Skip to main content

Silver is over estimating Romney's chances:

Well, 538 is projecting Obama's chances for a win as 68%. In today's column he mentioned that there are other sites that give Obama much higher chances than that and provided alink too.

There the projections are like "almost entirely to the right of 270 (Huffington Post)", "Nine to one for Obama (Sam Wang, Princeton Election Consortium)", "85% (DeSart and Holbrook election forecast)"

The votematic blogger concludes:
Nevertheless, The Daily Caller, Commentary Magazine, and especially the National Review Online have all run articles lately accusing Silver of being in the tank for the president. Of all the possible objections to Silver’s modeling approach, this certainly isn’t one that comes to my mind. I can only hope those guys don’t stumble across my little corner of the Internet.

Well, Nate made me stumble across his little corner.

America does not react instantaneously.

I think we political junkies watch things very closely and we are expecting large things polling and seven day moving averages to move quickly after the debate. Understandable, given 60 million people watched each debate. But people might form some instant opinion, but it gets reinforced are weakened by other influences, media narrative, fact checking, late night comedian's zingers, etc etc. A nation of 330 million people do not jump instantly after the debate.

After the first debate it took almost 1 week/10 days for Romney to reach his high watermark. The second and third debates came in quick succession, and it would take a while for the electorate to digest the debate, the media news, and the jokes and the stuff.

It looks very much like, Joe Biden stopped the bleeding, the second debate stopped the swing, and the third debate has reversed the momentum. I am very sure CO, NV, OH, IA, WI will move solidly blue in the next few days. VA will be light blue and Florida will be back in play. I think even if we lose NC we will come very close and make them sweat a little.

My take on what is going on:

In the Republican convention, Romney looked like a loser, and the Republican base stayed at home. The first debate provided the excuse the base needed to come home, and give Romney the liberty to shake the etch-a-sketch. None of the Republican splinter groups want to oppose Romney when he had the big swing and he appeared to be winning.

If that narrative is lost and once the Republican insiders give up on Romney, they will splinter and try to make sure their turf is protected. Glenn Beck sounded the discordant tweet first after the third debate. Now this senators are burnishing their "anti-abortion" credentials. Karl Rove must be apoplectic now, his coalition is unraveling. Soon the 2016 contenders would take positions to help their 2016 primaries. They will try to set up a narrative, "if only you guys have followed me... in blah blah we would have won". They will isolate Romney and make the loss his personal failure not a reflection of the suckiness of the Republican brand.

That is when we know what the internal polling is revealing to the Republicans.

Is it wishful thinking? Am I deluding myself? Perhaps but I seem to be good company. Anyway as Churchill said, "We can trust America to do the right thing, once all other options have been exhausted". Yes, America will do the right thing.

Your Email has been sent.
You must add at least one tag to this diary before publishing it.

Add keywords that describe this diary. Separate multiple keywords with commas.
Tagging tips - Search For Tags - Browse For Tags


More Tagging tips:

A tag is a way to search for this diary. If someone is searching for "Barack Obama," is this a diary they'd be trying to find?

Use a person's full name, without any title. Senator Obama may become President Obama, and Michelle Obama might run for office.

If your diary covers an election or elected official, use election tags, which are generally the state abbreviation followed by the office. CA-01 is the first district House seat. CA-Sen covers both senate races. NY-GOV covers the New York governor's race.

Tags do not compound: that is, "education reform" is a completely different tag from "education". A tag like "reform" alone is probably not meaningful.

Consider if one or more of these tags fits your diary: Civil Rights, Community, Congress, Culture, Economy, Education, Elections, Energy, Environment, Health Care, International, Labor, Law, Media, Meta, National Security, Science, Transportation, or White House. If your diary is specific to a state, consider adding the state (California, Texas, etc). Keep in mind, though, that there are many wonderful and important diaries that don't fit in any of these tags. Don't worry if yours doesn't.

You can add a private note to this diary when hotlisting it:
Are you sure you want to remove this diary from your hotlist?
Are you sure you want to remove your recommendation? You can only recommend a diary once, so you will not be able to re-recommend it afterwards.
Rescue this diary, and add a note:
Are you sure you want to remove this diary from Rescue?
Choose where to republish this diary. The diary will be added to the queue for that group. Publish it from the queue to make it appear.

You must be a member of a group to use this feature.

Add a quick update to your diary without changing the diary itself:
Are you sure you want to remove this diary?
(The diary will be removed from the site and returned to your drafts for further editing.)
(The diary will be removed.)
Are you sure you want to save these changes to the published diary?

