Princeton Election Consortium

A first draft of electoral history. Since 2004

Consequences of Iowa: Trump still strong, new life for Rubio, long-term trouble for Sanders

February 2nd, 2016, 1:21am by Sam Wang


My preliminary take on the Iowa caucuses is that they didn’t alter the trajectory of where things are probably headed for the Democrats: Hillary Clinton is still favored. However, the Republican field could potentially narrow to a three-way race (Trump-Cruz-Rubio) sooner than I had expected, thanks to a strong showing by Marco Rubio.

On the Republican side, Donald Trump and Ted Cruz switched places relative to polls. To compare the final polls with tonight’s counts, Trump underperformed by 26.5-24.3=2.2%, Cruz overperformed by 27.7-23.5=4.2%, and Rubio overperformed by 23.1-18.0=5.1%. The late swing for Rubio was visible in the final days of polling. All of this is well within the range of normal polling error in primaries. As expected, multiple delegates went to Cruz (7), Trump (7), Rubio (6), and Ben Carson (3). Numbers updated to reflect exact vote shares. After all the fuss, Cruz and Trump appear to be tied for first.

It is premature to say that Trump is doomed. However, he does look a little less inevitable. It is certainly possible that he can crash from his high position in New Hampshire, South Carolina, and nationally. But I think a bigger risk to him is the possibility that tonight’s results will pressure Rubio’s lower-tier rivals to get out sooner rather than later. As I’ve written before, if the field gets down to three candidates after New Hampshire, that opens up a narrow route to stopping Trump. In short, tonight kept Marco Rubio’s chances alive.

On the Democratic side, tonight was substantively bad for Bernie Sanders. After all the talk about hordes of Sanders supporters, in the end he only achieved a near-tie: 23 delegates for Clinton, 21 delegates for Sanders. Iowa is one of the most favorable states for him because of its ethnic composition. But it is not enough to win 50% of white Democrats. To have a chance overall, he needed a big win to (a) indicate that he can get enough white support to compensate for lack of support in nonwhite demographics in other states, and (b) create press coverage to boost him in the coming weeks. Outcome (a) didn’t happen. We’ll see about (b).

One of the most notable features of the Democratic race was the age gap. In an entrance poll, Sanders led by 70% among voters aged 18-29, while Clinton led by 43% among those aged 65 and over. That is a 113-point gap. This difference surely is on the minds of both sides for the weeks and months ahead.

Finally, a word about polling. There seems to be a persistent meme that polls are in trouble. There was no evidence for this. Primaries and caucuses are volatile situations – this is a well-known fact. I have been assuming that home-stretch polls can be off by an average of 5 percentage points. Any fuss tonight is based on the fact that in Iowa, with its tiny turnout and odd voting procedure, Trump was polling 3 points ahead of Cruz, and ended up losing by 3 points. It would be a mistake to conclude that Trump’s support is illusory in other states. Quite the opposite. A 6-point error would not affect his ranking anywhere else. For now.

Tags: 2016 Election

93 Comments so far ↓

  • Mark F.

    Hillary has already written off New Hampshire. What happens on the GOP side will be interesting. Bush, Christie, Fiorina, Carson, and Kasich all pretty much have to get at least 10% to be viable going down South, and also probably a top 4 finish.

  • Matt McIrvin

    People are currently touting individual polls showing *gigantic* shifts in the race after Iowa: a PPP poll showing Trump’s lead over Cruz down to four points, and a Quinnipiac one showing Clinton’s lead over Sanders shrunken from about 15 to two points. Can this even be possible?

    Notably, the PPP poll doesn’t show as large a shift as the Quinnipiac one for Clinton/Sanders, nor the reverse.

    • bks

      I think the performance of the Selzer poll reminded us that Sam’s poll-agnostic attitude is correct. We should expect the individual polls to wander around the truth, whatever that is.

    • Matt McIrvin

      Whoops, now TPM is touting some poll showing Cruz in free fall! Drama, drama, drama!

