It’s Entirely Possible…

…that Barack Obama wins this election (more at noon).

But as the situation among reputable polls shifts ever more to Romney, and as more and more hitherto “likely Obama” states flip to “Leans Obama” and “Toss-up” and even “Lean Romney”, it’s interesting to watch Nate Silver doubling and tripling down on his prediction; he’s still giving The Light Worker a 75% chance of winning.

I’m not a statistician – but I can read and reason, and I’ve been dinging on Silver’s polling, methodology and predictions for a couple of years now.  My beef – and I’d suspect the beef of any rational person who isn’t one of the incurious low-information voters at which Silver’s polling is aimed – is that he calculates his results based on weighting existing polls based on some proprietary secret sauce known only to him.

Is the “sauce” valid?  I don’t know – nobody does, really – but as I showed in the 2010 Minnesota Governor’s race, it involved giving exaggerated weight to polls like the absurd “Minnesota Poll”, the so-bad-it’s-out-of-business Humphrey Institute poll and the frankly left-leaning PPP poll, while systematically shorting polls like Survey USA and Rasmussen.

Is Silver right?  Even if I could check his math, I probably couldn’t check his math, if you catch my drift.  Maybe Obama still is a near-sure thing, even after this past three weeks; maybe the Dems and Silver know something we don’t (like how many dead people will be voting).  We won’t really know until next week.

But while there will be many things about a Romney win that I’ll applaud, one of the big ones for me, personally, will be dancing – rhetorically, of course – on Nate Silver’s professional grave.

13 thoughts on “It’s Entirely Possible…

  1. As I understand it, the wekaness in Silver’s methodology is that he keeps old results in his data for so long that they mask what is going on. It makes sense to keep results in for a while (for stability, and to reduce the effect of outlier blips or temporary bounces), but those polls, and especially their party ID weight throw off the results. This was not a problem for him in 2008 when he was so close to correct, because the polls mostly moved in only one direction. This year, when there has been a significant shift in polling and party enthusiasm (post first debate), his method fails to account for that.

    Don’t worry though. Nate Silver is not going to have a “professional grave.” The left will continue to rely on him for comfort, to tell them everything is going to be o.k. because people still like them!

  2. You are correct that it is entirely possible that the President will win a second term. But he won’t. Mitt will win. Not by a lot, but by more than enough. Mitt will also win WI, but alas, not MN. My dream scenario is that Mitt carries WI, and that is what pushes him over the edge. The President has a real problem with his base their, based on his failure to lead at all, in any way during the recall fiasco.

  3. Noticeably absent in the poll conversations this go-around is Zogby, which had a laughable record. They might still be around, but I haven’t heard a Zogby poll cited for some time.

  4. I know that Romney supporters have adopted a largely faith-based perspective on the state of the race, but Silver’s reason for believing Obama is ahead is hardly complicated.

    Romney needs to either win Ohio or run the table everywhere else. Every single poll shows Obama leading in Ohio; the chances that they’re all equally inaccurate are negligible. He also leads in enough other swing states that Romney’s never going to win all the rest.

    Now, the polls could still change in the next week. But given what the polls show us now, Obama absolutely has the lead. Romney has a slight lead nationwide, but that’s driven by massive margins in the South. In the swing states, Obama’s leading, and no amount of “poll denialism” will change that.

  5. Every single poll shows Obama leading in Ohio; the chances that they’re all equally inaccurate are negligible.

    No longer true, Jeff. Rasmussen has Romney up two and the poll over the weekend from the Ohio newspapers showed it 49/49.

    I also suspect Romney will win Wisconsin, Iowa, Colorado and New Hampshire. I have Romney with 295, counting Ohio.

  6. I know that Romney supporters have adopted a largely faith-based perspective on the state of the race,

    Wait – you guys think “Faith-Based” is a burn, right?

    but Silver’s reason for believing Obama is ahead is hardly complicated.

    Indeed, it’s conventional wisdom. No, Jeff, I got that. Ohio is a big deal. Stipulated.

    Every single poll shows Obama leading in Ohio

    Well, no, they don’t.

    Obama’s leading, and no amount of “poll denialism” will change that.

    Jeff, using a term like “denialism” puts a huge rhetorical target on your head. It implies that polling is an empirical, precise science, and those who criticize it are like the peasants who storm the observatory with pitchforks and torches to keep the witchcraft in the bottle.

