Princeton Election Consortium

A first draft of electoral history. Since 2004

Diagnostics

July 27th, 2016, 9:29pm by Sam Wang


The HuffPollster feeds are changing structure to offer more options. We are adapting. Tonight, some routine diagnostics… [Read more →]

→ 8 CommentsTags: Site News

Differences between models

July 26th, 2016, 6:17pm by Sam Wang


Some of you ask why various models are currently giving different probabilities. At the moment, everybody’s above 50% probability for a Clinton win, but with varying degrees of certainty.

Here at the Princeton Election Consortium we follow a relatively simple procedure:

  • use the median and distribution of state polls to estimate each state’s current margin and win probability (see the right sidebar),
  • combine the probabilities to make an exact snapshot of today (the EV count above),
  • calculate how far national sentiment is from an electoral tie (the Meta-Margin),
  • calculate what would happen if the nation were to drift in either direction by November (gives the random-drift probability), and finally
  • impose a prior that the final outcome will be somewhere in the vicinity of where it’s been so far, a.k.a. regression to the mean (gives the Bayesian probability).

Polls only, no “fundamentals,” no pollster corrections. All of the above steps have a history and a justification. It’s a transparent process. I’ll populate with links later.

PEC’s model works well when applied retrospectively to 2004-2008, and was extremely good in 2012. It is accurate and stable – but it does not generate that much news. There is not much pressure to attract eyeballs around here. It responds to changing opinion, but only after state-level data is available.

Other sites use national polls and other factors. Josh Katz at The Upshot gives a detailed comparison.

→ 35 CommentsTags: 2016 Election · President

The polls are always bouncing to the left and to the right

July 25th, 2016, 9:30am by Sam Wang


For those of you not into adolescent pop culture of the 1970s, today’s title is explained at the bottom.

Updated to reflect new polls. The bounce is smaller than the original estimate.

We have five polls that measure a post-convention bounce for Presidential candidate Donald Trump (R):

  • NBC/SurveyMonkey: steady at Clinton-Trump 46%-45% (July 11-17 to July 18-24).
  • CBS: an increase from Clinton-Trump 40%-40% (July 8-12) to 43%-44% (July 22-24), a 1-point swing from a tie to Trump +1%.
  • YouGov: A 1-percentage-point change from 40%-37% (July 15-17) to 40%-38% (July 23-24).
  • Morning Consult shows a 6-point swing from 41%-39% (July 14-16) to 40%-44% (July 22-24).
  • CNN shows a shift from Clinton-Trump 49%-42% (July 13-16) to 45%-48% (July 22-24), a 10-point swing from Clinton +7% to Trump +3%.

The measure of the post-RNC bounce so far is a median swing of 4 percentage points 1 percentage point. For stragglers, see HuffPollster.

One point is not an impressive change. Recall that in states won by Mitt Romney (R) in 2012, [Read more →]

→ 43 CommentsTags: 2016 Election · President

Convention bounces in a polarized era

July 22nd, 2016, 9:55am by Sam Wang


Measuring the effect of a particular political event is challenging. Any single poll spans multiple days, and multiple polls are necessary to get good accuracy. National surveys give the first indication, within a week. State polls (upon which the PEC snapshot and forecast are based) are more accurate when aggregated, but take longer.

In the case of the Republican convention, we will have a hard time knowing what its effects are in isolation. Certainly the event was distinctive. Last night, Donald Trump entered the general election campaign with a harsh 76-minute speech that painted the U.S. as a dystopia, and his opponent as a criminal. These claims do not hold up to scrutiny – but they do show his approach for the months ahead. One might like to know the net effect of that speech, that of his endorsers and various other Trumps, and of Ted Cruz, who called for citizens to “vote your conscience.” It would be interesting to know if such a convention would close his deficit with Republican voters.

Almost immediately, the Democrats now take the stage. Today, Hillary Clinton is expected to announce her vice-presidential pick. Next week comes her party’s convention. Many polling measurements will capture the combined effect of all of the Republican and Democratic events. The conventions are close in time, making it easy for viewers to see contrasts between the parties. My guess is that the net effect should be relatively large, favoring the Democrats. If I am wrong, undecideds should still decrease.

However, how large is “relative”? The effects of conventions have been declining. Based on Gallup data, the median “net impact” (more-likely-to-support minus less-likely-to-support) of conventions from 2004-2012 was 5 percentage points, compared with 16 percentage points for 1984-2000. I would characterize this as entrenchment of voters, a feature of political polarization. I will be interested to see if the net impact of this year’s conventions is an exception to the trend.

