Princeton Election Consortium

Innovations in democracy since 2004

Outcome: Biden 306 EV (D+1.2% from toss-up), Senate 50 D (D+1.0%)
Nov 3 polls: Biden 342 EV (D+5.3%), Senate 50-55 D (D+3.9%), House control D+4.6%
Moneyball states: President AZ NE-2 NV, Senate MT ME AK, Legislatures KS TX NC

Happy New Year!

December 31st, 2015, 6:22pm by Sam Wang

Happy new year, all!

Predictions for 2016, anyone? Ideally with some logical support, but oh heck, it’s New Year’s Eve – let your freak flag fly. Aspirations?

Tags: Uncategorized

30 Comments so far ↓

  • mtsw

    A team with <90 wins will beat the 100+ win Cubs in the first round of the MLB playoffs. Lock it down.

  • 538 Refugee

    President Sanders sounds good but my “official” pick is Clinton. Long coat tails would be nice and it’s possible Trump could actually provide that.

  • Amitabh Lath

    Regarding Trump’s chances in the GOP primary: . It’s like judging a Gorn beauty contest. You can’t very well say the front runner has no chance because she looks like a lizard, that just shows your lack of understanding of the audience. In this case the thing you might think as a disqualifying negative (lizard features) is actually what is propelling her to the top.

    A lot of analysts including Nate Silver are setting aside the poll numbers and saying Trump has no chance, because well, Trump. Since I do not know any of his supporters I am going to assume that Trump being Trump is not a negative with them.

    Going with the numbers: Trump in the top 2 in IA, winner in NH, SC, FL, and most of super Tuesday.

    • Josh

      That actually isn’t what Nate is saying at all. At last check, he was pegging Trump at about a 15-20% chance to snag the nomination. Given that there are a dozen candidates in the race, 20% a month before the primaries is about as good as anyone could ask for.

      As I’ve written earlier, Trump’s problem is that he has a high floor and a low ceiling. There are many, many Americans who would never vote for him–some even in his own party. He’s not religious, so the religious Republican voters will not cotton to him like they do with Cruz or Rubio, and moderate Republicans, such as they are, are turned off by his extremist views on foreign policy.

      Really, Trump is tapping into a vein of impassioned, disenfranchised voters, who feel as though they’ve been left out of the conversation: uneducated whites, largely men, blue-collar or unemployed. It’s the one demographic of American voters who’ve clearly been doing worse by most any metric over the last 30 years, and to them, they want someone–anyone–who cares what they have to say.

      That’s what Trump does, putatively, anyway–he speaks to and for this group, which, depending on the math you use, makes up anywhere from 30-45% of the GOP electorate. Of course, this group is also hard to get to the polls–which is why some of us are somewhat more bearish on Trump’s chances. But I don’t disagree that if he can mobilize this slice of the voting pie, he has as good a chance as anyone of winning in the nomination.

  • bks

    More articles will mysteriously disappear from PEC.

    • Amitabh Lath

      Articles do not disappear, they get their timestamps changed. The last one for instance is now somewhere in December 2014.

    • Sam Wang

      Come on, give me a break. There’s a problem. I’m fixing it. I am taking advantage of the low season, when only the hardcore are paying attention. You, for instance! (Amit is correct, sometimes I redate things back by exactly by one year so that they do not disappear entirely. If you ever notice such an oddity, feel free to check there. Consider it an Easter egg.)

      If you must know, I am trying to get a clean read on the actually-getting-rather-significant likelihood of a Trump nomination. If one only used primary-season polls as evidence, he is in Gore 2000 / Bush 2000 territory. If you recall, those two sailed to their nominations.

    • bks

      Okay, real prediction: On 31 May 2016 Trump will have more delegates than Cruz, and Cruz will have more delegates than Rubio. Democrats retain POTUS. Republicans lose Senate but retain House..

    • Amitabh Lath

      Sam, I have been hoping for a serious analysis from you of the likelihood of Trump being the Republican nominee. None of the analyses I have read that try to explain (or explain away) Trump’s support have any sort of intellectual depth.

      They tend to be “my gut tells me he won’t be the nominee” pieces, or “I assembled a focus group of Trump supporters, and they support Trump!” Or they do some half-assed claim that polls can be wrong with a vague reference to Giuliani 08 with no numerical credibility to back it up.

  • Paul

    Golf: Tiger Woods plays very little competitive golf and does not come close to winning.

    NFL: Arizona Cardinals win Super Bowl.

    College Football: Clemson defeats Alabama.

    GOP allocation of delegates: I have tried to find a definitive site which explains allocation of delegates, state by state, without any success whatsoever. My prediction is that no media source will adequately or accurately explain this process.