    • JayBoy2k

      Here is Washington Post article from yesterday showing 5 polls with Trump’s lead basically holding.
      https://www.washingtonpost.com/news/the-fix/wp/2016/02/05/new-hampshire-polls-are-all-over-the-place-just-like-normal/ and with the comment that NH polls are all over the place. In Field against Trump, the $$$ still goes to Trump with multiple candidates popping up and down.
      I do not see how Trump or Cruz voters end up switching to any of the others except maybe Carson.. As long as you have Rubio, Christie, Kasich, and Bush all fighting for the same voter, it is hard to see how Trump falls to 2nd here.

  • Mark Buckley

    On Cape Cod I’m following Boston’s ch 7 / UMass daily tracking poll which has Rubios numbers for the last four days at 8, 10,12 and 15 today suggesting that there might not be an establishment lane GOP candidate this year . Rubio’s comment yesterday on Obama’s visit to a mosque was ” divisive ” shows he’s a Theocrat who cares only about religious freedom when it applies to his own . Very different from Bush ( whose brother also visited a mosque after 9/11) or Kasich who are more moderate by comparison. Expecting Trump in an upcoming debate to bring up Rubio’s brother -in – law ‘s felony conviction for cocaine dealing . As in ” Hey Marco , where did the 12 million go and why did you get him a real estate license after he got out of jail ” for example . Never a dull moment and everything’s on the table this year .

    • Mark F.

      Rubio’s numbers are rising, so he is in a good spot to come in a very strong 2nd in New Hampshire and perhaps win it. Of course, Kasich might have a strong showing as well, which would keep him in the game.

  • A New Jersey Farmer

    Trump now accusing Cruz of fraud and of stealing the Iowa election. Donald is coming undone and his frustration is showing. This is the beginning of the end for him and he’s going to drag Cruz right along with him. Rubio and at least one of Bush, Kasich and Christie will gain.

    http://www.nytimes.com/politics/first-draft/2016/02/03/donald-trump-says-ted-cruz-stole-victory-in-iowa-caucuses/

    • 538 Refugee

      Also, if you look at towards the end of the article it appears ‘Team Cruz’ mailed intentionally misleading letters that appeared to be official government communications telling people that the Iowa government was grading people on their voting records and they would receive “F’s” if they didn’t attend the caucus.

      I guess I really don’t understand why the evangelicals support him. I thought evangelicals was some Christian thing built around the teachings of Jesus. Obviously I’m misinformed.

    • Mark F.

      I thought this for a few hours. But don’t underestimate Trump. It probably won’t matter much. People have been predicting that he is “done” from the moment he entered the race.

    • Anthony

      None of Trump’s antics have proven to hurt him yet, and there has been plenty.

    • Amitabh Lath

      Farmer, I believe accusations of fraud and unfair treatment are aimed at Trump’s demographic. These are people who feel themselves to be downwardly mobile and feelings of being cheated out of rightful gains (by immigrants, China, bankers etc) may resonate.

  • Amitabh Lath

    Paul drops out. Who gets his one delegate? A true libertarian would auction him off to the highest bidder.

  • bks

    Delegates are the name of the game. According to the The Green Papers The delegates were apportioned 7,7,6,3,1,1,1,1 (+3 uncommitted)
    http://www.thegreenpapers.com/P16/IA-R
    This differs from other sources. WTF?

  • Mark F.

    I am now reminded of the “Fraud At Polls” headline in “Citizen Kane.” And I take back my comment that Trump isn’t stupid.

  • bks

    This is the poll that I will be watching closely this week:
    http://www.uml.edu/Research/public-opinion/polls/default.aspx
    It’s a daily tracking poll in NH. I’m not interested so much in the absolute numbers but rather if it catches a surge of late-deciders.

  • Olav Grinde

    I was watching the results on the official Iowa Democratic Party website yesterday – with frequent page refreshes. That website had the most up-to-date count. Hillary Clinton’s lead over Bernie Sanders was steadily diminishing.

    When 91 % of the precincts were counted, that margin had decreased to 0.2 %.

    Then something very strange happened: the margin suddenly increased, all the way up to 0.8 %.

    The bizarre thing is that the margin changed so drastically, while number of precincts counted remained unchanged: 91 %!

    I do hope that Professor Sam Wang, or someone who is familiar with the Iowa caucus can explain this. I just don’t see how it can be possible.

