Princeton Election Consortium

A first draft of electoral history. Since 2004

Meta-Margins for control: House D+1.0% Senate R+4.2% Find key elections near you!

Super Tuesday open thread #1 – Hari Seldon edition

March 1st, 2016, 2:00pm by Sam Wang

Follow the evening’s results at the HuffPollster primary page. The calculation below, made in January, still applies under current conditions for today’s election.

For people who watch polls closely, many outcomes are foregone conclusions. We ignore the big wins, and instead look at the close races and the spots of genuine suspense that remain. Sometimes those races are not marquee items (American Samoa, anyone?). But tonight is different for me.

By coincidence, I planned major travel for Tuesday, March 1st. As this post goes live, I am starting a 16-hour-long flight to Hong Kong. My flight is on Cathay Pacific, which appears to lack in-flight wireless. Therefore I will completely miss the evening’s festivities. You’ll have to entertain one another!

This post is on timed release, and is the first of a sequence. I’ll play at being Hari Seldon. Seldon was a fictional figure in Isaac Asimov’s Foundation novels. He had identified laws that predicted the future many centuries in advance. Real-life political forecasting is not at that stage, to put it charitably. However, let’s see if we can manage a lookahead of one day.

I’ve tried to make the comments as well-grounded in polls, delegate rules, and likely reporting times as possible. Of course you are welcome to flay me for any mistakes!


The evening ahead, Republican side: Based on polls, Ted Cruz should finish first in Texas and maybe Arkansas. Marco Rubio has a shot at Minnesota. Donald Trump should come in first everywhere else.

The big caveat is that very little polling data comes after last Thursday’s debate. The last few days have seen some incredible meltdowns: several governors and a senator have endorsed Trump, and Rubio has gone hard after Trump with both substantive attacks and highly personal insults. Evidently Trump’s infection of the rest of the Republican field is complete. If election results deviate from the polls in either direction, these events might help explain why.

Early in the evening, look for East Coast states with larger margins to be reported earlier. Trump’s better performances should be Georgia (+13% over Rubio), Virginia (+14% over Rubio), and Massachusetts (+23% over Rubio). Massachusetts might get called for Trump fairly early.

Cruz’s best state, and most likely win, should be Texas (median of 7 polls: Cruz 33%, Trump 26%, Rubio 17%). Arkansas might be a close win as well (Cruz 27%, followed by Trump and Rubio 23% each, though that survey is 3 weeks old).

Rubio’s strongest states are Alabama, Arkansas, Georgia, Minnesota, and Oklahoma (21-23% each). He might have a shot at coming in first in Minnesota – but not the other states. Tonight is Trump’s night.

Democratic side: Expect some races to be called fairly early for Clinton: Alabama, Arkansas, Georgia, Minnesota, Tennessee, Texas, and Virginia, where she leads by margins of 24-48 percentage points.

Sanders’s one expected blowout is in Vermont, where he leads by 70 percentage points. Polls close at 7 p.m.; expect that race to be called immediately.

Sanders’s other possible wins are, from more likely to less likely, Colorado, Oklahoma, and Massachusetts. If he wins any of these states the margins are likely to be narrow.

Tags: 2016 Election · President

20 Comments so far ↓

  • Josh

    Now that the talk of a brokered Republican convention is heating up, I will be curious as to whether you see any reason after today to change your assessment of Trump’s likelihood of winning on the first ballot. After the first ballot, delegates can change candidates, which I think Rubio is clinging to as his only hope. Cruz is going to stay in for sure, given that he can (probably) claim two wins (IA and TX) to Rubio’s zero. Sounds like Kasich has no intention of leaving before Ohio, so Trump’s chances look better than ever, per your assessment. Safe travels! Thank you for the insights.

    • Matt McIrvin

      Is talk of a brokered Republican convention heating up? It seems to me that if these primaries go as expected, Trump would be on track to get an outright majority easily. A lot of Republicans just don’t like that.

      What I’m hearing is more rumbling about some kind of third-party conservative insurgency.

    • amit

      A brokered convention only happens if no one has more than half the delegates: three or more candidates standing. In this situation Trump will still probably be close to half (say, 45%) and Rubio and Cruz may be closer to a quarter each (say, 25%, 30%). Add Kasich, Carson to taste.

