Pretty Clear Trendline in GE Matchups between Hillary and Rubio

When Nevada Democrats entered caucuses this weekend, 18% of them (in the 25 precincts surveyed out of 1700+ total) told pollsters that “ability to win” represented the top trait they seek in a candidate. Hillary Clinton won these voters 80%-15%. These were enough voters to put her over the top in the caucuses. Are these voters correct though?


In hypothetical matchups between Clinton and the Robot Rubio, Hillary led in most polls early in 2015 and has trailed since the fall. Bernie has done much better in recent matchups but the data is sparse. As a Democrat, this concerns me a bit as does Hillary’s overall favorability since she left the State Department.

hillary-fav-02-22Those supporting Hillary Clinton might want to look at different reasons than electability.

Outsider GOP Candidates Stalled Out in South Carolina

A flurry of polls have come out in South Carolina heading into the GOP primary on Saturday. After opening up a large lead, the momentum for the “outsider” candidates (Trump, Cruz, Carson) has stalled out and the “insider” candidates (Robot, Jeb!, Kasich) have moved up. However, the outsiders will likely win as not one poll has shown the outsiders behind.


As far as the individual candidates, looking at the polls released from February 15-17, which represent 4,336 interviews, Trump holds a clear lead. Cruz and #RobotRubio are close, and ahead of Bush and Kasich. Dr. Ben is last. For Rubio to finish second significantly ahead of Cruz, he needs to dig into the 16-20% of the electorate currently sticking with Jeb! and Kasich. If he does, he has a chance to win this mess. If he fails, he’s toast. The opportunity is there though when it didn’t look like it a week ago.


What Changed in New Hampshire Once People Started Voting

The great thing about Presidential nominating contests is they provide new polling and election data every week. For general elections, we pour over mountains of polling data in anticipation of a singular event. Here, we can see trends and help shape our thinking going into the next contests.

Looking at the performance of individual candidates, as opposed to “lanes,” we can see how the vote shifts right at the end and how events at the end can be determinant (or that polls are systematically wrong). For Iowa and New Hampshire, I collapsed all the polls done in the last three days into one estimate and confidence interval. Pollsters published 3,998 interviews in New Hampshire (in 8 polls) and 2,667 in Iowa (in 4 polls). Caveat – collapsing different polls with different methodologies creates biased data. I get that, but can only go with what I can freely get.

In New Hampshire, Trump and Kasich, the top two, overperformed the polls. Rubio severely underperformed as his debate performance was universally panned by humans and robots.

11rubio-web1-master675Interestingly, despite the conventional wisdom that Christie created a “murder-suicide” pact by attacking Rubio so aggressively, he overperformed as well. The only problem was that he was doing terribly to begin with and had no chance at all to either win or break out of the pack. (Below, the black dot and black line represent the poll estimate and the confidence interval where we would expect the result to fall 95 times out of every 100 independently conducted poll of 3,998 voters. The red dot represents the final result.)


In Iowa, the story differed. Trump underperformed while Cruz and Rubio overperformed. How much of that had to do with the debate nonsense from the previous Thursday, where Trump bowed out of a Fox debate and held a competing rally to “raise money” for veterans instead, we can’t say with the data we have. However, it does show how volatile primary elections can be right up until the end.


While the “insider vs. outsider dynamic” has been stable in both places, how each candidate will perform is not. Polls give clues, but are not definitive. That what makes this fun (when you are not polling for the candidates).

New Hampshire Polls Pretty Much Right

At the macro-level, the polls in the New Hampshire GOP primary race were pretty accurate. The outsiders (Trump, Cruz, Carson, and Fiorina) beat the insiders (Jeb!, Kasich, and the Fat Man and Little Robot) 53.5% to 44.7%. A narrow victory but expected. The insiders gained and the regression line picked it up. The outsiders got about a click above projected.


The big question going forward is when do the insiders win. They trail big in South Carolina (only 10 days away) and other than Kasich’s native Ohio, the insiders have no advantage in any other state. Even as the insiders drop out, the remaining need to fundamentally alter the underlying dynamic of the race. If the strategy is to get it to Kasich or Rubio against Trump and Cruz and hope to narrowly prevail (~35-40%) in 3 way contests in the states with “winner take all” delegate rules, good luck.

Some New Polling Charts about the NH Primary and Beyond

The New Hampshire primary is coming tomorrow and with it, a flurry of polls. The difference between the “outsider” (Trump, Cruz, Carson, Fiorina) and “insider” (Rubio, Christie, Bush, Kasich) has varied wildly by poll but it looks like the outsiders have the edge.


As far as who leads, unlike in Iowa last week, Trump holds a lead over Rubio that is outside the margin of error in every recent poll.


Rubio came in third in Iowa because he ended up taking 78.5% of the “insider” lane vote. He comes nowhere near that share in any New Hampshire poll conducted over the last week.


Trump should win. If he doesn’t, never believe another public poll.

While the “outsiders” only narrowly edge the “insiders” here, as the campaign shifts to South Carolina, the “insiders” have a long way to

This poses a major problem for Rubio as his 3-2-1 strategy of coming in third in Iowa, second in New Hampshire, and first in South Carolina hinges on actually winning South Carolina. He is nowhere close to that and debate performances like this won’t help.