    Something very strange happened to the counting in Iowa.
    .

    PS. I still have the email that I sent to a friend at 11:15 PM last night, commenting this exact observation. So I am sure about what I saw – and I am sure that this happened just before I sent that email.

    • P G Vaidya

      I second your observation! I was watching the margin decrease steadily and I was calculating by what percentage Sanders would have to outperform in the remaining precincts to win and that margin was also decreasing very steadily and just around that 90 percent mark, it seemed almost certain that Sanders would eke out a win and then the mirror cracked!

    • Some Body

      Just guessing, but it could have been a batch of remote votes getting uploaded (there was an option of remote caucusing for Dems this time around).

  • A New Jersey Farmer

    Neither Trump nor Cruz will be the nominee. Trump’s campaign was built on inevitability and steamrolling the field. Now that he hasn’t, I don’t think his soul is in it and I don’t think his numbers will stay high. he was banking on anger and once that spell is broken, it’s difficult to get it back.

    Cruz is a perfect Iowa candidate, but not much else. He won’t have time to create the ground game necessary to win a sustained number of states, especially on Super Tuesday.

    Rubio is the one to beat now, unless one of the moderate candidates shoots to the top in NH. Won’t be Christie, though.

    • Anthony

      I totally agree with you on Cruz (I made a post a ways down outlining the same thing), however there is no data, as of right now, that Rubio has a chance of beating Trump. That won’t become clear until after NH. As a totally unconventional candidate that built his base of support on anger and bigotry there was an uncertainty that Trump voters will actually come out and vote, which he has proven even if he slightly underperformed.

      Even if he underperforms in NH, his lead is so large over Rubio (as of right now) and has such a large lead national that it’s hard to imagine his chances of winning the nomination being less than 80% assuming polls don’t move drastically in the next week (both NH and national numbers).

    • Sam Wang

      I agree, basically it’s a quantitative question. I am prepared to believe Trump’s numbers will go down. But will he and Rubio really switch places? That would be one heck of a swing. We will find out.

  • Olav Grinde

    I just heard on WGBH (NPR radio) that about a dozen precincts were decided coin toss!

    Sam, do you have data on this – and on how many coin tosses were in Hillary’s and Bernie’s favour?

  • Jack Tenold

    Sam, thank you a thousand times for taking the 2014 data off the top of the page. Now when I come to the site I can continue to drink coffee instead of searching for the gin.

  • Mark F.

    NH is critical for Rubio. He needs a first or very strong second place. The rest of the pack besides the top 3 need strong showings in NH or they are out (Paul, Christie , Kasich, Bush, Carson , Fiorina) There will be huge pressure for people to unite behind one “establishment” candidate after next week. Rubio looks like the best bet.

  • Andrew EC

    Isn’t “underperforming” polls by 2.2% exactly what we should have expected for a first-time candidate with no real ground infrastructure in a caucus state with (relatively) complicated rules?

    Obviously today’s media take is “Trump is A LOSER,” but I wonder if there isn’t time in the next week for a counter-narrative to emerge once the more data-driven pundits start to dig into the actual numbers.

    • Phoenix Woman

      I was surprised to see Trump do as well as he did against Cruz in Iowa.

      Remember, Cruz has the backing of a big chunk of Iowa’s Republican establishment, including the Bible wing run by Vander Plaats. He’d been there for a full year. And Trump comes within four percentage points of taking him out.

      Trump needed Iowa far, far less than Cruz did.

  • Surface Thought

    Sam and others,

    The demographic problem will certainly be a problem for Sanders going forward, and I will be interested to see to what extent he is able to make inroads with minority communities as he is able to campaign more specifically to them.

    That being said, I also wonder the extent to which Bernie will also see boosts in states that have Primaries as opposed to the opaque, closed caucus style of Iowa that would seem to help favor Clinton’s older, establishment base.

    Thoughts?

    • Mark F.

      Sanders needs to make huge inroads with older voters, women and minorities to be competitive. I don’t think he can do it. There is also the fact that Clinton already has a huge lead with “super delegates” counted. By the time we get to March 15 , I expect she will have the nomination all but locked up. Things could even be over long before then, if she can pull off a New Hampshire win.