      In the second round delegates may (or may not) change their vote, but Trump needs far fewer defections to him to win than either Rubio or Cruz.

      So unless there is a sound reason to believe the rate of Trump –> Rubio/Cruz delegate defections will be much larger than the rate of Rubio/Cruz –> Trump, there is not much hope for a non-Trump coming out of Cleveland.

    • Josh

      Thanks, Amit

    • Brian

      There is plenty of reason to expect Trump delegates to defect in large numbers. In some states like New York, party bosses can force unfaithful delegates onto candidates. In every state, the RNC members can declare for Trump and displace three of his loyal delegates but refuse to support him at convention (other than the first ballot vote, and that rule can be softened, too).

      Trump really needs about a 55% majority to avoid a brokered convention that will go against him.

    • bks

      55%! The party can only show so much contempt for the rank and file. If Trump has over 40%, there is no way to unseat him without destroying the party.

    • amit

      Brian, while it is possible that state-level Republican officials will install their flunkies and give them secret instructions to defect as soon as legally possible, note that this would mean going against their own self interests for the benefit of the national party. The state level officials depend on these Trump voters for their jobs. They might think twice before doing something that might lose them their jobs to benefit some DC area establishment.

    • 538 Refugee

      Thank you for participating in our ‘opinion poll’/primary. 50.01% said ‘Not Trump’ and we hear you loud and clear. Karl Rove has done the math and said Jeb Bush is the only ‘lock’ to win in November so we will go forward with his nomination. Again, thank you for participating in this wonderfully democratic process.

  • Mark F.

    If things go to a second ballot in Cleveland, I would imagine they might try for a “compromise” candidate rather than go with someone who has less delegates than Trump

    • Olav Grinde

      Like Condoleezza Rice or Colin Powell?
      Or Brian Sandoval?

    • Andrew EC

      I’m buying Paul Ryan on Betfair, particularly at 200:1.

    • Brian

      Romney gives every sign of running for the Republican nomination. He considered running early and deferred to Jeb! Now he’s attacking the frontrunner and refusing to endorse. I expect him to declare around the time of Florida and try to force a brokered convention.

    • Bob

      Romney gives every sign of running for the Republican nomination. He considered running early and deferred to Jeb! Now he’s attacking the frontrunner and refusing to endorse. I expect him to declare around the time of Florida and try to force a brokered convention.

  • amit

    Does this mean Trump = The Mule? You have to admit the shoe fits.

  • Matt McIrvin

    I’ve heard enough Democrats say they actually crossed over to vote for not-Trump that I’m wondering if it will actually have a measurable effect on the primaries. (Probably only if they voted specifically for Rubio, as Peter Beinart pleaded.)

    • Josh

      Matt, regarding your question below about brokered convention– I was going off of articles like this one:
      But maybe you are right– Trump will just get an outright majority. That’s why I asked if Sam has any reason to believe Trump won’t win on first ballot. Seems like if anything, his chances are better now than ever, given that endorsements are now coming in (which implies some superdelegates I assume)…

    • Andrew EC

      Seems like “shenanigans” (crossover votes for strategic purposes) are posited every four years and never amount to anything, so color me skeptical that anyone is following Beinart’s advice in measurable numbers.

    • Matt McIrvin

      @Josh: Thanks! I think political reporters love to speculate about brokered conventions in every cycle because they find the idea exciting, but it never seems to come to anything. This is an interesting year, though, just because the party insiders hate the frontrunner so much, which is the kind of thing that hasn’t happened in a very long time. But I think there would be hell to pay if they actually tried to somehow torpedo someone with a solid delegate majority.

  • SoddingJunkMail

    Sequencing is hurting the Republican efforts to winnow the field.

    Rubio remains their best bet to unseat Trump, but he needs it to be a 1:1 competition.

    Texas coming early like this is going to give Cruz a delegate lead over Rubio, and it’s hard to imagine him dropping out when he’s 2nd in the count.

    If Texas were late in the primary season, the Republican establishment might have been able to convince Cruz to drop out early.

  • Olav Grinde

    American Samoa is missing from The New York Times’ Super-Tuesday live results overview. This is scandalous!

Leave a Comment