Nate Silver (Photo credit: jdlasica)
One of the most distinct features of the 2012 Presidential election was the polling wars that took place. The copious amounts of data seemed to tell two completely different stories. The first was that President Obama would handedly win the election by a significant amount while the other side showed former Governor Romney delivering a victory by the smallest of margins. The division on how to read the polls caused quite a stir among the Right and the Left. Well…in all reality neither sides forecast was quite that accurate with the exception of NY Time’s blogger, Nate Silver whose probability model accurately predicted the results in 50 of 50 states as well as the margin of victory.
What makes Silver so special is that he is what is known as a sabermetrician, not your standard political polling operative. According to Wikipedia, Sabermetrics is the technique of “specialized analysis of baseball through objective evidence, especially baseball statistics that measure in-game activity.” He has adopted these SABR techniques and uniquely fit them to prediction models which are far more accurate than those presently used by standard polling agencies. Below is an article from Foreign Policy which highlights the effects sabermetrics are going to have political polling and how Silver successfully pioneered this innovation in the world of politics. Very interesting read if you are curious as to what happened.
Before Billy Bean and the sabermetric revolution upended baseball and ushered in a new era of statistically driven baseball analysis, old-timers insisted that the young eggheads and their spreadsheets were no match for a time-worn scout. Experience, gut feeling, and a sense of the intangible qualities that make up a quality prospect — these were the things that the old guard argued could never be captured by an Excel spreadsheet, let alone a statistical model.
By and large, they were wrong, and Billy Bean’s scrappy Oakland Athletics squads showed that the so-called eggheads could see further into the future and with greater clarity than had previously been thought possible.The same statistical revolution that changed baseball has now entered American politics, and no one has been more successful in popularizing a statistical approach to political analysis than New York Times blogger Nate Silver, who of course cut his teeth as a young sabermetrician. And on Nov. 6, after having faced a torrent of criticism from old-school political pundits — Washington’s rough equivalent of statistically illiterate tobacco chewing baseball scouts — the results of the presidential election vindicated Silver’s approach, which correctly predicted the electoral outcome in all 50 states. [...]
If the GOP was smart they would approach Silver and offer him triple the salary that the Times is currently paying him and get him to head up their polling operation. It would be a first and great step to catching up the with the Democrats.
Related articles
- Nate Silver Went 100% on Election Day: Super Accurate Statistical Analysis or Luck? (dailyfinance.com)
- Moneyball, the 2012 Campaign, and the Limits of Political Sabermetrics (themoderatevoice.com)
- Nate Silver and the Nerd-Haters (theamericanconservative.com)
I followed the polls closely for at least a good 6 months prior to the election. I also followed Nate Silver. The simple truth was Romney was never in the lead if you followed the polls in the battleground states closely. Silver had Obama’s odds of winning at about 90% going into election day. But you didn’t need Nate Silver to see the election was going Obama’s way. And that’s what really made me shake my head at all the Republican bravado of “landslide.” It just seemed, well, driving into a wall with your eyes wide open. The Real Clear Politics averages alone, state-by-state, clearly showed an Obama victory going into election day. These polls were a sea of blue as you moved from battleground state to battleground state averages on election day. Yes, they were close (within the margin of error in some). But the chance of, say, 6 or 8 individual state polls being wrong, all showing Obama ahead, even if they were close, is pretty slim. I’m not a betting man as I told a friend, but I was betting Obama was going to win. The odds were clearly in Obama’s favor.
Jeff, a lot makes you shake your head about “Republicans.” It’s a wonder you haven’t developed vertigo! It doesn’t surprise me that you were put off by the bravado. It’s pretty safe to say from where you stand a Republican or Conservative is, by definition of being a Republican or Conservative, always wrong.
At any rate, the polls as you remember them isn’t altogether accurate. The RCP is an average of all the polls, not a poll in itself. So the really crazy polls that had Obama winning by 10 to those having him winning by .5 or losing by 1 are all averaged.
Polls like Gallup and Rasmussen had the polls tight and with Romney winning the last two weeks. Something happened Thursday before the election. Obama’s personal approval numbers climbed from 47 to 52 by election day. There are a few factors that went into that. I just don’t want to mention them now. I know Gallup and Rasmussen may not be as sexy as Nate Silver, but they have been historically very accurate. In fact, Rasmussen was the most accurate in 2008.
Having said all of that, the polls, on average, wound up being correct. The race was tight, and Obama turned out his constituents at levels most did not consider to be possible. Meanwhile, as many as 4 million Republicans and 9 million white voters decided not to vote. And voters like yourself switched their vote from McCain to Obama.
