Month: May 2016

Who will win the 2016 US presidential election? (update)

Surprisingly, party approval ratings don’t seem to have much predictive value for presidential races[1]. But what about the candidates’ own favorability ratings? There’s good news and bad news here for the enterprising forecaster.

The good news: I have it on excellent authority that “the better-liked candidate has won each election since 1992, when Gallup began measuring favorable ratings in the current format”[2]. That’s only 6 elections, so look out for the small data set. Nevertheless, net favorability ratings seem like a pretty good predictive tool in presidential elections.

The bad news is harder to explain. Mr. Trump’s net favorability rating is currently -20%, and it has been in this neighborhood since August 2015[3]. HRC’s net favorability rating is currently -15%. But it has been positive as recently as 13 months ago and has been falling steadily about 1 percentage point per month since early 2013. If that trend continues, and Mr. Trump’s net favorability remains constant, then Mr. Trump will have a slightly higher net favorability rating than HRC does at the time of the election.

That’s a pretty big “if.” What does the future really hold? The following factors could affect HRC’s net favorability ratings over the next few months:

  • Reconciliation with Bernie or further alienation from Bernie[5]
  • The email scandal heats up even more[6]
  • Bill does something crazy[7]
  • The business about HRC being an enabler for Bill’s philandering catches on
  • Old FoB and FoH, Terry McAuliffe’s legal problems get worse[8]
  • Something (else) from HRC’s time as SoS turns bad (a la Lybia or Syria)
  • A scandal (or, if you prefer, “scandal”) from the past (Whitewater, Vince Foster, Benghazi, the hits just don’t stop) reemerges

That’s a lot that can go wrong, and not much that can go right. That said, HRC will probably get a boost coming out of the convention, provided that she’s able to unify (more-or-less) her party. I’m sure I’m missing some stuff, but it looks a bit bleak on donkey-back.

In contrast, the following factors could affect Mr. Trump’s net favorability ratings over the next few months:

  • Tax returns: either not releasing them makes Mr. Trump look worse (somehow) or he does release them and they contain something that is unforgivable to those who were fine with the racism, sexism, ablism, and so on[9]
  • Mr. Trump manages to say something that goes beyond the pale with his supporters (and, most likely, about his supporters)
  • Mr. Trump manages to pull apart the Republican Party, by – let’s say – attacking popular members of the party who are more-or-less bystanders to the election[10]
  • A scandal (no scare quotes this time) from the past (beauty pageant grossness, Trump University, mafia ties, etc.) reemerges[11]
  • Further evidence that Mr. Trump doesn’t know much about presidenting[12]
  • An exogenous shock to the US economy (which is almost always a boon to the party that is out of power) occurs
  • A large-scale attack on US civilians or military personnel (which might help the guy who said he’d kill the families of terrorists[13] or might hurt the guy who probably still doesn’t know what the nuclear triad is[14]) occurs
  • Retired high-ranking US military officers have a public freakout about Mr. Trump[15]

As with HRC, it’s a little hard to think of things that will improve Mr. Trump’s approval ratings. However, as I said before, they have gone up, so there you go. I suspect this is a matter of Republican voters lining up behind their candidate, and that will only go on for so long. Mr. Trump’s gross (as opposed to net) favorability rating is 38%. At the moment, 25% of likely voters consider themselves to be Republicans, 41% if you include those leaning toward Republicans.

I have to admit that it’s damn hard to know what sort of probabilities to assign these events. It would have to be a team effort. Or just – I don’t know – take a guess or something. Yeah, that’s what I’ll probably do.

Note that both of these candidates are pretty damn unpopular, by recent historical standards, with the voters. President Obama consistently had a net favorability rating of 20-or-more percentage points during his run-up to winning the White House in 2008, and his net favorability numbers didn’t fall below zero until 2013, until to oscillate from about +3 to about -3 percentage points since then[16]. In fact, the only major political figures other than President Obama who have a positive net approval are Sen. Sanders, who’s up by about 10% [17] and Joe Biden, who’s up even more [18].