    • The Raven

      Sanders has a long history of strong opposition to racism and racist policies, which Clinton does not.

      Is this line of argument valid?

    • Kevin

      Um, no. Sanders is not more distinguished than Clinton in this area.

    • Josh

      I actually think you have it backward. Iowa’s system is neither opaque nor closed: especially on the Dem side, voters argue openly with one another about which candidate to support, and everyone in the room can see who you choose. My guess is that this format allows the more zealous voters to convince (cajole?) people on the fence to come to their side. A primary, where you go into a booth, alone, might be less conducive to a candidate like Sanders over-performing?

      I do also think that a primary versus a caucus will help Trump.

    • Q

      I think the biggest beneficiary of a secret ballot is probably Trump, due to a variation on the Bradley effect where his voters may be reluctant to openly express their support to pollsters.

  • Amitabh Lath

    I’m with Sam here, I don’t understand how this is anything but good news for Trump. He is behind by one delegate, and Cruz has no numbers in NH and beyond.

    Bad news in my opinion would have been if Trump voters failed to show up (in other words massive LV fail). This did not happen. Primaries are easier to participate in, so they will show up again.

    Trump himself seems to understand this, ergo his gracious concession speech.

    • Mark F.

      Don’t think Trump is stupid for one minute. I’m sure he is getting very good advice right now. for all practical purpose, Iowa was a virtual 3 way tie.

    • Matt McIrvin

      It didn’t last. Trump is now claiming that Ted Cruz somehow rigged the caucus and demanding that it be nullified!

    • Matt McIrvin

      Apparently one of Cruz’s staffers was (falsely) telling caucus-goers that Ben Carson had dropped out. Trump is crying fraud and demanding a new caucus.

    • pechmerle

      Most of the story on this seems to be detailed here. http://thehill.com/blogs/ballot-box/presidential-races/268051-steve-king-i-had-obligation-to-tell-iowa-voters-about
      Rep. Steve King’s acting as he did isn’t exactly up to chivalrous ideals, but he does purport to have an (internally contradictory) excuse. The Carson campaign said that he had returned to Florida only because “he needed to get some fresh clothes.” How feeble is that!
      Trump now sounds like just sour grapes, but — as usual — will any of his supporters care.

  • Scott

    I think Trump suffered in the ground game. He has the support, he just didn’t have the ground organization out there knocking on doors and getting people to actually show up at the caucus. If this doesn’t improve, he’s in trouble everywhere.

  • Sheila

    Iowa is one of the most favorable states for him because of its ethnic composition. But it is not enough to win 50% of white Democrats. To have a chance overall, he needed a big win to (a) indicate that he can get enough white support to compensate for lack of support in nonwhite demographics in other states,

    *********************

    i have read this in many places and i am stunned how nonchalantly we talk about a DEM candidate who only can only draw white voters..as if that is OK…as if that is acceptable…

    • MarkS

      The most recent numbers I could find on Sanders support among black voters was 20% in South Carolina, in December (Fox poll). I expect that number to go up. This is very far from drawing “only white voters”.

  • The Raven

    Considering that Sanders has not taken big money and got amazingly poor coverage in the press, I think he made a strong showing. Strong enough to win? Perhaps not. On the other hand, when Marco Rubio is looking like the moderate alternative, you know the Republican Party has fallen off the edge of the world.

    • Kevin

      He sure pasted Dennis Kucinich.

    • Robert

      It is not accurate to say Bernie did not get fair coverage. neither he or Hillary got coverage like Trump did and all I heard for 3 straight weeks was how Sanders was surging in Iowa and Hillary was in trouble. I am not sure what more you want from the media. they are not there to help any candidate and if a candidate cannot navigate the media waters perhaps it is more of a signal about their unpreparedness than the media’s conspiracy to gang up on the Bern!

    • Mark F.

      “Sanders has a long history of strong opposition to racism and racist policies, which Clinton does not. ”

      A lot of Clinton supporters would strongly disagree with you. Regardless, what matters is whether or not minority voters have this perception and will vote for Sanders. He needs way too many of those voters to stage an upset IMHO.

