing it. Second, his translation of his analysis into a certainty factor to the nearest 10th decimal inferred an accuracy level that is not sustainable from the data.
But he was right.
Let’s not go too far here. Silver may have his own model, own way of doing things, but when it comes down to it, all Silver did was create a weighted average of poll results. Silver was right, not because he’s a genius, but because the polls were right.
Exactly. Nate Silver (and even more, Sam Wong) are just the messengers, the polls they were aggregating wrote the message.
My big take away from the Silver/Wong results over time: the race remained pretty much in the same please since last summer, with most actual eventual voters having made up their minds well in advance of Nov. 6th. and most of the dips and rises for either candidate reflected differences in survey response rates due to partisan enthusiasm.
The outcome, in light of the polling, makes a *lot* more sense if that was the case.
I too said to listen to Silver and was attacked for it. Do NOT question big data and predictive analytics. It is very powerful.
“My big take away from the Silver/Wong results over time: the race remained pretty much in the same please since last summer, with most actual eventual voters having made up their minds well in advance of Nov. 6th. and most of the dips and rises for either candidate reflected differences in survey response rates due to partisan enthusiasm.”
That does make sense. To his credit, actually, I believe Silver made the same point a while ago. He commented that some of the rise in Romney’s polls after the first debate could have come from Romney voters who, because of Romney’s strong performance, were more willing to talk to pollsters.
I’m not sure that all of the poll fluctuations are attributable to differing response rates, but it is a plausible hypothesis, especially given our “endless campaign” political culture. After 2 years of nonstop campaigning, and 24-hour cable coverage of that campaigning, it’s certainly plausible that most people made up their minds months ago.
“All we heard about was “That poll is crap. D+8? YEAH RIGHT!!!” All we heard yesterday was story after story of record turnouts.”
The “polls are oversampling Dems” argument was really one of the more frustrating things to see repeated over and over, here and on every conservative site. It’s like no one understood that the vast majority of pollsters do not weight their polls to a certain D/R/I sample, and it would not make sense to do so (since party ID is flexible, often changes with one’s voting preference, and is one of the things pollsters are trying to measure - weighing to a certain party ID target makes only slightly more sense than weighting directly to a certain result).
The simplest explanation for why virtually every poll this season had a party ID spread of D+5 to D+10 is that the party ID spread of the electorate was between D+5 and D+10. Occam’s Razor.
Sadly, writing anything like this before yesterday would have been enough to get people zotted around here. Hiding from the facts does not make them disappear. Denying the facts does not help us win. That must be one of the primary lessons we take away from this loss.
Exactly. I remember some on here trying to explain that (that polls just report the weight...not determine it). But...people didn’t listen...and unfortunately I was one of them. I repeated that mantra to my friends...and now appear to be a big ‘ol idiot.
I do not agree with you. Most models dealing with individuals have a shelf life. Behaviors change. ALWAYS question predictive analytical models.
Nate was right but he could have been right for the wrong reasons. It could have been
Exactly. Silver is a total hack -- he made his mark in 2008 by using campaign polling numbers fed to him by Obama, and then creating some "model" mumbo-jumbo to make it look like and independent prediction. That got him on the NYT staff so he could then use his "model" to influence thought -- all public polls are propaganda, not a measure of reality.
He missed 2010 by a country mile. Getting it "right" in 2012 was more a case of creating the outcome than predicting it.
Perhaps, but it is reasonable to question whether the polling was correct and if there was some common flaw in the polls done by the various polling places.
Lets try a thought experiment. Lets assume I did a random sample of the US population and the results were that I received responses from 90% men and 10% women. Although, such a result is possible for a random sample I would need to question whether my method of selection was in fact random.
When the current crop of polls showed a stronger democrat response than what would be expected, it is perfectly normal to question whether the sampling methods were flawed. Unfortunately for us we do not have access to the actual methods these pollsters use in order to see if they are applying a different methodology. We also have instances in the past where these pollsters have erred.
One such reason that made sense to me that could have caused a common error for all polls was the greater emphasis on cell phone contacts. Being a newer method I could understand that wrong assumptions on who the population was could have an unknown effect on the sample.
Did Nate have a statistical method to determine if the sample internals were representative? Were the samples actually flawed like many thought and the polls were actually wrong but the result was affected by other factors?
Nate ended up being right.
Of course it's reasonable to question whether there was some sort of common flaw in the polls. The problem is that the typical reaciton around here (and on other conservative sites/talk radio/etc.) was not to question whether the polling was correct, but rather to assume that the polling could not possibly be correct, because people simply could not believe that the electorate would look anything like 2008. People took anecdotal evidence of enthusiasm for Romney (and against Obama)--lawn signs, crowd sizes, the sheer hatred of Obama that people see in conservative areas and on conservative news sources--and assumed that this observed enthusiasm could be extrapolated to the electorate as a whole, despite polling evidence to the contrary.
In a way, this election was sort of like a reverse-2004. Just like the Bush-haters on the left, some conservatives--here and elsewhere--have been so blinded by their own desire to get rid of Obama, that they simply could not accept that a (slight) majority of the electorate did not feel the same way.
Only means he was in on the fix. Rats KNEW they would get however many votes they needed from Detroit, Milwaukee, Cleveland, Philadelphia, etc. Until voter ID is the law of the land this will continue - rats will never lose another election that they care about winning.
Good luck with that. It happens every election here.
Top 5 poll lies told here on FR:
1. Polls are crap.
2. People lie to pollster.
3. Undecided breaks for the challenger.
4. Republican’s do worse on the weekends.
5. The polls are used to demoralize not report.
Absolutes are a problem both ways:
1. SOME Polls are crap.
2. SOME People lie to pollster.
3. MANY Undecided breaks for the challenger.
4. (No contest on this one) Republicans do worse on the weekends.
5. ALMOST ALL POLL SPONSER’S HAVE AN AGENDA AND SOME OF The polls are used to demoralize not report.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.