Of course, one must put these numbers in context, especially those concerning Biden. I might approve of both Chris, the local police officer, and Pat, the local high school English teacher because each does her job well. But would I continue to approve of Chris if he started teaching English too? I don’t know, but my approval of her work as a police officer is little guide here. Chris might be very good at giving out speeding tickets to out-of-towners but abysmal at getting teenagers excited about John Donne and Gerard Manley Hopkins. Likewise, I might approve of Mr. Biden in his role as Vice-President, but I might also turn on a dime if he were at the top of the national ticket, even though Mr. Biden has all the same virtues and vices. Something similar (though not identical) is true of Sen. Sanders. I might approve of him as the head of an insurgent movement within the Democratic Party but disapprove of him as someone who’s one election short of being the Leader of the Free World (as we used to say back in the old red-under-the-bed days).

tl;dr – HRC’s net favorability ratings will probably go up a bit after Bernie is ushered from the stage, while Mr. Trump’s seem unlikely to get much better. And there’s good reason to think that the candidate with the higher net favorability will win. Since HRC already has a lead on Mr. Trump, that’s further reason to think she’ll beat him in November. But all of this is strictly ceteris paribus. So many highly unpredictable factors (see above, ad nauseum) can cause these numbers to fluctuate that it would be recklessly imprudent, I believe, for forecasters to put much weight on them, at least at this stage of the game. And that’s the bad news.

[1] https://www.gjopen.com/comments/comments/243836

[2] http://www.gallup.com/poll/157292/obama-challenge-higher-likability-approval.aspx

[3] http://elections.huffingtonpost.com/pollster/donald-trump-favorable-rating

[4] http://elections.huffingtonpost.com/pollster/hillary-clinton-favorable-rating

[5] https://www.washingtonpost.com/blogs/plum-line/wp/2016/05/24/can-bernie-sanders-actually-hurt-hillary-clinton-on-his-way-out/

[6] http://www.nbcnews.com/news/us-news/what-state-department-email-report-means-hillary-clinton-n580516

[7] http://www.courierpress.com/columnists/bills-tough-on-crime-approach-could-hurt-hillarys-chances-31000e4d-20f4-03e1-e053-0100007fd89b-376553881.html

[8] http://www.cnn.com/2016/05/24/politics/terry-mcauliffe-hillary-clinton-2016/

[9] http://www.nationalreview.com/article/432326/donald-trump-tax-returns-must-be-released

[10] https://www.washingtonpost.com/blogs/post-partisan/wp/2016/05/26/trumps-new-mexico-setback/

[11] http://www.theatlantic.com/politics/archive/2016/04/donald-trump-scandals/474726/

[12] http://foreignpolicy.com/2016/04/04/trump-is-the-ignorant-candidate-ignorant-americans-deserve/

[13] http://www.cnn.com/2015/12/02/politics/donald-trump-terrorists-families/

[14] http://www.politico.com/blogs/live-from-the-venetian/2015/12/2015-gop-debate-rubio-trump-nuclear-triad-216828

[15] http://www.thedailybeast.com/articles/2015/07/20/retired-admirals-torpedo-president-trump.html

[16] http://elections.huffingtonpost.com/pollster/obama-favorable-rating

[17] http://elections.huffingtonpost.com/pollster/bernie-sanders-favorable-rating

[18] http://elections.huffingtonpost.com/pollster/joe-biden-favorable-rating

Advertisements

Who will win the 2016 US presidential election? (update)

The Republican Party is very, very unpopular and has been so for quite some time. It has a net favorable rating of -29%[1] and has not had a net positive favorable rating since 2005[2]. Currently, we’re currently looking at a 20-year low in the party’s popularity[2].

Republican_Party_Favorability___Pew_Research_Center

Ouch. In contrast, the Democratic Party has a net favorable rating of -1.3%[3]. That’s nothing to write home about, but it’s 22 times better than the Rs.

Democratic_Party_Favorability___Pew_Research_Center

Will any of this make a difference come November? Less than you might expect. Here are a few data points, going back to 1992 (the first year for which I could find information). I note at the outset that we need to be careful about overgeneralizing, of course.

  • 1992: Rs = -3 and Ds = +28, Ds win
  • 1996: Rs = +8 and Ds = +2, Ds win
  • 2000: Rs = +10 and Ds = +22, Rs win
  • 2004: Rs = +11 and Ds = +19, Rs win
  • 2008: Rs = -10 and Ds = +24, Ds win
  • 2012: Rs = -8 and Ds = +13, Ds win[5]

Begin by noting that there’s not much correlation between the party’s approval rating and the victory of the party’s candidate. The candidate from the party with the higher approval rating won in 1992, 2008, and 2012, and the candidate from the party with the lower approval rating won in 1996, 2000, and 2004. The winning candidate’s party has an approval rating of an average of no more than a 4 percentage points higher than its rival.