  • Olav Grinde

    There is one Black Swan waiting to happen: Who do Americans vote for if there is a full financial meltdown this year – a much worse meltdown than the last one?

    Dr Wang, I would be very interested to hear your take on this. How do you believe a financial meltdown, before the Presidential Election, might influence the results?

    • Mark F.

      Another possible “black swan” is a major terrorist attack.

    • RDT

      I think if we’re discussing the possibility of something happening, it’s not really a black swan…

    • Matt McIrvin

      A “much worse meltdown than the last one” would be an extraordinarily bad meltdown, a 1929-level catastrophe. I could see any number of shocking things happening in American politics as a result of something like that.

      I expect Republicans would be swept into total power, and probably swept out again as soon as possible when the real misery began to bite.

  • Mark Buckley

    Guessing Trumps support is real and will show up again as he did get the second most Iowa caucus votes in Republican history – over 43, 000 !

    • JayBoy2k

      This may be the start of a surge for an “Establishment” Candidate” to compete with Trump/Cruz winner. This is not the end. Trump is currently up 20-25 points in the NH polls. Is anyone putting money on the idea that either Trump or Rubio closes that gap in the next 7 days?

      I do think Cruz is hoping for 2nd or 3rd and it will be interesting if his ground game works in NH. If Rubio continues to surge, the “losers” may well be all the other Establishment candidates.

    • Some Body

      Well, recall Obama vs. Clinton 2008. Polls went from a double-digit Clinton lead to a double-digit Obama lead within a week (but, of course, Clinton won, more narrowly, in the end).
      [Caveat: I'm citing figures from memory and might be exaggerating a bit]

  • Olav Grinde

    Does anyone know why the official Democratic and GOP results are both missing some precincts. You would think they would have managed to tally 100 % of the caucus votes by now!

  • The duke

    Thanks for responding to your commenters. My comment from last night was more about your Jan. 5th post which was a bit more confident about using just polls (which is a “modeling” choice, sorry – potatoes potahtoes) to figure out the path to the nomination than other folks who incorporate other information (endorsements, pollster weights, and others who use qualitative inputs, etc.). I was looking for a bit of self-reflection, and I guess I’ll take this: “That’s the way it goes when outcomes are uncertain.” Thanks again for taking the time to do all of this. I learn a lot from it.

    • Sam Wang

      I tried to be careful in the delegate-rules calculation to make it generically about anyone who is leading in a 4-candidate field. Obviously I meant Trump, but the same calculation would apply to any leader. The question is a simple one: how to (a) estimate opinion, then (b) convert it to delegates. I don’t see any problem with what I did, unless you want me to remove the word “Trump.” Many people want that word removed.

      In regard to non-polling data, I think one has to be careful about rules that appear to have predictive value – but only until they are upended. Endorsements and prior office-holding experience are examples of that. How can one figure out when those rules collapse? Since modern election prediction only dates to 2004/2008, there is not that much track record to go on. To my understanding, a econometric modeler might just sweep it under the rug with a prior that says “10% of the time, The Party Decides will fail,” or something like that. Anyway, in this domain Drew Linzer does the kind of job I would aspire to, if I were completely immersed in it.

      Here’s another way of thinking about the predictive-value problem: to a scientist like me, non-polling variables are unsatisfying because they are not linked to outcomes with what I would call a clear mechanism. Then it becomes a question of taste, intuition, and attending to a tremendous number of details.

      In my view, polls, whatever their faults, directly measure opinion at some point in time. Except at special moments (NH primary, IA caucus, national disasters), opinion usually takes multiple weeks to move. That is a good anchor for a calculation. Of course polls have their limitations, but I think those limitations can be listed in a finite way, so it is possible to specify most of the problems in one essay, or one comment thread, or one MATLAB script. Mmmm, finite.

  • DaveM

    I’ll be interested to see whether the “humbling” defeat for Trump in Iowa goes the way of the Mexican rapist remarks, the McCain putdown, the Muslim immigration prohibition, etc.—that is, as an event seized upon by the media as the long-awaited smoking gun and thereby turned into yet another galvanizing force for Trump fans.

    Are there any good graphics available representing the correlation, if any, between Trump’s “gaffes” and his poll numbers?

