Polls: Who got it right?

[GiancarloC]

http://fivethirtyeight.blogs.nytimes...idential-race/

Our good old friend Nate Silver has yet another article lambasting and criticizing various pollsters... and describing which ones were the closest to reality.



Nationally at least... Quinnpac, SurveyUSA, IBD/Tipp and We Ask America were all extremely close. The biggest offenders of the reality? Gallup (horrendous) with a 7.2% republican bias, American Research Group (4.5% bias in favor of GOP) and the much touted by the political right wing Rasmussen (3.7% bias in favor of GOP).

And quoting from the article:

Rasmussen Reports uses an online panel along with the automated calls that it places. The firm’s poor results this year suggest that the technique will need to be refined. At least they have some game plan to deal with the new realities of polling. In contrast, polls that place random calls to landlines only, or that rely upon likely voter models that were developed decades ago, may be behind the times.

Source: http://fivethirtyeight.blogs.nytimes...idential-race/

Rasmussen continues to get it wrong and got it wrong because of the way they conduct polling. They randomly make calls on landlines... a lot of younger voters don't have landlines with many only having cell phones. It's an outmoded way of polling and constantly had a republican bias. I believe pollsters like Rasmussen and ARG gave Romney a false sense of security and victory... along with his own campaign's internal polling. Republicans have often shot accusations that polls don't sample enough republicans... but the reality is there may not be justification to include republicans as they are a shrinking portion of the electorate.

It would be interesting if Nate breaks it down regarding certain battleground states and who got it right... but for now it's interesting to see how polling companies did.

[GiancarloC]

Correction: I mentioned "We Ask America" and "Survey USA" as one with less of an error, but Google and CNN would be better examples. Bias is one thing... the average error ratio is more important.

The real winner of the whole thing? IBD/Tipp. Notice... IBD/Tipp calls cell phones too! Rasmussen does not. And while Gallup and ARG did call cell phones, their sampling was total crap.

[andysayshi]

Fascinating info, thanks for posting it. Amazing that Gallup are the worst of the worst. It just goes to show how little can really be predicted with the very small polls used in modern politics. Nate Silver has proven that it takes more hard work than a few hundred phone calls to really gauge the electorate.

[rareboy]

Some of the pollsters really do need to take another look at their methodology.

[GiancarloC]

Yea, Rasmussen needs a serious update and it's my belief daily tracking polls need to be eliminated. They are subject to wild swings. But yeah, every time I heard about Rasmussen being the most accurate it was frustrating. Rasmussen has been wrong in 2008, 2010 and 2012. One would think they would get the message by now. And Gallup is just inexcusably bad.

[loki81]

I thought this was a good read... http://www.politico.com/news/stories/1112/83672.html

"On the Republican side, this was the worst cycle ever for polling and there’s nothing that even comes close to it... It was a colossal disaster and it wasn’t confined to the presidential campaign."

[JayHawk]

This is a connected article about why Romney chose to ignore other forms of reality and why they may be continuing to 'pretend' that they had no idea.

The article is called "Rethinking Shellshocked Romney"

It lays out a very convincing argument that Romney knew and he kept it hidden because Republicans risk destroying their money machine.

The two reason the Romney campaign is sticking with the shellshocked story are:

Option A: admit that you were operating in a bubble, that your pollsters were making faulty assumptions, and that despite the fact that your pollsters were coming up with numbers that didn't look like anyone else's, you were so reliant on gut feelings about voter enthusiasm that you didn't bother to seek a second opinion.

Or Option B: admit that your data looked much like everyone else's and that you're smart enough to know that all along that you were losing, but that the rules of the game prevented you from publicly admitting that. That's partially because, via the 'bandwagon effect,' it might depress turnout, but mostly because it would depress contributions from big money donors who don't want to waste their money -- thus becoming a self-fulfilling prophecy because you then wouldn't have the money you'd need to even have a shot at winning.

I am cynical when it comes to politicians so I would imagine they were well aware they were losing but didn't want the spigot of cash flow to stop... remember before the first debate many wealthy donors were reluctant to part ways with cash on a losing proposition.

[GiancarloC]

JH, I think Option A would be the honest one. I really think they believed in what their pollsters were saying, but also realized it wasn't telling the full story. One quote I brought up recently is Romney ran a copy of Ronald Reagan's 1980 campaign. He didn't realize one thing... this country is far different demographically than it was in 1980. Romney picked up the same majority of white male voters that Reagan did. Reagan won in a landslide... Romney lost by 2.5% in the popular vote and arguably, lost in a landslide in the electoral college.

I think there may be a third option. Republicans were trying desperately to bring back the days of Ronald Reagan when some had all of the cake and others were left with the crumbs.

On a different note... the shellshocked story could be true... looking at Romney's face when he gave his concession speech. Body language says a lot about a person... and even Mitt Romney couldn't hide it. I really think he thought he was going to win. Those around him gave him a false sense of victory.

[JayHawk]

The odd thing is the article gives a couple different examples of the campaign giving away the fact that they knew they were going to lose.

For instance in the weekend leading up to Tuesday on many shows the same pundits from each campaign were ask chances. Obama's team was thoroughly positive on a win. Romney's camp would say thing like "it is close" and "we are praying" which are things the loser says not the confident winner who is branching out into other states.

No I think it far more likely that the Romney Campaign and Karl Rove saw the writing on the wall but didn't want to admit that they were wrong so they chose to blame their internals.

[GiancarloC]

You are perhaps correct on that point. But I don't know... looking at Romney's face... I think at least he thought he was going to win. The guy was pretty shocked when he was giving his concession speech, which he didn't even have written at the time. Jon Stewart had a segment on Fox News where several analysts said "Romney is going to win in a landslide"... the one with Karl Rove's meltdown on TV.

Perhaps you're right though... they didn't appear too confident to begin with.

[JayHawk]

Yeah there is other opposite evidence such as them ordering a huge fireworks display for the win.

Perhaps Romney was that easily deceived by those around him... :O

[GiancarloC]

What I ultimately think happened is his advisers didn't tell him how seriously screwed they were... they gave him a false picture, but while perhaps knowing they were in trouble. It's funny how many of his staff had their campaign provided credit cards cut on the night of the election lol. Perhaps that was punishment.

[MystikWizard]

I really have to question reports of how much Romney was really "surprised" by the results. Towards the end of the campaign, I really didn't get the impression he really wanted to be President. I don't know if these are just stories or what ... but I honestly don't get the feeling Romney was truly surprised by the results.

[Orlandude]

The problem with polls is that don't only sample likely voters or whomever, they shape public opinion.

[GiancarloC]

The problem with polls is that don't only sample likely voters or whomever, they shape public opinion.

Some do better than others shaping public opinion... obviously Gallup and Rasmussen failed miserably.

[CoolBlue71]

Rasmussen Reports is a Republican pollster. No question. It's sad, though, what has become of Gallup, which I sensed for some time had become really fucked up. (New owner?)

[LuvFindsAndyHardy]

It's sad, though, what has become of Gallup

They seem to subscribe to the "if it ain't broke..." methodology but they need to update and re-think the way they conduct polling. The problem is that they haven't changed with the times and reliable information can no longer be gathered in the same old ways.