That said, a few points stand out. No candidate from a party with a net negative favorability rating has won the White House since Ronald Reagan was in office. Now, it’s possible to make too much of this because *both* parties regularly had positive favorable ratings until pretty recently. Of course, if neither party improves its favorability ratings in the next 5 months, that streak is going to come to an end in 2016. Currently the approval ratings gap is 27 points. In fact, the last time the Republicans won the White House (i.e., 2004), their net favorability rating was 11%, a whopping 40 percentage points higher than their current status.

At the risk of excessive curve fitting, let’s give a further tweak to the data. I’ll add the incumbent’s approval rating to his party’s rating. Here’s the result:

  • 1992: Rs + GHWB = -30 and Ds = +28, Ds win
  • 1996: Rs = +8 and Ds + WJC= +26, Ds win
  • 2000: Rs = +10 and Ds + WJC = +52, Rs win
  • 2004: Rs + GWB = +18 and Ds = +19, Rs win
  • 2008: Rs+ GWB = -52 and Ds = +24, Ds win
  • 2012: Rs = -8 and Ds + BHO = +19, Ds win[6]

That’s a little closer to the historical facts of the last few decades. On average,the winning candidate’s party has an approval rating of an average of a 22.67 percentage points higher than its rival. But the variance is huge: 42 percentage points!

So what does all of this mean for 2016? First, we’d be very unwise to put much stock in the fact that the Democratic Party and its President are much more approved of than the Republican Party. The sum of the net approval ratings for President Obama and the Democratic Party is about 3%, and that’s a lot higher than the Republican Party’s approval rating of -27%[7]. But, first, the Republicans were able to overcome a 42%(!) gap in 2000, coincidentally(?) the last time the Democrats had a sitting president who was not running for the highest office in the land. (Honesty, how the hell to Mr. Bush get elected in 2000? A better question is how the hell Mr. Gore did not. This was a Boer War level failure.)

Second, does the electorate even consider Mr. Trump to be an honest-to-goodness Republican[8]? The heavily tarnished image of the party might not rub off on him. Sure, Mr. Trump has monumental problems of his own[9]. But his approval ratings have been trending upward[9] as he becomes more closely associated with the party (hell, he’s about to become the head of the GOP, something that seems impossible even as I type these words), while the party’s numbers have been fairly flat. There might be a lesson here for those who breathlessly declaim that Mr. Trump will irredeemably ruin the GOP’s credibility[10]. There’s already some reason to believe that voters are compartmentalizing their views Mr. Trump and the Republican Party, which is human, all too human[11].

[1] http://elections.huffingtonpost.com/pollster/republican-party-favorable-rating

[2] http://www.pewresearch.org/data-trend/political-attitudes/republican-party-favorability/

[3] http://www.businessinsider.com/poll-republican-party-favorability-ratings-2016-4

[4] http://elections.huffingtonpost.com/pollster/democratic-party-favorable-rating

[5] Data taken from http://www.pewresearch.org/data-trend/political-attitudes/republican-party-favorability/ and http://www.pewresearch.org/data-trend/political-attitudes/democratic-party-favorability/ – Note that I took the favorability ratings from the time closest too (but not after) the given election; for the 1996 election I had to use 1995 data. Also who knows what to say about 1992 with the third-party craziness and all? Not me.

[6] http://www.gallup.com/poll/124922/Presidential-Job-Approval-Center.aspx – An even better way to do this would be to add the candidate’s party’s approval rating to the candidate’s own approval rating and then compare it to the sum of his/her own approval rating and his/her party’s approval rating. But I lack that information and am too lazy to try to get it.