  • AEKH

    Main question now is how Trump and his supporters in other states handle him taking a hit, and how the GOP shakers and movers react.

    I am still very wary of Cruz. He’s slick, cowardly, clever and can manipulate the media and narratives. I would much rather have Trump as the nominee against Clinton.

    • Todd S. Horowitz

      Maybe, but on the other hand, Cruz is a true believer. Trump doesn’t give a shit about consistency or ideology; he can turn on a dime in November and change his story without shame or backlash. I think Trump’s utter lack of principles or shame makes him much more dangerous.

  • Amitabh Lath

    My zeroth order take on this is there is no big pollster fail like UK or Israel (it’s a minor point that they didn’t get the horse race right, but other than Ann Selzer losing her aura of infallibility I don’t see what difference that makes overall).

    Results within a couple of sigma of expected.
    Apply this to NH, SC, NV and the contours of the race become clear.

    • bks

      Yes, upon reflection the polling is not that bad for such a complicated process. What’s the deal with the three unpledged delegates?

    • Sam Wang

      It looks like Republicans have 27 Iowa delegates, all to be allocated proportionally. Currently, 26 have been assigned, according to the NYT. Democrats have 44 delegates, all assigned.

    • bks

      I’m pretty sure that Iowa GOP has 30 delegates. Apparently three party honchos go as unpledged delegates.

    • bks

      I see. There will be 3 unpledged GOP honchos for every state: the the RNC apparatchiks. So Iowa has 27 plus these three.

    • Sam Wang

      Yes, that’s correct. Sorry, up too late.

    • Some Body

      Just had a detailed response submitted with the gist that Israel is actually not a good example of polling failure (pollsters did quite well, considering limitations, and certainly no worse than in previous elections). It didn’t get through (maybe because of the length, triggering moderation?) Anyway, Sam, if you publish my original comment, do erase this one :-)

    • Sam Wang

      I’m sorry, Some Body. For reasons unknown, WordPress has swallowed this comment.

      I agree that Israel was sui generis. However, I do think it’s worth thinking about how to poll situations with many candidates and possible strategic voting. Or at least train journalists to understand the topic.

      In Iowa, we have the advantage of lots of data points, which allowed the detection of a late surge for Rubio.

    • Some Body

      Here’s a retry of my comment from yesterday:

      @Amitabh: Going on a tangent, can’t say much about the UK, but in Israel (where I happen to live), the polling this time around was actually quite solid (and certainly no worse than in previous elections), and the hype about the failure of polling is more spin than reality.

      First of all, you should take into account that Israel has a multi-party parliamentary system (and a very fragmented one at that), which makes things substantially more difficult to begin with. Secondly, the latest polls published are in the field at least five days before the election (because of a legal restriction), so late movements, which in a multi-party system tend to happen quite often, cannot be captured.

      One result is that there is practically always a party that registers a surprise gain of up to a dozen seats (of the total 120) not reflected in polls. Moreover, this gain is usually only partially reflected in exit polls. That’s precisely what happened this time around, but of the last eight parliamentary elections in Israel, surprises on the same scale were registered in at least three (2006, 2013, 2015), and if you include somewhat smaller effects—in all eight. (Note that previously, Israel still had effectively two major parties dominating the field, with around 40 seats each, and with no minor party passing into the double digits, which made things easier for pollsters. Since 1996, there was a rapid shift towards the current situation, where there are up to five parties in the 10–30 seat range, and, correspondingly, surprises got more surprising).

      What we had this time around was one party, Likud, successfully “cannibalizing” its own political block to jump from a seat count in the low 20s according to polls to the 30 seats it has now. This was most likely a late movement phenomenon, resulting from strategic voting on the right. The polls were actually right on the spot (no more than 2-3 seats off) in predicting the seat tally for other parties, and even more precise in estimating the size of major political blocks. The only thing they got wrong was the identity of the largest single party in Parliament, which is mostly meaningless in a coalition-based system.

