Princeton Election Consortium

A first draft of electoral history. Since 2004

Trump 1st+2nd choice reaches 50%; Carson least upsetting

September 10th, 2015, 9:07am by Sam Wang


This is interesting, from CNN/ORC: those who list Trump as their first choice added to those who list him as their second choice now add up to 50%. In principle, this opens the possibility of a clean win if the whole narrowing-down process were to play out immediately, using today’s opinion among GOP voters.

However, an actual Trump win by next year’s convention is still highly unlikely because of opposition by Republican Party leaders – and primary voters too. Of GOP voters surveyed, 21% said they would be “upset” if he were the nominee, not far from the number who would be upset about Jeb Bush.

Who was least upsetting? Ben Carson at 5%, followed by Marco Rubio at 7%. If the primary process were allowed to play out in an orderly manner (a big if, with Trump acting as a wrecker, to use the old Soviet term), these two would seem to be in a better position to make it to the final stages of the primary season.

Tags: 2016 Election

28 Comments so far ↓

  • bks

    Last pre-debate poll (YouGov/Economist) — Trump 33, Carson 17, Cruz 8, Bush 7, Rubio 7, Fiorina 6, Walker 5 …
    https://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/4ranljpkz8/econTabReport_v1.pdf
    –bks

  • pechmerle

    Dept. of Unstable Numbers.
    Ben Carson is now up to 23%, just 4% behind Trump.

    http://www.cbsnews.com/news/new-poll-donald-trump-still-leads-but-ben-carson-makes-strong-gains/

  • A New Jersey Farmer

    The latest NYT poll is a real corker. It shows Trump still on top with Carson next. The Times, true to form, then says that Hillary’s lead over Sanders has gone from 41 points to 20, which is important, why? I told my students not to pay attention until at least Halloween. Debates will be fun to watch.

    http://www.nytimes.com/2015/09/16/us/politics/gop-support-for-donald-trump-rising-as-ben-carson-gains-poll-finds.html?hp&action=click&pgtype=Homepage&module=first-column-region&region=top-news&WT.nav=top-news

  • Amitabh Lath

    I hear Pataki will not sign the Republican primary loyalty oath. Not that Pataki ever mattered, but the loyalty oath might lead to awkward questions for someone like Jeb Bush: If he is so pro-immigrant and compassionate how can he sign on to (even remotely) support a guy who says he will deport 10^7 people? Does he put party over deeply held principles?

  • Kim Golderg

    Here is another interesting poll:

    http://talkingpointsmemo.com/polltracker/joe-biden-hillary-clinton-msnbc-telemundo-marist-early-results

    might not make it on RCP as it doesnt support media driven narrative.

  • Josh

    Thanks for the response Sam. Of course you are correct.

    I should have been more clear: I do not believe that the moderate (but statistically significant) drop in what little support Carly Fiorina had to this point will have any real influence on the race in general. If we assume the 2% of support lost is divided proportionally among the remaining candidates based on the current polling, it changes nothing; but even if we assume one candidate (Jeb!? Scott Walker? Ben Carson?) receives the lion’s share of that 2%, I’d argue that it still fundamentally changes very little if anything about how we see the race at this point.

    I don’t think I’m wrong about this, but I’d be grateful to be shown otherwise.

    • Sam Wang

      Could be her rise and fall doesn’t mean anything…I agree that it’s likely of little significance. I forget who said it where, but she seems to be an example of a brief boost arising from media exposure. It’s what John Sides was implying when he claimed that Trump was purely a media exposure-driven phenomenon.

      It is worth thinking about what went wrong with Sides’s assessment. I think he missed the fact that media coverage is not administered at random. It is just one step in a feedback loop: if a candidate says something exciting or motivating, that attracts media coverage; if he/she repeats the performance, then that gets spread around, attracting more (or less) media coverage. It’s a dynamic process.

      In the case of Fiorina, the question is whether she can come up with anything else to attract more coverage. In the interval since the first debate, she hasn’t really sizzled as far as I can tell.

      At some level, this is why Carson is interesting at the moment: he’s had a fairly fanatical following from the start, in the face of basically no media coverage. Yet here he is in the top tier of candidates. His level of support suggests that he is doing something interesting. But then what will media coverage get him? I think he’ll make it fairly far, then maybe implode in the home stretch. Just a guess.

  • Amitabh Lath

    CNN general election poll has Trump tying Clinton but losing to Biden.

    This brings up some red flags.

    Biden may be considering entering the race, and this sort of poll may tip the balance.

    And Biden entering the race would be very good for CNN ratings.

    Hmmm….

    • 538 Refugee

      Biden may get in just as a backup in Clinton falters. I’m interested about the timing. Are there stats that show when the best time to try and torpedo the front runner is? I’d have thought that playing the email card would have been better much closer to the election.