[7] http://www.nbcnews.com/meet-the-press/obama-hits-highest-job-approval-second-inauguration-n578866

[8] http://www.usnews.com/news/blogs/ken-walshs-washington/2016/02/22/is-trump-really-a-republican

[9] http://elections.huffingtonpost.com/pollster/donald-trump-favorable-rating

[10] http://thehill.com/blogs/ballot-box/gop-primaries/264175-george-will-trump-nomination-would-destroy-gop

[11] http://www.simplypsychology.org/cognitive-dissonance.html

Who will win the 2016 US presidential election? (update)

On the one hand, there’s been quite a lot of public hand-wringing lately about the narrowing of the polling gap between HRC and Mr. Trump[1][2]. On the other hand, the GJ consensus has been been 2-to-1 in favor of HRC (or the D of your choice) for more than a month. What to think? Is the GJ consensus out of alignment with the changing reality of the 2016 campaign for president? Or is the world of media overreacting?

Well, the unweighted average of polls for May still has HRC up by 3 points. Of those 20 national polls this month, HRC has won 16, Mr. Trump has won 3, and there has been 1 tie. Yet if we stick with polls that were initiated since May 15 the results are a bit different. HRC wins by an average of about 1.6 points, and she actually wins 4, losses 2, and ties once[3]. If you just look at the those who say they will vote for HRC/Mr. Trump and ignore those who are either undecided or prefer another candidate, then there are 2 percentage points between the candidates with a standard deviation of slightly less than 2 points. That…makes HRC about a 2-to-1 favorite. Well, it would if those numbers had a lot of predictive value, and they probably don’t since we’re still too far away from the election, and some claim that May polls are especially unreliable[4]. At any rate, the betting market has similar numbers[5]. So, no, my limey friends, it’s “not time to panic” just yet, if you have some hopes for HRC, the survival of western civilization, or – what the hell? – both.

A few things have gotten missed in the rush to put on sackcloth and smear our faces with ashes: First, this isn’t the first time that HRC and Mr. Trump’s numbers have converged. They were even closer (HRC>Mr. Trump by 1.2%) on September 12 of last year. They were even closer than that on December 3, 2015. None of that stopped HRC from recovering to lead by more than 11% recently. So we might be seeing something cyclical, rather than some inexorable trend toward parity (or worse) for the Clinton campaign. So that takes the a bit of the edge off the most recent changes in the polls: small advantage HRC.

Poll

Second, the nice people at vox.com point out that if you look beyond the who-will-you-vote-for polling, you begin to see data that is much friendlier to HRC. For instance, “A majority of voters (59-33) think Clinton has a better personality and temperament to serve as president”[7]. I guess the idea is that, over time, voters will change their mind about whom they will vote for, so as to align themselves with their beliefs about job qualifications, etc. “But wait,” I can hear someone say, “why not think that voters will change their minds about job qualifications, etc., to align themselves with their beliefs about who they’re going to vote for?” Good question, someone. I don’t know the answer, but it’s worth noting that HRC comes out as favorable to Mr. Trump in no less than all 10 of the questions that the Vox piece mentions. If you’re in a Quinean mood[8], you might think that economy of belief revision favors changing one belief over changing many, though of course it matters rather a lot how close those beliefs are to the center of the web: very small advantage HRC.

Third, there’s an increasing likelihood that that some Republicans will see voting for the Libertarian Party’s candidate as a viable option[9]. Gary Johnson isn’t going to move into the White House next year (unless he marries either Mr. Trump or Mrs. Clinton). But Republicans who are justly appalled[10] by Mr. Trump’s demagogic lunacy yet don’t want to vote for a third Clinton term (or a third Obama term, slice and dice this one as you will) are likely to find some common ground on economic policy with the Libertarians. Might Democrats (or those who would otherwise vote for the Democratic Party candidate) also vote Libertarian? Some have thought so[11], but I’m blowed if I can see it. Sanders, in my opinion, is a socialist first and foremost. People of my extreme age sometimes use “socialism” as a term of abuse. Cold (War) habits die hard, I guess. But I mean this only as a descriptive term. If you’re a Democratish voter who’s attracted to Bernie because of his economic populism, you’re going to have a very rude awakening when you get a load of this little gem from the 2012 Libertarian Party platform:

A free and competitive market allocates resources in the most efficient manner. Each person has the right to offer goods and services to others on the free market. The only proper role of government in the economic realm is to protect property rights, adjudicate disputes, and provide a legal framework in which voluntary trade is protected. All efforts by government to redistribute wealth, or to control or manage trade, are improper in a free society.[12]

Now, no one is going to confuse HRC with Eugene Debs (especially since Mr. Debs died in 1926), and soaring passages in praise of HRC will not figure in Woody Guthrie’s songs (Since he died in the 60s; hell, I’m just dating myself badly), but I think Sen. Sanders’ supporters[13] can tell a hawk from a handsaw. Bernie bros and Bernie babes[14] are not going to be able to make peace with this unholy love child of Bob Nozick and Eugen Böhm von Bawerk. I conclude that Mr. Johnson and Mr. Trump are essentially playing a zero-sum game: small advantage HRC, which could become fairly large if Mr. Johnson catches on with the anti-Trump Republicans[15].