      And now to the spin: this time around, there was a major pre-election media spin, coming especially from the “left” (in reality, Israeli Labour Party is about as left-wing as Donald Trump, but that’s an unrelated topic), that the size of the largest single party matters more than the division of votes between blocks. The rationale behind it was, I guess, that Labour thought they had a chance to beat Likud for the first spot, and then perhaps could use the perceived mandate to join Likud in a coalition with some level of shared power, while they never had any chance to form a coalition outright. Of course, once right block voters realized they might just pull that trick through, they flocked to Likud, and away from other right-wing parties. Eventually, Labour fell victim to their own spin, and were portrayed as hopeless demoralized losers, even as their political block actually gained a few seats relative to the last election. And a collateral victim of all this spin game were the pollsters. In reality, they did quite well, and certainly no worse than in previous rounds.

    • Phoenix Woman

      Amitabh Lath: Ann Selzer didn’t lose anything.

      The discrepancy between her final numbers and the caucus results is easily explained by seeing most of O’Malley’s supporters going to Sanders after O’Malley’s sudden and unexpected decision to quit mid-caucus.

  • pechmerle

    Lot of interesting exit poll figures here:
    http://www.nbcnews.com/politics/2016-election/primaries/IA
    Most notable to me: Clinton’s very large margin with married women. To a lesser extent, married men, which I suspect (from household experience!) is partly an effect of the married women. A distinct contrast with the unmarrieds of both genders, though they are likely also younger, a group with whom Sanders did much better.

  • pechmerle

    He didn’t. You go to the top of this page, and arrow left two posts and there it still is.

  • SocraticGadfly

    You’re still spinning, Sam, you’re still spinning. I guess you and Nate, although heated rivals, have this in common.

    Fox shows Bernie’s support among black voters rose 7 percent the second half of this month. Plus this gives him name recognition and momentum.

    • The Professor

      And you’re making excuses for Bernie, who couldn’t make a strong standing on what was essentially home turf with the demographics in his favor.

    • Billius

      You sound exactly like Romney supporters did on this and 538 in 2012.

    • Josh

      Bernie’s support among black voters is up to 8% now? Hillary better watch her back.

    • Hal9000

      Is this the point where all the black voters in the South jump onto the Sanders bandwagon? Because they’re galvanized by a statistical tie in an all-white state?

    • Phoenix Woman

      You had me at “Fox shows”. Um, not.

  • Anthony

    Is there any evidence that Cruz actually has staying power? He is losing in both NH and SC, and is still far behind Trump in national polling. While trump underperformed slightly tonight we see now that his supporters are real people who will actually show up to vote. From all the evidence I see, Iowa will be Cruz’s high point and will only go down from here. I don’t see a path to his victory at all.

    Too early to tell for Rubio. He overperformed substantially in Iowa and I think NH will be the test to see if he can also overperform in more moderate/establishment states. If he can, I think the Republican nomination will be a two man race between Trump and Rubio. If he doesn’t, I still see Trump walking away with the nomination fairly easily.

    • Josh

      I agree with all of this. The real test of Trump will be in NH, where he generally leads in polling by anywhere from 10-20% with a week to go. If he holds on to win NH–and especially if his winning margin is real and sturdy–I still think he has the best shot of anyone to win the nomination. If someone (Rubio?) catches and overtakes him, he’s in real trouble.

    • 538 Refugee

      It’s possible Cruz ‘punched himself out’. I don’t follow closely enough to really know what his finances and ground game in the rest of the states is like. Did he put everything he had into this effort to look like a winner hoping that would bring him more support? I simply don’t know the answer to this question.

  • Violet

    I agree. The press is spinning a near tie as a disaster for Clinton, but this was Bernie’s demographic, and, other than NH, it can only get worse from here. I’m still betting on Hillary.

    • Kenny

      I’m a Bernie supporter and I think Clinton will likely get the nomination, but I think there are a couple things to consider. One, Bernie outperformed the polls. Two, Bernie has steadily decreased Clinton’s lead over the past several months. Is it possible that he can continue to gain momentum and do better in states where he is polling behind Clinton? I certainly think it’s possible.

    • Sam Wang

      Considering the likely break of O’Malley supporters toward Sanders, I think polls on the Democratic side were off by 1-2 percentage points at most. Also, consider the fact that there was a pool of ~5% who were classified as undecided.

Leave a Comment