  • bks

    Who is more likely to to go the polls in Texas, a Bush supporter or a Trump supporter? If the GOP nomination is not decided on 6 June, California and New Jersey have winner-take-all GOP primaries on 7 June 2016. If Trump is one of the players, doesn’t he have the edge over all other candidates in California (the single largest block of votes)?

    Romney was #1 in New Hampshire wire-to wire. There was not a single poll that showed a different candidate winning in NH. None of the 16 GOP dwarfs will be able to make this claim.

    Just blue-skying. –bks

  • Lee

    When can we expect there to be state-wide polling useful for predicting winners in the Democratic and Republican primaries? These national polls make nice headlines but barely reflect the realities of the election process.

  • Rob

    I think everyone’s going to need to reset their expectations based on this info. Trump the nominee is now a very real possibility

  • Amitabh Lath

    One more argument in the Trump-numbers-seem- unreal basket: the general election matchups where Trump is near or passing Clinton. Really? This may be pointing towards some catastrophic problem with polling filters, because Trump beating Clinton is equivalent to the faster than light neutrino (ie, results so ridiculous that they point you to the bug).

    • bks

      Polling can stay irrational longer than you can stay sane. –bks

    • Sam Wang

      Along those lines, Survey USA crosstabs indicate 25% support for Trump among “the blacks” as he would call them, and 30% among “the Hispanics.” I am curious to see how those settle down once the general election campaign begins between Trump-Carson and Clinton-Dayton (or whoever).

    • Matt McIrvin

      Is it ridiculous? Trump’s got all the energy and media fascination in the campaign at the moment. Clinton is basically still in the fundraising stage. And even if it won’t be anything like that come Election Day, right now, a lot of people are basically not paying a whole lot of attention to the race.

    • Rieux

      Clinton-DAYTON, Sam?

      I’m a Minnesotan who has voted for Mark Dayton every chance I’ve had (I also did some street-level volunteering for his gubernatorial campaign), and even I think that Dayton would make a terrible VP candidate. He’s old, he’s (ahem) low-energy, he’s melancholic as hell, and he talks like he’s got a mouth half-full of marbles.

      When it comes to actually governing, actually doing his job, Dayton’s been a very good governor. (I’d vote for him again, though I believe he’s said he’s not going to run for a third term in 2018. I doubt he’d even accept a VP nod.)

      As a campaigner, however, Dayton stinks. Clinton (or whoever) should certainly pick someone else.

      On the other hand, extolling Minnesota’s performance under Dayton, especially as a helpful contrast to Wisconsin’s contemporaneous faceplant under a certain collapsing Republican presidential candidate, would work just fine.

    • Sam Wang

      Heh. I just pulled that off a list of governors. Sorry, I was being too random.

      Other options include Jerry Brown (successful but old), Hickenlooper (loopy but then again, Colorado is valuable), Malloy (survived in CT in a GOP wave year), Beshear (Kentucky), Inslee (Washington). I dunno, Beshear?

    • Matt McIrvin

      So far these head-to-head questions with Clinton vs. anyone other than the previously assumed major candidates (Bush, Walker, Rubio, etc.) are somewhat underpolled, and subject to small-number effects and outsize influence from loopy organizations like Gravis Marketing.

      It’s interesting to see that MSNBC-Telemundo-Maris poll, because the headlines are all touting it as a reason for Biden to run, but they’re burying the lede that it doesn’t show Trump beating Clinton. But enough recent polls had shown that that it was starting to maybe look like a solid result.

    • Craigo

      If Clinton beats Sanders and chooses Beshear, the “democratic wing of the Democratic party” would take it as a slap in the face.

      Hickenlooper is a decent option. O’Malley fits the runner-up who’s not too much of a runner-up mold. Kaine, Vilsack, and Warner are capable moderates from a swing state. Castro and Heinrich are fresh blood. She has good options.

  • bks

    Fiorina is at 3% down from 5% in the same poll three weeks ago. Even Nate Silver is still talking about the Fiorina surge even though she’s down in the noise. –bks

    • Josh

      A 2% difference over 3 weeks in one poll 14 months before an election is meaningless.

    • Sam Wang

      Josh‘s statement is statistically not quite right, and deserves a response.

      For very small fractions, the uncertainties are smaller than for typical 2-way horserace results where support is in the 30-70% range. The formula for one-sigma is sqrt(p*(1-p)/N) where there are N respondents. For example, N=1000 respondents and 3% support, the 1-sigma error bar is 0.5% and the margin of error is 1.0%. In the case of a change from 5% to 3%, the difference is outside the margin of error. To the extent that we are interested in ups and downs at all, Carly Fiorina’s decrease is likely to be real.