    It’s just not so.

    Could Obama win? Sure. You Democrats had almost best hope not; if he does win, he may well be the weakest President in history;; against a solid GOP House and a Senate that’s going to be deadlocked if not GOP, another Obama term will be a disaster, and will set the stage for the most devastating electoral riposte in history in 2016, and likely take down even more of your governors and state legislators in the process. It also increases the chance that the Dollar will completely collapse on the Democrats’ watch, leading either to the marginalization-unto-extinction of the Democrats (good) or the collapse of the country (likely bad, depending on what part you live in).

  7. Jeff is under the 100% belief that Ohio is going for Obama. Jeff not only has Romney caught Obama in Ohio one poll has Romney ahead by two and two tied there is something more to be worried about if you’re an Obama fan. There has been careful studying of the early voting in Ohio and Obama is down something like 250,000 votes from his early vote totals in 2008 (basically his margin).

    But lets play a game of pretend here and magically give the state of Ohio to Obama. Since we know that Obama is losing Florida (Obama trailing), North Carolina (Obama trailing), Virginia (tied but I’m giving to Romney for a reason I’ll explain in a second). That gives Romney 248 electoral votes. You think that Obama has enough of a margin to win the swing states especially after he wins Ohio. Here’s where the second problem you’re encountering Jeff is that the only person who is going to be voting for Obama is somebody who has already made up his mind right now.

    In Virginia Obama has 47.8% of the vote which is why I give it to Romney (not to mention early voting for Obama is way down in key areas he has to win). In Colorado Obama according to OCB has just 47.8% of the vote and his early voting is down compared with 2008. Give Romney 9 electoral votes. In Iowa Obama does have 49.0 a RCP thus giving him a shot to win until you realize that is skewed by a Wall Street Journal poll two weeks ago. Throw that out he is way under 50.0 and four major Iowa newspapers have endorsed Romney. Not to mention Bush did win Iowa in 2004. Advantage Romney and he gets their 6 electoral votes. That’s 15 electoral votes.

    Now according to my rule Minnesota and Michigan will both go to Romney. But lets get to Wisconsin. There has been one very recent poll (other polls are two weeks old) and it’s tied 49-49. Now if it was a matter of life and death to get Obama reelected in Wisconsin it was more then life and death to the lef that they kick Scott Walker out when the recall took place. Scot Walker’s organization and voters see we have to do for the country what was done for Wisconsin. Wisconsin and their 10 votes for Romney.

    So Jeff I don’t see how Obama is winning even if Ohio magically appears in his column. What you’re looking at is a 53-47% popular vote win for Romney and Romney breaking 300 electoral votes.

    Walter Hanson
    Minneapolis, MN

  8. Jeff, even if you continue to ignore the polls that show Romney ahead in Ohio and elsewhere, the polls that show him tied/very close have lots of good news for Romney. Particularly (as noted above) the President’s collapse in early voting. More notable is the President’s failure to sway independents, and his loss of their confidence on the economy.

    Given that, things look very bleak indeed for the President. Your faith based denial of discouraging polling data won’t save your savior.

  9. Jeff, I want to be delicate about this, but you have to go out beyond polls of your friends and family. The grownup polls show Romney ahead.

    You need to stop relying on the New York Times for your “news”.

  10. I once read a story in which a scientist did no actual research of his own, but merely read the studies performed by others, then decided which he thought sounded most convincing. Sounds as if Nate Silver has a similar approach. Did I mention it was a Science Fiction story?

  11. Joe, wasn’t that report the one on the doctoring of gloBULL warming reports out of East Anglia U? 🙂

  12. Jeff Rosenberg:

    A quick question for you: WHY, in terms of methodology, do you believe Silver?

    Just trying to figure out who is actually “Faith-Based” here…

    Also – you refer to a bulge of support for Romney in the South. OK – do you think that overwhelms the bulge of voters for Obama in NY, IL, CA, MA and PA?

    If so, please explain why.

  13. Mitch:

    I’ll add this question for Jeff. If he thinks Wisconsin is in the bag for Wisconsin how come the left when it was a matter of life and death couldn’t get Scott Walker voted out of office ealier this year? They had the money and man power they wanted which suggests that Obama can’t win Wisconsin.

    Walter Hanson
    Minneapolis, MN

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.