And that gets me back to more or less where I was when I started to rethink this question: A little more than a 2-to-1 advantage for HRC, though there’s a lot of room for change in the next few months. The media has (so far) been over-reacting to the latest polls.

[1]http://www.theguardian.com/us-news/2016/may/23/donald-trump-poll-hillary-clinton-presidential-election-2016?CMP=share_btn_tw

[2] http://www.nytimes.com/2016/05/25/us/politics/trump-electoral-map.html

[3] Data taken from http://elections.huffingtonpost.com/pollster/2016-general-election-trump-vs-clinton

[4] http://election.princeton.edu/2016/05/22/february-national-polls-are-the-best-you-get-until-august/#more-15718

[5] https://www.betfair.com/exchange/plus/#/politics/market/1.107373419

[6]http://www.realclearpolitics.com/epolls/2016/president/us/general_election_trump_vs_clinton-5491.html

[7] http://www.vox.com/2016/5/24/11747294/donald-trump-ahead-poll

[8] http://www.ditext.com/quine/quine.html

[9] http://fivethirtyeight.com/features/pay-attention-to-libertarian-gary-johnson-hes-pulling-10-vs-trump-and-clinton/ – Mary Matalin seemed like an early adopter (http://www.politico.com/story/2016/05/mary-matalin-registered-independent-222882) but has since tumbled down from the moral high ground because she, I guess, thinks it’s okay to have a presidential candidate who advocates torture (http://www.thenation.com/article/exactly-what-kind-of-torture-does-donald-trump-want-to-use/) and killing children (http://www.independent.co.uk/news/world/americas/donald-trump-reiterates-desire-to-murder-terrorists-families-a6912496.html), and who thinks that global warming is a “hoax” (https://www.washingtonpost.com/news/morning-mix/wp/2016/05/24/donald-trump-calls-global-warming-a-hoax-until-it-threatens-his-golf-course/) as long as he has”attitude” and “strategic chutzpah.” Thanks a lot, Mary. You’re an early contender for the Heinrich Brüning award.

[10] http://nymag.com/daily/intelligencer/2016/04/america-tyranny-donald-trump.html

[11] https://alibertarianfuture.com/2016-election/five-reasons-bernie-sanders-fans-should-support-gary-johnson/

[12] https://www.lp.org/files/LP%20Platform%202012.pdf

[13] http://www.thenation.com/article/bernie-sanders-will-make-the-economy-great-again/

[14] https://www.instagram.com/babesforbernie/?hl=en

[15] I’m assuming Mr. Johnson will be the Libertarian Party nominee, but there’s a small chance it’ll be someone else.

Will Montenegro become a NATO member in 2016? (update)

This question has come back to life as a result of NATO’s immanent signing of the accession agreement with Montenegro. But what does this mean for the question of whether Montenegro will become a member of NATO before 2017 begins? “Diplomatic sources say they expect it will take the 28 member states some 18 months to ratify the Montenegro accession accord, which Russia has condemned as another case of NATO encroaching on its strategic interests,” one media outlet tells us[1]. That would put Montenegro in NATO near the end of 2017, much later than Vesko Garčević, Montenegro Coordinator for NATO membership, has suggested[2]. So why 18 months? The outlet doesn’t say. Thanks a lot, dude. Other outlet who report this piece of news don’t mention the amount of time that they expect the ratification process to take[3]. So I’m inclined to think the reporter pulled the number out of his butt. By the way, Garčević’s Twitter feed is full of encouraging retweets from the likes of Denmark[4]. Really adorable stuff.