On the Democratic side, the Bernie Sanders surge has ended and he has trailed off a bit, but he holds such a massive lead over Hillary Clinton, her campaign is getting ready to decide who to throw overboard.


Sanders has no time to rest. Even as there hasn’t been a South Carolina poll in a month, he trails significantly there and only has 2.5 weeks to make up the gap.



Iowa Polling Not as Bad as Reported

Last night, the Iowa Democratic and Republican parties held the first in the nation caucuses to select the next President of the United States. As a high profile election, with lots of public polling, Iowa serves as yet another tedious referendum on the accuracy of polling (see here – “Will Trump Save the Pollsters” – and here – “Iowa Is the Hardest State to Poll” for some of many examples).

As with any poll, the first question a consumer should ask is “what is this poll telling me about the story of the race?” This is far more important than the 10 individual estimates of candidate support. The meta narrative about the Republican race in Iowa has been that caucus voters overwhelmingly prefer candidates from outside of government, or have positioned themselves as opponents of government from within (Trump, Cruz, Carson, Fiorina) than candidates from within government and have been involved in high profile bipartisan deals on important issues like immigration (Rubio, Kasich, Christie, Jeb!).


As far as this narrative, the polls were dead on. The outsiders got 63.1%, right on with the final estimate. The insiders got 29.5%, just slightly higher than the estimate.

The polling looks “off” because it got two unexpected results from a straight forward analysis of the estimates of voter support of each candidate. Every poll had Trump “ahead” of Cruz and Rubio significantly behind them. Cruz won and Rubio came in a strong third (a position that launched the presidencies of Lamar! Alexander in 1996, Alan Keyes in 2000, and Ron Paul in 2012).

Were these really “mistakes”? Only if you assume that the result in a poll is a precise estimate, (which it is not) and there is no sampling error (which there is). In the last nine polls in Iowa, Trump had a statistically significant lead over Cruz in just two of them. (Trump’s margin is in blue, Cruz’s in red). For polls with overlapping confidence intervals, you can not say that Trump led Cruz:


As far as individual polls, the Des Moines Register poll has a reputation for being the most accurate. How did they do?


Cruz and Rubio overperformed their poll result. Trump and Christie underperformed (albeit at barely the level of statistical significance). The pollster boldly declared that “there’s no indication of a [Rubio] surge: His support declined during the four days of polling.” That did turn out to be wrong, but entrance polls show that 29% late deciders broke to Rubio compared to 14% for Trump.

A late poll, by a college in Massachusetts using automated phones to call just 298 likely Republicans (not marks of reliability), caught the Rubio surge but offered no clear guidance which of the top three candidates would win (and got Carson’s support way off).


Rubio ended up getting 78.2% of the “insider” vote, much higher than his share of that vote in polling. The “insider” vote didn’t grow, it just coalesced around Rubio. If that happens next week in New Hampshire, a state where the Governors Kasich, Jeb!, and Christie have spent considerably more time and resources, he could come close, if not win. The outsiders edge here is much narrower than in Iowa (albeit is growing).


Polls will be much more useful for consumers if both the media outlets and readers adjust expectations in what is observable and what is not. Getting precise estimates of 10 Republican candidates is not realistic and should not be either the goal or the measuring stick. Close poll results will not predict win, place, or show. But they will give context to what is going on in the race and which factors will be determinate in answering the big questions.

One last note, polling showed a close Democratic race, and the result was a virtual tie. That should not be conveniently forgotten.

IA-NH Democrats: It’s Not Us Hillary, It’s You

With all the caveats about the problems in election polling (much of which is overblown), Bernie Sanders has caught, and maybe passed, Hillary Clinton in Iowa.


Regardless of who ends up with the most precinct delegates tonight (there will actually be no overall vote tallied in the Democratic race), Iowa’s caucus voters are pretty much split, with Bernie gaining support during the entire contest. Sanders continues to dominate in New Hampshire.


Hillary’s lead nationally and in the future primary/caucus states could rapidly dissipate if she loses both of these states, particularly if she loses New Hampshire (a state where she beat Barack Obama) as badly as it looks like she will.

GOP Voters Really Don’t Want Insider Candidates

This hardly qualifies as the most insightful analysis of the 2016 GOP nomination contest.  With all the caveats about the unreliability of national polling, using the absurd nomenclature of GOPers involved in the race, the “outsider lane” (Trump, Cruz, Carson, Fiorina) took the lead over the summer and has dominated the “insider lane” (Rubio, Jeb!, Kasich, Christie).


The insiders haven’t budged at all in more than 6 months and countless debates and media appearance. The same trend has played out in Iowa.


The outsiders now take about two-thirds of the vote in pretty much every poll while the insiders are stalled out. New Hampshire is different. The outsiders have dipped recently while the insiders have brought it to a dead heat. The only problem is that Jeb!, Rubio, Kasich, and Christie all get roughly the same – high single digits, low double digits, the 8-13% range. Thus, Trump beats them in every poll.


A fun race. Cue the singing