The Other McCain

"One should either write ruthlessly what one believes to be the truth, or else shut up." — Arthur Koestler

Polls Continue to Show Trend Toward Romney — Nate Silver Notwithstanding

Posted on | October 25, 2012 | 33 Comments

Excuse me for my continued attention to Nate’s graveyard-whistling, but no matter how clear the evidence of a pro-Romney trend, the Grand Swami at the New York Times won’t stop. He’s now raised the likelihood of Obama’s re-election to 71.0%. (The one-tenth of a percentage point being necessary to the pretense of scientific exactitude.)

Is Nate Silver hustling an insider-trading scam with InTrade? Or is he merely acting as an Obama pompom girl? Either way, the poll-watcher at the nation’s most influential newspaper cannot be unaware of how his coverage functions to shape elite opinion, which is in turn reflected in other media coverage that then influences mass opinion, and believing that Nate Silver is acting as an honest neutral broker in this transaction requires a faith in human goodness that I lack.

When I looked at the Real Clear Politics national average this morning, my eye was automatically drawn to the ABC/Washington Post poll that shows Romney now leading 49-48. This strikes me as significant because two weeks ago (Oct. 10-13) the same poll had Obama leading 49-46. Thus they are reporting a net pro-Romney shift of +4. It doesn’t matter for this purpose, by the way, whether ABC/WaPo is accurate as to the actual number, so long as their method is consistent — the trend is what it is. Meanwhile, both Gallup and Rasmussen tracking polls have Romney leading by 3 points, and the latest Associated Press poll has Romney ahead by 2 points.

“But, but, but . . . Ohio!”

Yeah, I know: Obama’s poll numbers in the Key Battleground State of Ohio — as it is always called — are consistently better than in the national polls, but I do not believe that any amount of advertising, campaigning or ground-game organizing in Ohio is going to enable Obama to overcome at a purely local level what is a clear national trend toward Romney.

That is to say, if Mitt’s pulling ahead nationally, Ohio will inexorably move in the same direction, even if the shift is smaller. So when we look at the Ohio polls and see three that show it at tie, and five others showing Obama ahead by 1%-5% margins — a net advantage of 2.1% in the RCP — we have to weigh two facts (a) Obama’s small but persistent lead, and (b) the national trend toward Romney. If the trend continues, Ohio will follow that trend, and subsequent polling will reflect it.

By the way, that Time poll of Ohio is so hinky as to be dubious.

Time hasn’t previously polled Ohio at all this cycle, and for them to come in during the last two weeks of the campaign with a poll on the high side of the margin — matching the Obama +5 of PPP and Quinnipiac — is highly dubious. You can go chew on Time’s sample if you want, but the statistical details are irrelevant to my basic suspicion. If it walks like a scammy poll and talks like a scammy poll, it’s probably not a duck.

We do not know, by the way, whether the Romney trend will continue over the next 11 days. Polls are a lagging indicator and you can’t extrapolate numbers forward. Welcome to the Twilight Zone phase of the campaign, where things become mysterious. But be encouraged by Ed Morrissey’s analysis of the new AP poll:

Now Romney leads without leaners by three points, 45/42 — and a firm number of 42% for a incumbent is a disaster 12 days out from the election. When the “could change mind” voters are taken out, it’s even worse — 43/41 for Romney, with the incumbent barely above four out of ten voters.

Too early for predictions, but an incumbent persistently below 50% in October is usually what polling experts call “a loser.”

UPDATE: Ted Frank offers an informed statistical critique of Nate Silver’s methodology, which is different than my own criticism, namely that sabremetrics doesn’t work for polls, because politics ain’t baseball. Political opinion and voter behavior differ from the performances of baseball players, and “weighting” polls based on past performance cannot overcome the fact that polls are a lagging indicator. Nate Silver’s pretense of scientific exactitude — his constant and increasingly obnoxious reference to his own self-proclaimed expertise — is itself a factor that may shape public opinion, and he ought to damned well know it.


33 Responses to “Polls Continue to Show Trend Toward Romney — Nate Silver Notwithstanding”

  1. Evi L. Bloggerlady
    October 25th, 2012 @ 11:39 am

    It is going to be a close race.

    What is with these swearing Democrats?

  2. Kelly
    October 25th, 2012 @ 12:00 pm

    Thanks for this! Being a Republican statistics major in an election year is not easy when Nate Silver is the end-all-be-all of celebrity statisticians

  3. Adobe_Walls
    October 25th, 2012 @ 12:12 pm

    Another “trend” is the polls of less than 1000 LV have Comrade President ahead those of over 1000 have Romney leading. Breaking that trend today at RCP is AP/GfK with Romney at plus2. Perhaps a new trend developing.