[1] http://www.defensenews.com/story/defense/international/europe/2016/05/18/nato-montenegro-sign-accession-accord/84533698/

[2] derstandard.at/2000034771967/Montenegro-Unterzeichnung-des-Protokolls-zum-Nato-Beitritt-im-Mai (translated from the original German by Google).

[3] One exception is the Daily Star, but it repeats the Defense News article almost word-for-word, which means one is cribbing off of the other or both are just taking dictation from a source who wants to stay off the record, probably the latter – https://www.dailystar.com.lb/News/World/2016/May-18/352743-nato-montenegro-to-sign-accession-accord-stoltenberg.ashx

[4] https://twitter.com/VeskoGarcevic

Will a majority of voters in Britain’s upcoming referendum elect to remain in the European Union? (update)

A number of folks here have drawn out attention to a recent article at the Guardian in which one is told about

a widening disparity between phone and Internet polling, with one producing a 10-point lead for remain and the second reporting that the leave campaign is ahead by four.[1]

What to make of this? One possibility is that both of these polls are outliers. An unweighted average of polls this month has both “stay” and “leave” at 43%, with “undecided” at 14.3%. That’s close to the April numbers: “stay” = 44.43%, “leave” = “41.71,” “undecided” = 12.38%[2]. The Guardian phone poll has “stay” at 47 and “leave” at 39 (which is an eight-point spread, not a ten-point spread as the quote above says, I must have missed something), and that’s about 1.5 standard deviations from the mean for “stay” and about the same for “leave.” The Internet poll was less out of line with the numbers we see elsewhere, with only “leave” more than a standard deviation away from the mean. I don’t know enough about the merits of phone vs. Internet polling to have an informed opinion, but my best guess is that we don’t yet have enough experience with Internet polling to speak to its accuracy in any reliable way[3]. Maybe I’m too pessimistic; who knows?

So that was something of a dead end, I think. But another way to try to get traction on the Brexit question is to look at party membership. The same Guardian piece gives us the following breakdown of UK party membership:

  • UKIP: 13%
  • Torries: 36%
  • LibDems: 7%
  • Labourites: 34%
  • ScotNats (SNP): 5%
  • Greens: 4%
  • Other: 1%

Let’s assume – why not? – that voters turn out for the referendum in numbers proportional to their party membership. (Will they? Probably not. But it’s not a terrible working assumption.) I’m willing to bet that UKIP is all in for breaking with Brussels, so that 13% for “leave.”

What about the ScotNats? This is a trickier question than I imagined. It might seem like a no-brainer for members of the SNP to vote for an exit to the EU and then follow that up with a second referendum on Scottish membership in the UK[4]. But SNP leadership is painfully aware that (i) losing a second referendum so soon after losing the first would probably cripple the independence movement for a generation (see Quebec) and (ii) the ScotNats might not yet be in a position to win[5]. In fact, some who favor Scottish independence would like to see it be part of the EU! (That’s not as odd as it might sound. Scotland has a long and troubled history with England, but that doesn’t mean the Scots wouldn’t want voluntary shared governance with other European powers.) So there’s no clear reason (that I can see) to think that ScotNats will vote overwhelmingly for Brexit. Indeed, one recent poll suggested that the Scots (who are not the same as members of the SNP, of course) favor remaining in the EU 3-to-1[6]. That gets us up to 15.5%.

The Tories seem deeply divided over Brexit – another anglophone conservative party in jeopardy because of a populist uprising[7]? Take these numbers for whatever they’re worth, but according to one recent poll, “45 per cent of voters said [pro-Brexit Troy Boris] Johnson was “more likely to tell the truth about the EU than [anti-Brexit Tory David] Cameron, while 21 per cent said the opposite”[8]. Boris has, as we know, said some extraordinary things of late:

Napoleon, Hitler, various people tried this out, and it ends tragically. The EU is an attempt to do this by different methods.

For reals, ya’ll. That’s the voice of truth, according to 2 out of 3 Tories. Well, suppose they do break for Brexit in slightly lower numbers – say 3-out-of-5 Tories vote “leave.” That gets “leave” up to 37.1%.

The Greens look like a pretty solid bet to vote “stay,” since EU environmental protection and regulation is bound to be much better than anything you’re likely to get in the UK alone[9]. The same is true, mutatis mutandis, for bankers and their City cronies, who are no friends of the Party of Kermit [10]. Suppose then that the Greens go overwhelming for “stay” but a few vote “leave.” That gets “leave” up to 37.75%.