Comment Preferences

  •  Nate's model is good (10+ / 0-)

    Nate's model is different than everyone elses and will produce different results.

    Take all the models together and you'll get a good idea of the race.

  •  Nate Silver (4+ / 0-)
    Recommended by:
    Floande, Gooserock, tapu dali, greengemini

    we all like Nate and his volume of work. However, being connected to the MSM now  maybe he is a little like all the rest of the MSM: the closer things look the more interest in my NY Times column and everything else put out their. I think we are foolish if we don't consider Mr. Silver pretty much like anyone else tied to the NYT or any other giant medai outlet.

  •  asdf (13+ / 0-)

    Nate Silver and Sam Wang were the two most accurate predictors in 2008.

    And their current models agree almost perfectly.

    Obama with 290+ EVs and about a 2-1 chance of victory.

    Sadly, everything Communism said about itself was a lie. Even more sadly,, everything Communism said about Capitalism was the truth.

    by GayIthacan on Wed Oct 24, 2012 at 05:20:46 PM PDT

    •  Both (1+ / 0-)
      Recommended by:

      of their projects match oddmakers.  And the models were developed separately.  

      Over time the math will add up or it won't.

      "There are four boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order." Ed Howdershelt

      by Lava20 on Wed Oct 24, 2012 at 05:23:58 PM PDT

      [ Parent ]

      •  Wang has Obama at 90% (2+ / 0-)
        Recommended by:
        Iseeurfuture, elwior

        Nate is the outlier of the objective statistical aggregators.

        •  asdf (3+ / 0-)


          OBAMA - 289  ROMNEY - 249


          OBAMA - 293  ROMNEY - 245

          Sorry - I see little differece.

          Sadly, everything Communism said about itself was a lie. Even more sadly,, everything Communism said about Capitalism was the truth.

          by GayIthacan on Wed Oct 24, 2012 at 05:37:31 PM PDT

          [ Parent ]

          •  But Wang projects 90% odds for Obama (1+ / 0-)
            Recommended by:

            Though they both predict the same mean, the probability they project are different. Nate 68-70% for Obama. Wang 85-90% for Obama.

            •  I believe Wang is like Nate's "Now-Cast" (1+ / 0-)
              Recommended by:

              That is, he is trying to predict "if the election were held today."

              Not that he and Nate will always agree on that basis, but we should compare apples-to-apples.

              Numbers are like people . . . Torture them enough and they'll tell you anything.

              by Actuary4Change on Wed Oct 24, 2012 at 06:00:53 PM PDT

              [ Parent ]

          •  If (1+ / 0-)
            Recommended by:
            Demi Moaned

            you use more numbers then you increase certainty.

            Promise you Wang would not place any money on a 90 percent win with the stunts and stupidity of the people we are dealing with.

            "There are four boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order." Ed Howdershelt

            by Lava20 on Wed Oct 24, 2012 at 05:41:42 PM PDT

            [ Parent ]

          •  Wang has stated his belief that Obama has 90% odds (1+ / 0-)
            Recommended by:

            He doesn't have a break-out probability like Nate does, but Wang believes, based on his model, that Obama should be seen as about a 90% favorite to win.

            The pleasure of hating...eats into the heart of religion...[and] makes patriotism an excuse for carrying fire, pestilence, and famine into other lands. - W. Hazlitt

            by rfahey22 on Wed Oct 24, 2012 at 05:42:52 PM PDT

            [ Parent ]

            •  i (0+ / 0-)

              Also think there is a warren effect that needs to be consider with both.

              Birthers have been working at it for four years.

              Popular vote odds seem a factor here.

              "There are four boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order." Ed Howdershelt

              by Lava20 on Wed Oct 24, 2012 at 05:57:02 PM PDT

              [ Parent ]

        •  When (0+ / 0-)

          money is on the line, markets and models may vary.  I believe in markets and look for the subtle trends etc. for each one.

          One must factor in stunts and the campaign itself.  You have to consider that everything will have to go according to plan the entire time for that 90 percent to happen.

          No one would risk the loss Wang projects.  You have to factor in the stupidity of people and the shenanigans they will try.

          "There are four boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order." Ed Howdershelt

          by Lava20 on Wed Oct 24, 2012 at 05:40:07 PM PDT

          [ Parent ]

    •  True. But that was BEFORE Nate sold (0+ / 0-)

      538 to the NYT.