  4. Mark Jordan
    October 25th, 2012 @ 12:28 pm

    Nate Silver’s secret collusion with the Obama campaign in 2008 renders his entire model suspect. It seems to me what he did then — and for NY Times to then hire him — borders n criminal.

  5. Charles
    October 25th, 2012 @ 12:49 pm

    My favorite is Nate’s invocation of favorite-long shot bias. Obama is certainly his favorite, but Romney is a long shot in his mind only. A long shot is 50 to 1 or 100 to 1. Jill Stein is a long shot. Virgil Goode is a long shot. Romney is running neck and neck.

  6. Roxeanne de Luca
    October 25th, 2012 @ 12:49 pm

    ….AND Romney has $169 million cash on hand, and the RNC has $83 million cash on hand….

    Welcome, mother of all final-week advertising! Bwaaaaahhhh!

    October 25th, 2012 @ 12:52 pm

    “[B]ecause politics ain’t baseball.”

    But if it were, Romney hit a Grand Slam in the first debate and Obama got intention walks by the media in Debates #2 and #3.

    Score: Romney-4; Obama-zero. ;^)

  8. Adjoran
    October 25th, 2012 @ 1:06 pm

    BTW, that Time poll is D+11 sample, so there’s that.

    Silver is in for a heapin’ helpin’ of scorn, derision, and mocking, I hope.

    For the record: No one who has ever carried the national popular vote by as much as 0.5% has ever lost Ohio. Ohio goes with the national trend more than any other state.

  9. CalCon10
    October 25th, 2012 @ 1:14 pm

    Nate Silver reminds me of the juveniles who run baseball sabremetrics sites. “We’re right; the numbers we invented prove it. And if you don’t believe it, we’ll show you even more numbers we invented to prove it.”

    That’s like a Catholic (like me) telling an Evangelical, “You’re wrong, because the Pope says…” Silly, offensive, and utterly dishonest.

  10. Wombat_socho
    October 25th, 2012 @ 1:18 pm

    Well, fortunately there are a lot of adults who realize there’s more to sabermetrics than inventing useless new statistics…which is really not what sabermetrics was intended to be at all, according to its inventor Bill James.

  11. Mikey NTH
    October 25th, 2012 @ 1:20 pm

    It is one thing to look at what an athelete has been able to physically accomplish and take into account aging, injuries, and so on to predict what he will do in the near future. It is much of another thing to predict where an opinion is going to land because a lot of people do not have solid opinions on anything political, but rather have fuzzy preferences, and it is difficult to tell what event out of any event may sway a preference (and it may be something that no campaign could influence anyhow).

  12. jwallin
    October 25th, 2012 @ 1:40 pm

    Yeah, exactly. Since he’s basically spit balling the “weight” he gives each “poll”, those extra 2-3 decimal places are window dressing for the mathematically challenged Liberals he writes to and for.

    Nothing like going out 5 decimal places to sound like you’ve really cooked up some accurate math.

    Trouble is and any computer tech can tell you; Garbage in == Garbage out.

  13. Romney still has momentum, Nate Silver hardest hit « The Daley Gator
    October 25th, 2012 @ 1:59 pm

    […] but I know people like Silver so you do not have to, and The Other McCain certainly knows Silver, and his whistling in the graveyard act Excuse me for my continued attention to Nate’s graveyard-whistling, but no matter how clear the […]

  14. K-Bob
    October 25th, 2012 @ 2:30 pm

    Yes. People keep thinking winning Ohio is critical, because you can’t win without it. Well, it’s the other way ’round. If you won, you got Ohio. You can’t win by focusing on Ohio, you win by focusing on winning.

  15. EsausMessage
    October 25th, 2012 @ 3:04 pm

    Interesting thing about ABC/WaPo, other than general trend, is that the previous week’s poll showing a 3 point O lead was D+5, but this week’s poll showing a 1 point R lead was D+9. So the trend toward Romney may be even more pronounced than the top-line number indicates.

  16. Jack Marchetti
    October 25th, 2012 @ 3:45 pm

    If you think of Nate Silver as a weatherman, which in essence is what he’s doing, then perhaps people wouldn’t get so upset about it. If he’s right, his career takes off even more. If he’s very wrong, then he’s done. So I’m not sure where the hostility comes from.

    If you don’t like his approach, don’t read it.

  17. Jack Marchetti
    October 25th, 2012 @ 3:47 pm

    Well he is a sabremetrician who sold his baseball modeling program when he was like 20 years old. If you’re not a SABR guy that doesn’t really matter to you.

    Not sure what your catholic/pope analogy even means.

  18. Wombat_socho
    October 25th, 2012 @ 3:59 pm

    But he’s not acting like a weatherman, he’s acting like a sabermetrician, and trying to apply the predictive techniques that worked fairly well with baseball players to politics. This gives his work a veneer of believability that it doesn’t deserve, for reasons other I and other commenters have mentioned. Worse, he’s doing this in a high-visibility position at the NYT, which is still used s a reference by a lot of lazy sods in the MSM, so a lot more people see this than if he was some young swot posting at DU or Daily Kos. So failing to thrash him early and often for his Juicebox Mafia aspirations would be a betrayal of our mission to whack the MSM for doing stupid crap like this.