Today the electoral snapshot is at Clinton 312 EV, Trump 226 EV. Margins in many states are quite narrow – note the pastel appearance of the electoral maps in the right-hand column. Consequently the Meta-Margin (how much swing toward Trump would be needed to create an electoral tie) is only 2.5%. This close lead, in PEC’s approach to prediction,  makes Clinton’s November win probability only 80%, in a range (20-80%) that I call uncertain. This is our starting point for evaluating the weeks ahead.

→ 26 CommentsTags: 2016 Election

Politics And Polls #4: An Unconventional Convention

July 21st, 2016, 6:34pm by Sam Wang


Politics & Polls #4 (podcast): in which Julian Zelizer and I chew over the convention’s role in shaping the race. Bonus overtime minute: Sam suddenly realizes how the Cruz/Trump smackdown is basically wrestling theater (be sure to click – must-reads from Washington Monthly and Balloon Juice).  Guess who’s the heel? (hint: not Trump.)
[Politics & Polls on iTunes] [on SoundCloud]
Courtesy of the Woodrow Wilson School.

→ 5 CommentsTags: 2016 Election · President

Senate update

July 20th, 2016, 3:51pm by Sam Wang


Indiana is updated to be Young (R) vs. Bayh (D). Bayh is a massive favorite, and his entry into the race pushes the overall Senate snapshot to a median of 50-50. [Read more →]

→ 9 CommentsTags: 2016 Election · Senate

Paths of Glory

July 19th, 2016, 10:27pm by Sam Wang


Welcome, New York Times readers! Josh Katz unveils The Upshot’s model for the November election – complete with comparison to other sites: electronic bidders (PredictWise), polls-only (FiveThirtyEight and the Princeton Election Consortium), and professional prognosticators (Charlie Cook, Stuart Rothenberg, and Larry Sabato).

Qualitatively, PEC and The Upshot mostly agree, though PEC shows less confidence in Nevada, Ohio, and Mississippi. That is likely because [Read more →]

→ 22 CommentsTags: 2016 Election · President

The party conventions begin

July 18th, 2016, 8:01am by Sam Wang


The Republican Party’s national convention starts today [schedule]. Conventions are a chance for a political party to showcase their unity, their candidate, and their policies. Next week the Democrats take the national stage. Viewers will get a fairly direct contrast.

As measured via state polls, the Presidential race shows Hillary Clinton slightly ahead of where Barack Obama was at this point in 2008 and 2012. So far, the 2016 race has been stable.
History of Popular Meta-Margin for Obama
National polls suggest that undecideds account for about 5 percentage points more of the electorate than the Obama-Romney race, and Trump is particularly weak in states won by Romney. Based on these numbers, Trump needs to bring Republican voters home.

Downticket, in many key Senate races, general-election candidates are not known yet. I will not have anything to say about that for a few weeks. For now, a more interesting indicator of downticket sentiment is the generic House preference, which is currently at Democrats +7.5%. If that were to hold up, Democrats would have a good shot at retaking control of the House. However, preferences can change. Based on past years, it should take at least a month for us to get a fuller picture of where the national House race is headed.

→ 32 CommentsTags: 2016 Election · House · President · Senate

FOO (freakouts over outliers)

July 15th, 2016, 1:00pm by Sam Wang


A CBS/NYT poll released yesterday indicated a 40-40 tie between Clinton and Trump. Cue media freakout. This illustrates the point that news organizations habitually report on outlier events, a bad move when it comes to data points when other data points are available.

Four other surveys with mostly overlapping dates show Clinton +1% (Morning Consult), YouGov/Economist (Clinton +3%), AP-GfK (Clinton +4%), and Clinton +12% (Raba Research). Five data points, with CBS at the extreme end. Oldtime PEC readers, all together now: take the median. The median is Clinton +3.0 ± 1.3% (± estimated one-sigma uncertainty). So the race may have narrowed from a 5% gap – maybe because of FBI director Comey’s public announcements? Anyway, it’s not a tie yet.

Same story with state polls. [Read more →]

→ 39 CommentsTags: 2016 Election · President

Politics and Polls #3: A Referendum on the Incumbent Party?

July 14th, 2016, 3:30pm by Sam Wang


In Politics & Polls #3 (podcast), Julian Zelizer and I talk about the extent to which the election is a referendum on the governing party…and who that might benefit. Courtesy of the Woodrow Wilson School.

[Politics & Polls on iTunes] [on SoundCloud]

→ 8 CommentsTags: 2016 Election · House · President