Nate Silver explains himself

November 4, 2012

Or, rather, defends state-by-state polling quite convincingly:

Of the 77 states with at least three late polls, the winner was called correctly in 74 cases. (I exclude Missouri in 2000, where the polling average showed an exact tie.) There has been little tendency for the state polling averages to overrate either Democrats or Republicans, or either incumbents or challengers. The state polls also performed fairly well in two years, 1996 and 2000, when the national polls were somewhat off the mark.

The chances of a miss are higher, of course, when the polls show a closer race. Even among the 33 cases where the final polling margin in a state was within five percentage points, however, the polling average identified the winner correctly in 30 cases.


Mr. Obama is at about 50 percent of the vote in the polling average in Pennsylvania, Wisconsin, Nevada and Michigan; at close to 49 percent in Ohio; and at about 48 percent in Iowa, New Hampshire, Virginia and Colorado.

There are not really any recent precedents in which a candidate has led by something like 49 percent to 46 percent in the final polling average, as Mr. Obama does now in Ohio, and has wound up losing the state. That does not mean such misses cannot or will not occur: there have only been a few elections when we have had as much state polling data as we do now, which is why the model allows for the possibility of a 1980-type error based on how the national polls performed that year.

But the reasonably high level of confidence that the model expresses in Mr. Obama’s chances of winning Ohio and other states reflects the historical reality that the polling average normally does pretty well.

There is only one way Mitt Romney wins on Tuesday: If the state polls are as biased towards Democrats this year as they were towards Republicans in 2010. Most swans are white, but black swans do exist.

8 Responses to Nate Silver explains himself

  1. Rdm on November 4, 2012 at 12:30 pm

    Except its not convincing. This swan is blatantly black due to flatly ridiculous samples.

  2. Jim,MtnView,CA,USA on November 4, 2012 at 12:41 pm

    “There is only one way Mitt Romney wins on Tuesday: If the state polls are as biased towards Democrats this year as they were towards Republicans in 2010.”
    Wait. State polls overshot in 2010 in favor of Repubs?

  3. Rdm on November 4, 2012 at 12:52 pm

    The same state polls that dramatically underpredicted republican gains, especially based on silvers model?

  4. oh boy... on November 4, 2012 at 12:54 pm

    Thing is Nate Silver has to explain how these polls are so strange. Romney up 10 with indies but losing? or how their early voting answers don’t match reality.

    Gallup and Rasmussen have much higher samples and more sophisticated likely voter screens.

    It’s easy to find inaccuracies in Gallup polls because they have done so much polling, but their track record is excellent.

  5. muckdog on November 4, 2012 at 1:10 pm

    Yeah but they dont think Romney is leading in independents

  6. Yancey Ward on November 4, 2012 at 1:19 pm

    The polls are actually starting to converge towards a 50/50 election that greatly resembles the 2000 election, with Ohio standing in for Florida. If you actually bother going through the various state polls, you find Obama with a narrow lead in Ohio, and Romney with narrow leads in Florida and Virginia. At this point, if I were going to use the state polls as a proxy (what RealClearPolitics calls the No Tossup Map), it gives Obama 290. Which means that even without Ohio, he can win.

    Now, if Romney wins the popular vote by 1% or more, I fully expect him to win Ohio and Colorado, and the election (and he may well get NH, too). For me, it still comes down to who wins the popular vote. I think it highly probable one candidate or the other will get over the 1% margin.

    For me, the real question is internals of the polls. There really is a big disconnect in the top lines and the support from independents. Either the polls are greatly overestimating Democratic support, or not.

    • Earl T on November 4, 2012 at 7:29 pm

      Where’s the so-called convergence”? Averaging polls as RCP does is like comparing apples to chainsaws. There is absolutely no way to compare the disparate manners of sampling and statistical weighting among the various polls. They can’t even agree on which election year to best model this year’s turnout, yet ALL of them assiduously avoid the 2010 turnout: Oops! 2010 just doesn’t exist!

      Further, nobody can predict how many voters who were totally disillusioned with McNutz in ’08 stayed home and whther they will return to vote “Republican” in ’12, particularly the evangelicals, whom McKrazy told to “F#*k off”. They don’t trust pollsters and aren’t even considered by the Lame Stream pundits and pollsters. GOTV is already favoring Repubs in early voting and the “R”s are famous for whupping up on Dems on Election Day!

      Hay, Barry! You close on that Hawaiian mansion yet?

      As for lil’ Natty Silver? It ain’t baseball, numbnuts!

  7. Dan on November 4, 2012 at 2:51 pm

    I am confident. Romney will win Florida,Virginia and Colorado,getting him to 257. If he wins Ohio,of course,it’s over. Now,more options seem to be out there just in case.
    As for Silver,he is pretty much everything I hate about pollsters,and in his previous day job,baseball.