  19. CalCon10
    October 25th, 2012 @ 4:12 pm

    Your snarkiness indicates a need to have things explained to you very slowly and carefully. I’m pleased to oblige.

    The “catholic/pope” (both proper nouns, gentle jack: I notice you were careful to capitalize SABR but not Catholic or Pope) analogy just means: it’s futile, and dishonest,to base arguments on the personal opinion/authority of someone whose validity your opponent doesn’t accept.

  20. R.V. Levin
    October 25th, 2012 @ 4:31 pm

    To irrationally optimistic Republicans: Haven’t you heard of the electoral college?

  21. Adjoran
    October 25th, 2012 @ 7:29 pm

    We know more about it than Obama and Biden put together know about math and finance.

  22. Adjoran
    October 25th, 2012 @ 7:31 pm

    Just so. He – and the NYT for hiring him – must be mocked and humiliated relentlessly when he turns out to be wrong, else he will retreat to the dishonest defense of the errant pollsters that there was an unforeseeable late move.

  23. Barack Obama at 47% in the Polls | hogewash
    October 25th, 2012 @ 9:50 pm

    […] Barack Obama at 47% in the Polls Posted on 25 October, 2012 by wjjhoge The President seems to be stuck at around 47 percent in the polls. (See the chart here.) […]

  24. DonaldDouglas
    October 25th, 2012 @ 11:06 pm

    Silver’s an idiot.

  25. RANDER1010
    October 25th, 2012 @ 11:15 pm

    well if you think Romney has momentum, here’s your chance to get rich because the books are giving you 2-1. Oh, free advice, your bookie knows more than you do.

  26. Dodd
    October 26th, 2012 @ 12:32 am

    I found out everything I needed to know about Silver’s “analysis” when I learned that he was overweighting a week-old PPP poll that showed Obama well ahead and underweighting a fresh poll *by the same pollster* that had Obama’s lead almost gone in his “model.” The adjustments were not small. So he was purposefully and consciously giving a large amount of additional weight to old data compared to new data despite the source of both being the same.

    This statistical legerdemain is justified by an ipse dixit assertion that PPP’s Dem lean has diminished over the cycle (a change he would have us believe has significant, observable effects over a 7-day period). Then there’s the grotesquely Obama-positive Marist poll to which he gave a huge statistical bump compared to a slew of more recent data for almost two weeks. That poll had a D+11 sample–over twice Obama’s 2008 ID edge and therefore something no rational poll observer could possibly argue with a straight face was likely to be valid. Nate loved it, though, giving it massively outsized influence over his overall results long after it was stale data.

    These are indicia of someone working the refs, not someone trying to make accurate, objective predictions.

  27. Evi L. Bloggerlady
    October 26th, 2012 @ 3:31 am

    You might be sad on election night.

  28. Are You a ‘Nate Silver Truther’? : The Other McCain
    October 26th, 2012 @ 7:16 am

    […] among independents.Back to the question, then: What the f–k is Nate Silver talking about? A commenter on yesterday’s post suggested it’s non-random ”weighting”:I found out everything I needed to know […]

  29. Larwyn’s Linx: Benghazi-Gate Boils Over: Evidence that Obama himself denied security | Preppers Universe
    October 26th, 2012 @ 10:02 am

    […] CowboyWI: Assaulted openly gay GOP campaign volunteer talking to FBI, sources say: WatchdogMediaPolls Continue to Show Trend Toward Romney — Nate Silver Notwithstanding: RSMThe Slow, Painful Coming Death Of The Independent, Liberal Blogosphere: RWNAP poll: Romney […]

  30. me
    October 28th, 2012 @ 1:25 pm

    Also, don’t forget they said you couldn’t win the Republican nomination without winning South Carolina.
    This race is a very different animal than previous ones. There are just too many unknowns floating about, such as… where the hell is all the Obama swag from 2008? The absence of it must mean something at least..

  31. K-Bob
    October 28th, 2012 @ 3:33 pm

    Fraud is less expensive than swag.

    That’s my guess.

  32. Mark October 30 on Your Calendar: Democrat Panic Hits Pandemic Stage : The Other McCain
    October 30th, 2012 @ 2:17 pm

    […] 25: Polls Continue to Show Trend Toward Romney — Nate Silver NotwithstandingOct. 23: Signs and Omens: Obama’s Fading Hope and the Graveyard Whistling ChoirOct. 22: The […]

  33. Pete Stream
    November 8th, 2012 @ 3:02 pm

    2 days after the election makes this blog post sad.