The fact that a lot of well informed folks thought Romney would win is not a case of “driving into a wall with their eyes open.”
There were many factors pointing toward a likely Obama defeat. And what do you know, a few very respected polls seemed to give evidence to that. They were wrong. I was wrong. A lot of us were wrong. Doesn’t have a damn thing to do with being Conservative or Republican. The election victory for Obama was a break from historical and political trends. In other words empirical observation that pointed towards a Romney victory wound up being an anomaly for Obama.
Yes, I said I would bet on a Romney victory. Give me the same scenario, with all the negative factors facing an unpopular incumbent, I’ll make the bet again. I’ll play the averages on that every time. And I would win 99 out of 100 times.
Jason,
First – Where I stand is where I stand. I am an independent, unlike you. I don’t have to carry the party line and support nonsense when it’s obviously patent nonsense. And right now, yes, there is a whole lot of nonsense in the “current” GOP. It is enough to give any independent minded person “vertigo”!!
Second – I cannot follow your logic about the polls. Nate Silver and others did statistical and probability based models and many of them showed Obama ahead for most of the race and showed an Obama victory the night before the election. Now, yes, RCP does publish an average of the polls. That’s better that just having Rassmussen and Gallup, both of whom were shown to be consistently more wrong than right. If I can average 6 or 10 polls together, continually, I have a better picture that just relying on 1 or 2 specific polls. That’s just good intelligence gathering policy. RCP, without all the probability and statistical models, was showing (through polling averages) who was favored over the entire election and going into election day. Obama was in-fact ahead just about the entire time polling averages were taken.
Third – The bigger point here is that while Silver and his crew were very accurate, we didn’t need them to tell us who was going to win. The RCP polling average told us without all the complicated modeling who was going to win that day. I believe, based on all the polls that were out there showing Obama ahead in the battleground states, that some Republican supporters we seeing what they wanted to (at, say, Gallup and Rassmussen) and not necessarily what was in plain sight through a more broader view of the overall polling averages — that didn’t give them the picture they wanted. Please just scroll down the page (in link below) and see what others were seeing in the polling — a sea of blue. It just is what it is. Obama was favored just about the entire election based on averages. His victory was really no surprise. As for the size of the victory, by the way, it’s now up to roughly 4 million votes and growing. Obama’s victory margin is now bigger than Bush’s (43) in both 2000 and 2008. A much bigger victory than I anticipated. And remember the Republicans have lost the popular vote in the last 4 out of 5 Presidential elections. It’s just not looking good…something I’ve been saying for over a year now. A course correction is needed.
http://www.realclearpolitics.com/epolls/2012/president/us/general_election_romney_vs_obama-1171.html#polls
Jeff, you advocate and argue for things that are left of center. Something you didn’t do as recently as two years ago.
You said this in 2010 in response to a post I wrote on the Keynesian school of thought
I think you like the idea of being independently minded but what you reveal suggest otherwise. Look, your positions nor your politics is at judge here. It doesn’t bother me in the least where you stand. Doesn’t make me think any less of you. But to be fair to me, you know I do not toe the line. Now I freely admit that I fall on the conservative side of major issues. In that regard, I am an independent mind as a member of the right. As a matter of fact, I was a registered Independent until 2008. But I don’t consider that some great virtue. At some point, you have to state what you are for, realize what you believe, and move forward with them.
I’ve disagreed plenty with the GOP over the years and stated it when I have. I vote Republican because, well, I sure can’t vote for a liberal Democrat for numerous reasons. I don’t consider that a partisan issue, but one of principle. There are two or three main things that prevent me from accepting most Democrats as a leader I would feel comfortable voting for. That’s just a personal position I have, not a personal indictment on you or anyone else who can or has.
You’ve developed in your mind recently a caricature of conservatives that just isn’t true. We are all a bunch of Limbuagh listening, flat tax advocating, government hating, ideological maniacs. It’s not true.
Meanwhile, you have towed the line as of late with many of the talking points and positions of the left. Yet, because you are an “independent” and a moderate, your path to these discoveries are superior to mine because you came about them through deep reflection, whereas I hold to my positions simply because I am conservative. I don’t agree with that. And I’ll disagree with you and if you wag your finger and say things that aren’t true or greatly debatable, I will debate them with you.
And what I find amusing about that, the more we dig in on our positions, according to you the more of a partisan I become while you are simply an independent mind in pursuit of the simple truth. When really, you are hiding your partisan side by appealing to the superiority of being an Independent.
As for the polls. What more can I say. They were right. Obama won. He broke a long held truth regarding incumbents associated with high unemployment, gas prices, unpopular legislation among other things, etc.