      I know you believe you understood what you think I said, but I'm not sure you realize that what you heard is not what I meant. -- S.I. Hayakawa

      by tapu dali on Wed Oct 24, 2012 at 05:59:44 PM PDT

      [ Parent ]

    •  Odds... (0+ / 0-)

      Sam Wang has the odds of an Obama win much higher -- he said 9:1 on NPR, but that seems high from his graphs.

      The real question is whether the errors are going to be dominated by random errors related to the statistics of polling, or by systematic errors (voter suppression etc).  Nate attempts to correct for systematics he can identify, Sam doesn't, but tries to incorporate it into his "confidence" windows.

  •  given the closeness of polling (10+ / 0-)

    at both the state and national level, a model that gives Obama over 2-1 odds of winning, as Nate's does, is actually kind to us, and feels about right.

    oops. I hope the gate wasn't too expensive.

    Twitter: @DanteAtkins

    by Dante Atkins on Wed Oct 24, 2012 at 05:20:50 PM PDT

    •  Actually, no. (0+ / 0-)

      Early voting in Ohio is off the charts.  Poll after poll shows that Obama leads early voting by 2 to 1 (somewhere around 60% to 30%) and that  about 20% of Ohio has already voted.  As Ohio comes out to the tune of around 63%, that means almost a third of the Ohio vote is already in, and it isn't pretty for Romney.  If the numbers of Obama's advantage are correct (and today's TIME poll confirmed the findings of a 2 to 1 Obama advantage shown earlier by other polls)  Ohio is at this point almost unwinnable for Romney.  He would have to surge to a 10% lead in Ohio by election day, something like 56% to 44%, to make up for the votes Obama has already banked.  That is not going to happen.  The TIME poll shows Romney and Obama exactly tied amongst voters who haven't cast their votes yet, Romney would have to surge by 11% from that basis to get to where he needs to be within the next 12 days.  

      Then you have Nevada, which is going absolutely great for Obama.  Early voting is off the charts there.  Nevada is close to be taken off the battleground list as it is, given the realities on the ground there.  Romney has not led in a single poll of Nevada all year.    

      So, with Ohio and Nevada looking great, basically "done"  and solidity in PA, MI and WI, you are already at 271 EC votes.  That before you get to Iowa, which is looking good for Obama, Florida and Virginia, which are both tied, as well as Colorado, New Hampshire and North Carolina.  

      I think in light of the realities we are seeing here, 90% seems about right.   A sudden Romney surge, now less likely than ever in light of the last debate and the Murdock issue, in Wisconsin could undo that certainty, but Wisconsin has universally polled badly for Romney, who hasn't led or tied in the state over the last 16 polls, dating all the way back to 8/16, and that was a slight 1% lead in what looked like an outlier poll from Ras.  Ever since then Obama has been on solid ground, and gets at 50% or above in almost all the polls we have seen.

  •  how DARE nate not go along with the rw script! (1+ / 0-)
    Recommended by:
    KJB Oregon

    kkkarl rove doing what's worked in the past: pushing a fake narrative about imaginary momentum.

  •  The line that keeps running over and over (2+ / 0-)
    Recommended by:
    Persiflage, highacidity

    In my head.....rats jumping ship, rats jumping ship

    It's festering as we speak.  Only a few days before the circular firing squad starts.

    Republicans===the party of the 1% rich people in America. Or in other words..The Party of NO!

    by jalapeno on Wed Oct 24, 2012 at 05:23:34 PM PDT

  •  Relax (2+ / 0-)
    Recommended by:
    wishingwell, TooLittleSleep

    That was yesterday and it was before the first post-debate polls came in. Now that they are coming in it's going for Obama and the Update later tonight will reflect it.

  •  He hasn't updated his forecast with today's polls (4+ / 0-)
  •  I think Nate is pretty much on target (2+ / 0-)
    Recommended by:
    wishingwell, ladybug53

    I do expect some incremental shifts towards Obama between now and election day so I expect his lead in the battleground states to firm up and Nate's model will reflect that as... and if... it happens. Whether it will be enough to secure Virginia, Florida and North Carolina for the President remains to be seen. I suspect so consequently I have not moved from my own early prediction of 347-191. But I would not be surprised if it ends up 303-235 either.

    Cuz, ya know, it's kinda close.

    "Do what you can with what you have where you are." - Teddy Roosevelt

    by Andrew C White on Wed Oct 24, 2012 at 05:26:11 PM PDT

  •  I'm Not Sure We've Factored In the Ad War. (8+ / 0-)

    It was ramping up at & after the 1st debate. I notice that Mitt has been reported several times having great days when he didn't do anything newsworthy in several days.

    This is the first Citizens United Presidential election. In that sense historic examples are all incomplete.

