April 9, 2016
When we last visited New York, Wisconsin hadn’t voted yet. Our two questions were if Hillary Clinton could keep Bernie Sanders from pulling an upset on her home turf and if Donald Trump could stay over 50% in his domain.
At the time, both were in good shape, but with three weeks before New Yorkers make their final decision, had plenty of time to lose ground. We wondered what difference Clinton and Trump losses in Wisconsin might make.
Three polls have arrived in the past week. The first two, from CBS/YouGov and Monmouth, were taken pre-Wisconsin. The other, from Emerson College, after. We’ll look at these to see if the tide is coming in or going out.
Trump did lose ground. In mid-March, Emerson had him at 64%. Now he’s at 56%. That’s still a big number, but a similar 8 point drop between now and Election Day and he’s below 50%. We know he almost never gets a large percentage of undecided voters. Normally you can figure Trump winds up within a point or two of his final polling average.
A few weeks ago, Kasich and Cruz combined for 13 points (Marco Rubio was still in the race). Now Emerson has them at 39. There’s no chance anyone beats Trump here, but based on this one data set, it’s easy to see him winding up below 50%.
Monmouth has Trump at 52, as does CBS/YouGov. I’d figured we would see something from Siena College by now, they surveyed at the end of January and February. Those polls had him under 50%. The trend line two weeks ago would have put him over the line with them, but we don’t have the data set. I’d expect him to poll under the line with them now. Either way, we wait.
CBS/YouGov is normally high on Trump. Their number adjusts to closer to 48 to 50 percent. If Trump lost any ground in the past week (survey ended 4/1), he’s lower.
This is a long way of saying Trump is right on the 50% line at the moment. If he rallies over the final week, he’ll be ok. If not, expect to see him barely ahead of where Ted Cruz and John Kasich were at home.
Some of this is about momentum and indicating what we might see on April 26. If you figure Trump should do better at home, a 47% finish would increase the odds of a loss in one or more of the five states voting the next week. The other component is delegate math.
If Trump finishes over 50%, he gets the full compliment of 14 statewide delegates. The larger remaining group are apportioned by congressional district. If he finishes over 50% in a district, he gets all 3 delegates. If not, the second place candidate gets one.
In a scenario where Trump is just over 50% statewide, he’ll give up a delegate in several districts. If he recovers back to 55% or so, he should wind up over 50% in the majority of districts. Below 50%, he’ll give up a delegate in many districts and could lose a couple.
The fight for second is important too. Cruz and Kasich would each like to pick up some delegates, but the bigger concern there is making sure Trump loses out on a few. For them it’s about bragging rights and positioning as the anti-Trump choice.
After Wisconsin, Cruz has momentum. If he finishes well ahead of Kasich here, it’s likely April 26 voters flock to Cruz unless they absolutely detest him. On the other hand, Eastern voters are generally more comfortable with Kasich, so if he finishes noticeably ahead of Cruz, they’ll likely move his way, complicating Ted’s efforts to consolidate the #NeverTrump crowd.
If the poor Wisconsin finish hurt Kasich, it’s not visible in the data yet. In mid-March, Emerson had Kasich trailing Cruz by 11. Now he’s 5 back. The two are within a point in the Real Clear Politics average. Pollsters who tend to favor Cruz have him ahead. Those who are traditionally more sympathetic to Kasich have him up.
Leaving Ohio aside, Cruz almost always closes better. The only exception I can think of is Vermont. Normally that isn’t the best example, but it does border Upstate New York. If Kasich is leading in next weekend’s polls, particularly the most publicized ones, by 5% or so, he’ll likely close well. If they’re still tied, or if Cruz is ahead, expect Kasich to lose most of his more strategic voters.
Hillary is still ahead. About a week ago, CBS/YouGov, which normally favors Bernie, gave her a 10 point advantage. Emerson’s new poll has her ahead by 18. All good for Team Clinton, right?
She needs to hope the increased pressure and media scrutiny over the next week or so prove troublesome for her opponent. Here’s why:
During the same time as the New York survey (3/29-4/1), CBS/YouGov polled Wisconsin. They found a 2 point advantage for Bernie. He won by 13. Subtract 11 points from Hillary’s New York margin and she loses by a point. Things rarely calculate out that neatly, but it does indicate she isn’t safe.
The Emerson poll is actually the more dangerous for Clinton. In mid-March, she led by 48 points. When we discussed it 10 days ago, I pointed out that surveys showing this type of Sanders deficit are traditionally very inaccurate. However, in order to have a chance at winning, he’d need to close the gap to 20 to 25 points within 10 days of the vote.
He cut it to 18 with 12 days remaining. I’m not saying they’re tied just yet, but it’s much closer than the polling averages make it look. If I’m right and he has materially closed the gap, Siena should have him within 5 to 7 points when their next survey is released.
Bernie is now close enough that if Hillary wins it’s because she closed better, not because he ran out of time. Stay tuned for the next batch of polls. They’ll include several organizations who haven’t surveyed New York yet, and will cover the Wisconsin result and the impact of a full week of campaigning in the Empire State.