    GOP Iowa: (1) Cruz (2) Trump (3) Rubio. Fiorina, Huckabee, Santorum, and Paul quit race by the end of Iowa caucus.

    GOP NH: (1) Trump (2) Rubio (3) Cruz. All other candidates quit race.

    GOP nominee: Cruz

    DEM Iowa: Clinton

    DEM NH: Sanders

    DEM nominee: Clinton

    President: Clinton

    House: GOP

    Senate: DEM

    • Sam Wang

      Amit – Generally, I agree with this assessment. Like you, I am bothered by the fact that even people who are data-based are really stuck on the weirdness of the Trump phenomenon. At some level, models based on existing measures (polls, campaign finance, party endorsements) are fine…until they are not. Those are types of data that really became available in abundance starting in 2004 — a short baseline to build models from.

      The thought bubbling in my mind is to find a way to sever all those biases from an analysis. It crept into my own thinking in that old post because I left out Gore (2000) and Bush (2000). One way is to find a way to collect all post-2000 data. That is pretty simple.

      Another way is to use delegate selection rules. Paul, you can find the rules at The Green Papers. Writing code to assign delegates in any simulation is a chore. There is nothing pretty about it.

      In regard to who drops out, why wouldn’t Bush stay in until Florida? See this set of guesses by James Pindell.

  • Amitabh Lath

    Either Run 2 of the LHC finds evidence of new physics like supersymmetry or a dark matter candidate, or there is an influx of young physicists into Wall Street.

    The last time this happened (early 90’s when the SSC was cancelled) they proliferated strange financial instruments like derivatives. What would it be this time? Lambda calculus? Holographic analysis techniques?

  • Paul

    Problem with the Green Papers is the disclaimer: “These pages contain a combination of official, unofficial, and estimated data. The information posted here is subject to change. This is a work-in-progress.” The delegate allocation process is arcane, may or may not bear any relationship to state totals, and differs state to state. In a year when total number of delegates may very well mean something, national and even state polling may not tell the story.

    As to why would Bush drop out (I figure a 5th in Iowa and 6th in NH), my answer is that the establishment will want him out as he will compete with Rubio, especially in Florida.

    • Amitabh Lath

      I read speculations that the RNC might fiddle with the nomination rules to keep Trump from clinching. This can happen anytime, even at the start of the convention.

      I do not think any faction controls the Republican party to that extent. If Trump is seen to be bringing disaffected white voters (of the sort that Sean Trende identified when he rebutted the GOP postmortem about hispanic outreach) then any agreement to box Trump out will be difficult to achieve.

    • Sam Wang

      I am also uncertain about this site’s sourcing. However, where documented the process is plainly labeled with current dates. So I think that your concern, as stated, does not make sense: of course the delegate process is arcane – that is the reason for this site’s existence. Or are you simply claiming that the rules are not described correctly?

      Article citing The Green Papers: in Salon/Alternet. Also, we can check rules against this.

      Your concern about polling is credible, but let me just say that there are ways to understand this process that do not require those polls to be accurate, at least not in the way that you are thinking.

  • A New Jersey Farmer

    Happy New Year to you all. I’m looking forward to another political year.

    I will give Trump the benefit of believing that he is indeed leading the polls and would win Iowa and NH if…wait for it…the election was held today. Seeing as we’re not going to see the first votes for a month, I will say that his support will slip and it’s more likely that Cruz, Rubio, Christie (NH) or even Bush will finish above him, though I do see Trump in the top three in both Iowa and NH.

    Cruz, Trump and Rubio will be the most competitive through Super Tuesday and I think Rubio will be the eventual nominee.

    Clinton will win the Democratic nomination and defeat any Republican in the fall.

    GOP keeps the House through 2022.
    Senate goes Democratic if Trump or Cruz is the nominee. Tossup if it’s Rubio.

    Christie comes back to NJ and resigns as Governor.

  • Fel Martins

    I’m not a Trump fan (far from it) but I think he’s underrated. He’s been playing a different game. If everyone is playing Chess he’s playing 3-D Chess. He’s said and done things that would have doomed any other campaign.

    For anyone who doubts me, let me just say this: remember when he first jumped in? People said he was unelectable, he had high negatives and was far underwater when you compared approval/disapproval. Keep in mind that, unlike some other candidates, Trump is a known entity, changing people’s perception of you is extremely difficult once set in (look at Hillary, who has just as high negatives now as she did 8 years ago). But Trump has managed to convert people to his side. What’s so say he won’t convert more between now and voting time?

    Trump is a big gamble. If he’s nominated, the GOP is either going to lose big (landslide Clinton which is what I hope for, along with Dems taking the senate) or they win big similar to Reagan in 1980. Keep in mind that, up until those last moments, Reagan and Carter were pretty equal in polls.