    We are called to speak for the weak, for the voiceless, for victims of our nation and for those it calls enemy.... --ML King "Beyond Vietnam"

    by Gooserock on Wed Oct 24, 2012 at 05:26:32 PM PDT

  •  Back in September this site was afire with (4+ / 0-)

    diaries making fun of GOP "poll truthers" and whining about skewed polls.  For the past 3 weeks if I've read one diary claiming that the polls don't accurately reflect Obama's lead, I've read 100.

    Not much shame or sense of irony, I'd say.

    The only poll that counts is 13 days from now, and the post mortem on who voted, who didn't, and which demographic voted for whom will be interesting.

    I'll be looking for the stats on women and those under 30 with much anticipation.

    Oregon:'s cold. But it's a damp cold.

    by Keith930 on Wed Oct 24, 2012 at 05:32:21 PM PDT

  •  i think nate is wrong (0+ / 0-)

    i say obama has a 85% chance of winning. silver's is good with numbers, but he is not always correct.

  •  68% is pretty damn good, (5+ / 0-)

    and I bet that number goes up tonight. Today was a very good polling day for the president.

  •  Recent polls are going to push it higher. (4+ / 0-)
    Recommended by:
    KJB Oregon, elwior, ladybug53, shamuson

    Nate is just reporting what his statistical models are producing. The numbers are what they are, based on the polling data that is input into the model.

    The recent polls are moving in Obama's direction today, it seems. Notably so. If that's the case, his model will begin to reflect it. I suspect that likelihood value will go up today when Nate does his update, and that we'll see further climbs in the coming days.

  •  Ya know, I'm not a poll junkie because no matter (3+ / 0-)
    Recommended by:
    ladybug53, evangeline135, highacidity

    what the numbers say, almost no one out there on main street cares.  And, they're the ones that actually vote.  I can't imagine a poll saying obama is winning or losing would cause me, or anyone I know, to either not vote or change their vote.  So, while it's nice to see data saying what one hopes will be true, it doesn't change much.  Tomorrow I'll be at the local OFA office...and I won't be quoting polls.  

    The longer I live, the clearer I perceive how unmatchable a compliment one pays when he says of a man "he has the courage to utter his convictions." Mark Twain

    by Persiflage on Wed Oct 24, 2012 at 05:51:15 PM PDT

  •  He has been kind to Romney (3+ / 0-)
    Recommended by:
    elwior, greengemini, Orlaine

    and I'm not simply referring to his numbers, but also to his accompanying punditry about those numbers.

    Whether or not people want to admit it, Silver is now toeing the MSM line, which is to go easy on Willard, and hard on President Obama. Nate described Obama's numbers as "awful" after a slew of flawed Republican leaning polls were released following the first debate. Yet he never came close to describing Romney's numbers in that way even after the DNC convention when Obama was enjoying a surge in the polls.

  •  Nate will update tonight and Obama will get back (1+ / 0-)
    Recommended by:

    to 70%.  He looks for movement in a candidate's direction.  The more uniform the movement across all types of polling, the clearer the indication that the candidate is actually surging.  Today's polls show movement towards Obama in all but 2 polls.  The state polls released are few but definitely positive for Obama.

    If this bounce is sustained through the weekend, then you'll see Nate increase Obama's chances closer to 80%.

    He would really need to get a couple of great polls in VA, CO and FL to exceed 80%.  If Romney has to flip 2 states to get to 270 he is pretty much done.

    Alternative rock with something to say:

    by khyber900 on Wed Oct 24, 2012 at 06:05:29 PM PDT

    •  I really don't see how polls in VA, CO and FL (0+ / 0-)

      should change the "win" probability.  Ohio and Nevada are looking incredibly good for Obama in early voting to the point where we all should be very bullish about those two states.   Romney would need next to a miracle to overcome his early vote deficits in both states.   Then you have extremely solid polling out of Wisconsin, where Romney hasn't led all year, and where Obama is at 50% or above in almost all the polls, and you are at 271 EVs.  That is not counting Iowa, Colorado, New Hampshire, Virginia, Florida.     Something extraordinary would have to occur to undo this, a major surge towards Romney in Wisconsin, which seems highly unlikely.  

      Frankly, I am surprised at a win probability below 70% at this point, I think the Princeton model at 90% is much closer to the real picture at this point.   I suppose Nate Silver's model does not have the means to include early voting info we have seen out of Ohio and Nevada, which make a Romney win in either state next to impossible, if the numbers we have seen are even close to accurate.  