It’s hard for me to suss out exactly what the LibDems are up to. Some of their leadership has made noises up the increased strain on the NHS if the UK leaves the EU[11]. LibDemster Tim Farron has called pro-Brexit thought leaders “downright pathetic”[12]. Suppose the LibDems go for “stay” in roughly the same numbers as the Tories go for “leave.” That gets “leave” up to 40%.

What about the Laborites? “Nearly two thirds of Labour voters say they are likely to vote to Remain – around six million people,” according to one source[13]. That sounds pretty good until you realize it also means that one third of Labour voters say that they are likely to vote to leave. Even if Jeremy Corbyn is able to step up – no sure thing as he’s rather ambivalent about Brexit [14] – that still means you’ve got 51% voting “leave.” So…wow…goodbye to all that.

Or not. All that was very back-of-the-envelope. But it’s not crazy. I might have (okay, I’m sure I do have) some of the numbers wrong. It’s been almost a decade since I lived in England, so I don’t exactly have my finger on the pulse of the UK. Still, this seems like a model I can work with, even if I have to modify the numbers a bit. Suggestions?

For now, I’m lowering my forecast for the UK remaining in the EU to 61%.

[1] http://www.theguardian.com/politics/2016/may/16/phone-poll-finds-10-point-lead-for-remain-but-web-survey-puts-leave-ahead-brexit

[2] The numbers from April and May are taken from https://ig.ft.com/sites/brexit-polling/

[3] http://www.wsj.com/articles/polarized-over-polls-internet-vs-phone-1454063404

[4] http://www.telegraph.co.uk/business/2016/03/25/would-britain-face-break-up-after-brexit/

[5] http://www.theguardian.com/commentisfree/2016/jan/14/brexit-golden-opportunity-nicola-sturgeon-nightmare and http://www.theguardian.com/commentisfree/2016/feb/25/brexit-vote-scotland-out-uk-scottish-independence and more recently http://www.thetimes.co.uk/article/snp-cools-on-referendum-after-brexit-0qdt27mg5

[6] http://www.dailyrecord.co.uk/news/politics/eu-referendum-record-poll-shows-7934908#qlCQwzRlRIzyCYhF.97 and http://www.theweek.co.uk/eu-referendum/65461/eu-referendum-poll-will-scotland-swing-the-vote

[7] http://www.ft.com/intl/cms/s/0/6e5737f6-defa-11e5-b67f-a61732c1d025.html#axzz48wm1DAHK

[8] http://www.theweek.co.uk/eu-referendum/65461/eu-referendum-poll-will-scotland-swing-the-vote

[9] http://www.theguardian.com/politics/2016/mar/14/green-party-loud-proud-backing-britain-europe-brexit-lucas

[10] https://www.greenparty.org.uk/news/2016/05/16/sian-berry-brexit-would-threaten-bankers-bonus-cap-in-london/

[11] http://www.ibtimes.co.uk/eu-referendum-lib-dem-leader-tim-farron-warns-pressures-nhs-after-brexit-1543112

[12] http://www.libdemvoice.org/liblink-tim-farron-cameron-and-corbyn-stance-on-brexit-downright-pathetic-47871.html

[13] http://www.independent.co.uk/news/uk/politics/jeremy-corbyn-could-hold-key-to-eu-referendum-vote-polls-shows-a6965751.html

[14] http://www.theguardian.com/commentisfree/2016/apr/15/jeremy-corbyn-britain-brexit-labour-david-cameron

Who will win the 2016 US presidential election? (update)

Will something like the Bradley Effect play a role in this 2016 presidential election? Recall that the Bradley Effect is supposed to manifest itself in the tendency for at least some sizable percentage of voters to say that they’ll vote for a black candidate but to vote for the white candidate instead[1]. I realize that it’s controversial in some circles whether we actually saw the Bradley Effect in recent presidential elections, much less have seen it in the election for which it was named[2]. But such controversies don’t matter for my present purposes. I merely want to think about the possibility that

  • (a) Voters are more likely to vote from Mr. Trump than tell pollsters that they’re going to vote for him.
  • (b) Voters are less likely to vote from HRC than tell pollsters that they’re going to vote for her.