    • Josh

      I respectfully disagree. Trump has, as they say, a high floor and a low ceiling. His rabid following can’t grow much beyond a certain point because a large portion of the electorate simply will not vote for him. Reagan was much close to the middle, ideologically speaking, and had an easier time capturing voters who weren’t especially psyched about another four years of Jimmy Carter.

    • Matt McIrvin

      1980 was a three-way race, in which John Anderson came in in the spring and siphoned off a bunch of voters basically exclusively from Carter, and then in the late fall started losing some of his supporters to Reagan. That suggests that Carter’s support was soft in the first place, though, as approval polling would suggest it was.

  • Paul

    Thanks to Sam and others for commenting.

    Last comment for now on delegate allocation. My point about the “The Green Papers” was not questioning the veracity of the information but rather that the process is complex and still evolving, and that the caveat on the page reinforces that idea clearly. You really have to understand the rules of each state separately, and that neither national polls nor state wide polls are necessarily geared to make predictions about the delegate allocation. Further, the GOP process appears far more complex than the process used by the Democratic party.

  • Amitabh Lath

    I looked at the Republican delegate selection rules and my eyes glazed over. I suspect the rules are allowed to be this arcane because they are never ever supposed to matter. By the time of the convention, the winner must be obvious to all.

    Because if delegate assignment rules become the story at the convention, then the party has lost the narrative. No matter how it turns out, half the party will go home thinking they were robbed; depressed turnout, bad news all around. Only winners will be the hyenas of the press.

    So look at the dates of the primary. April is loaded with large states which have winner-take-all (or winner-take-most) primaries. Starting with WI, then NY, PA, IN, WA, CA, NJ, positioned to give an insurmountable advantage to the front runner and ensure it does not come down to a coin toss. That seems to be the plan, anyway.

  • 538 Refugee

    Does this put Bernie within one sigma of Hillary? ;)

    Bernie Sanders raised more than $33 million during the last three months of 2015 to underwrite his bid for the Democratic presidential nomination, his campaign announced Saturday. His haul puts him within $4 million of the $37 million raised by Democratic front-runner Hillary Clinton during the same period.

    In all, the Vermont senator has raised nearly $73 million for his White House battle, campaign officials said. Aides said more than 1 million donors made more than 2.5 million contributions to Sanders’ campaign, surpassing President Obama’s previous record for individual contributions. (Obama had collected 2.2 million contributions at this stage in his re-election campaign.)

  • Matt McIrvin

    My gut says that what the polls say now may not happen, but it’s the way to bet, no matter how weird it may seem. So Trump gets the Republican nomination, Clinton gets the Democratic nomination, Clinton wins the election fairly handily, though not in a 40+-state landslide.

    I am actually pretty certain about Clinton’s nomination; I like Bernie Sanders but he’s not getting nominated, no way.

    I am less certain about Trump, but I do think that he and Ted Cruz have the best chances. It’ll be much clearer what’s going on after Super Tuesday.

    I think that the chances of Hillary Clinton actually winning the general election are considerably higher than most pundits and prognosticators are willing to say. If pressed they’ll usually say it’s 55, 60 percent in her favor, which is almost the same thing as saying nothing. But usually, if you ask them you find out this is predicated on someone like Jeb Bush or Marco Rubio getting the R nomination, which they still think is more likely than not.

    • Matt McIrvin

      Just to be clear, the reason Bernie Sanders will not be the nominee is not that there’s a giant conspiracy to fix the nomination contest; it’s that not enough people will vote for him.

    • Amitabh Lath

      Hi Matt, happy new (presidential election!) year. I am a little (a lot) fuzzy on this concept of “high floor/low ceiling” that you ascribe to Trump in a post below. What are these and how are they determined? What are they for the other candidates?

    • Matt McIrvin

      That wasn’t me, it was Josh.

    • Matt McIrvin

      I think the thing that distinguishes Trump and Hillary Clinton from all the other candidates is that they’re very, very well-known and people already have well-formed ideas about them. They’re the only two whose favorable plus unfavorable ratings are over 90 percent.

      Which is not to say those opinions are immutable. Hillary Clinton had huge favorables a few years ago but dropped into net negative territory as a result of becoming a presidential candidate again. Trump seems to have become more positively regarded when it became clear he was a serious candidate, but now his favorables are creeping down again, maybe because his extreme statements are polarizing his non-fans against him.

    • Matt McIrvin

      (…And, I guess I should make clear, Hillary Clinton’s favorable/unfavorable is still several points better than Trump’s.)

Leave a Comment