  •  "I don't trust Nate because Corporations" (7+ / 0-)

    I know that's not what the diarist was saying, but there's been a lot of that mindless guilt by association crap going on around here in DKos, and it's really unfortunate - "Nate joined NYT, so Nate is now corporate."

    Give me a break.  There is no evidence to suggest that Nate is doing anything other than trying to build the most accurate forecasting model he knows how.  You might have issues with his methods.  But to go after someone's intellectual honesty for no good reason other than you don't like his results is, ironically, pretty damn intellectually dishonest.

  •  I don't have a problem... (2+ / 0-)
    Recommended by:
    itsbenj, ladybug53

    ... with what Nate does.  We were the ones going bonkers when he had Obama up to 87% before the first debate.  I suspect none of us would argue that things have gotten closer since then, and there is more of a "risk" now than there was before the debate.  68% isn't a horrid place to be.  As he indicated yesterday, it's a very narrow amount that's holding Obama to where he is:

    If Mr. Obama’s head-to-head polling were 2 percentage points higher right now, he would be a considerably clearer favorite in the forecast, about 85 percent. A 1-point bounce would bring him to 80 percent, and even a half-point bounce would advance his position to being a 75 percent favorite in the forecast.
    If we see a 0.5-1.0 drift towards Obama, I suspect we'll all feel a lot more comfy.

    In turn, if those undecideds break the wrong way in some of the later polling, I'd be worried.  Seriously... if you've seen some of the undecided talk about what they're rolling over in their head, you'd realize that Nate's 2/1 Obama favorite isn't unreasonable in a still close race where the Margin of Stupid is pretty thin.

    Also, in some of the states like NC, there is evidence of either good GOP voter enthusiasm or good GOP GOTV.  While they may not be winning the "raw numbers", relative to 2008 they're actually a bit more impressive.

    Cautious optimism isn't a bad thing until we see more and better polling.

    •  3 things (0+ / 0-)

      Ohio, Nevada, and Wisconsin.

      Ohio and Nevada have incredible early voting numbers for Obama.  Both states are almost out of the picture because of that, can't be credibly considered toss-ups anymore, but very likey Obama's at this point.   Plus, in Wisconsin Obama polls at or above 50% almost every time a poll is conducted in that state, which means that late-deciders don't mean a thing there if that continues.  

  •  It's not really the math (0+ / 0-)

    Every model is based on likely voter assumptions. The more the GOP moves right, the more we get headlines about voter suppression (Like Jacksonville FL and Indiana today) we get people angry.

    Nate's numbers can't take into account that anger!

  •  better to take Nate's word for it (1+ / 0-)
    Recommended by:
    Williston Barrett

    and work our butts off and win, then not and we don't

    Those who make peaceful revolution impossible will make violent revolution inevitable. ~JFK

    by TheUrbanRevolution on Wed Oct 24, 2012 at 07:42:15 PM PDT

  •  Do you ever wonder? (0+ / 0-)

    What if there was some habituation going on?  

    Meaning, Nate was far more accurate when he was someone who didn't get much press.  

    But now - since every pollster knows who he is - they will consider how they are judged relative to Nate's model and want to get a better weighting.  You could get real tin-foil hat and believe that corrupt pollsters will want to game his system to shape the narrative.

  •  538 (0+ / 0-)

    Depends upon poll aggregation and some other "fundamentals" calculations.
    If Ohio begins polling Romney in the lead the topline number goes to ~ %50% Obama.

    It is all about Ohio at this point.
    There is but one poll.


  •  Nate Silver had a good model in 2008, (0+ / 0-)

    so he had extraordinary performance in 2008.

    However, the dynamics this year are very different than 2008.  In 2008 we had an unpopular Republican President, who could not run for re-elevtion, with the economy going through a extraordinary financial crisis.

    This year we have a well liked Democrat as an incumbent, with high unemployment and unusually slow economic growth.

    It is not clear that the same model will work under today's very different environment.

    The most important way to protect the environment is not to have more than one child.

    by nextstep on Wed Oct 24, 2012 at 09:09:00 PM PDT

  •  I don't think you are deluding yourself too much (0+ / 0-)

    MSM is slowly waking up from their first debate slumber as well calling out others within for getting duped by Romney camp for their momentum meme.

    Beck wasn't only one discordant after 3rd debate. O'Reilly was too. There are only 2 weeks left and Romney is too scared to take network interviews. As long as his surrogates like Murdock keep the gaffe machine going, then  the momentum meter is wil eventually fall. The DNC Romney/Murdock is as damning as the Romney America ad that ran early on in the game. It is only a question of when, not if, the President gets re-elected.

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site