Short answer: We don’t have, as far as I can see, good reason to think that either (a) or (b) is true. That removes (baring contrary evidence currently unavailable) one source of skepticism about the accuracy of the polls.

Long answer: Let’s start with (a). The thought goes something like this[3]: They are some voters who who recognize that Mr. Trump is seen by many as a racist[4], sexist[5], proto-fascist[6] thug[7], who might well be, according to Fox News(!), mentally ill[8]. Indeed, these voters might even think that Mr. Trump is some or all of the above.[9] Hence, they do not want to be seen as supporting him. However, the theory goes, they prefer Mr. Trump to HRC because they share these repellent attitudes, because think that HRC would be even worse, or for other reasons I cannot quite fathom. Hence, these voters will say to pollsters, to their friends, to their neighbors, etc. (according to the theory) that they won’t vote for Trump, but they will do so anyway.

The LA Times[10] discusses one poll that suggests a cautious “yes” to (a), while Red State expresses mild skepticism[11]. I think that Red State gets the better of this disagreement since the dispositive value of a single poll is slight. But the idea behind the poll seems like it’s worth exploring further. But we’ll have to await the work of clever (and well-funded) pollsters. Finally, it’s worth remembering that Mr. Trump consistently under-performed his polls until last month[12]. His recent tendency to over-perform his polls really dates, as I understand it, from his win in the New York primary.

What about (b)? I haven’t found any discussion of this possibility, and that fact surprises me. Perhaps the apparent failure of the Bradley Effect to surface in the 2008 and 2012 elections has soured people on this idea. But I’m not sure we should give up so quickly. It seems entirely possible that some people – especially some men – will want to sound right-thinking and enlightened by saying that they will vote for the first female president in US history while either refraining to do so or, quelle horreur, voting for Mr. Trump. Again, we haven’t seen much evidence for this. HRC has pretty steadily outperformed her polls[13], the opposite of what one would expect if the Bradley Effect were in operation.

 

[1] https://ballotpedia.org/Bradley_effect

[2] http://fivethirtyeight.com/features/persistent-myth-of-bradley-effect/

[3] http://www.cbsnews.com/news/commentary-are-polls-underestimating-trumps-support/

[4] http://www.huffingtonpost.com/entry/donald-trump-racist-examples_us_56d47177e4b03260bf777e83

[5] http://www.independent.co.uk/news/people/heres-all-the-sexist-things-that-donald-trump-has-ever-said-10452180.html

[6] http://www.slate.com/articles/news_and_politics/interrogation/2016/02/is_donald_trump_a_fascist_an_expert_on_fascism_weighs_in.html

[7] http://www.salon.com/2016/04/22/donald_trump_will_never_be_presidential_partner/

[8] http://www.politicususa.com/2016/03/18/fox-news-suggests-donald-trump-mentally-ill-obsessed-megyn-kelly.html

[9] https://www.washingtonpost.com/blogs/plum-line/wp/2016/03/24/gop-elites-think-trump-will-be-a-disaster-for-the-party-gop-voters-disagree/

[10] http://www.latimes.com/politics/la-na-trump-polls-20151221-story.html

[11] http://www.redstate.com/streiff/2015/12/24/donald-trumps-polling-show-reverse-bradley-effect/

[12] http://www.redstate.com/neil_stevens/2016/04/11/new-york-polling-continues-look-disappointing-trump/

[13] http://www.cnn.com/2016/04/15/politics/hillary-clinton-new-york-poll/ – This is true of primaries, but less so of caucuses.

 

 

 

Will Montenegro become a NATO member in 2016? (update)

“Montenegro [expects] to become the 29th member of the Alliance this year” (i.e. 2016) according to one source[1]. However, the article provides no evidence for this claim, and is mostly concerned with another question – namely, whether or not Georgia will ever become part of NATO. Still, I think it’s worth mentioning this only because it suggests there’s still a healthy difference of opinion among people who are (or ought to be) well informed about this question. This question is rather different from a lot of the GJ questions I’ve given some thought over the last few months. Here it’s not so much a matter of how do I analyze and weigh all of the data that I have; it’s a question of what don’t I know yet, and how might I shed some of my ignorance.

[1] http://www.europeanleadershipnetwork.org/georgia-and-nato-perpetual-threshold-_3741.html