March 15 is either the day that Hillary Clinton pushes out to an insurmountable delegate lead, or the day that Bernie Sanders confounds the polls a second time. I wouldn’t bet my firstborn on either scenario anymore. But despite Michigan, polls still exist, and we’re still going to look at them. Whether you want to discount them or not is your call, but it’s no longer unreasonable to do so. Here we go! First of, the national picture:
As of last week, there has been precisely two new polls. It’s as if most pollsters said “the hell with it” and quit in protest.
There were four polls this week, all from solid pollsters. All of them showed Clinton with at least 60 percent of the vote. One of them, from Bendixen & Amandi for the Washington Post and Univision is noteworthy because they are one of the two best outfits at polling Latinos anywhere (the other is Latino Decisions). In it, they found that Latinos preferred Clinton by a 68-21. SUSA has Clinton winning Latinos 70-30. You can’t lose Latinos by those margins and expect to win a Democratic primary in Florida.
Illinois (156): Clinton 67, Sanders 25.
Illinois is the state that most closely mirrors our national demographic profile. In fact, that’s a great argument for making Illinois the first-in-the-nation primary state if we really need to make some people’s votes count more than others (and I’m skeptical).
Illinois might share many of the dynamics that drove Sanders’ upset victory in Michigan, but Clinton gets to play native daughter here. Another poll-defying crazy upset victory is not out of the realm of possibility, but certainly not likely.
Missouri (71): No polling
Come on guys, this isn’t a caucus state. What excuse is there for NO ONE polling it?
In any case, a persuasive case could be made as to why Clinton will win here, and same for Sanders. Both campaigns are seriously contesting the state. It’s anyone’s ballgame.
North Carolina (107): Clinton 54, Sanders 34
A 20-point loss would be a victory of sorts for Sanders, given his poor performance in Southern states. On the other hand, a 20-point loss is still a 20-point loss. At least he’s not completely abandoning the state. He needs to close those margins if he wants to escape the evening without an insurmountable delegate deficit.
Ohio (143): Clinton 55, Sanders 40
Ohio is culturally similar to Michigan, but less diverse (Ohio: 81 percent non-Hispanic white, Michigan: 76 percent). No matter what the polling says, I’d venture to guess that this one now leans Ohio Sanders. It’s certainly a must-win state for him, since a clean Clinton sweep of the evening would be beyond devastating, both from a mathematical perspective, but just as importantly, from a narrative perspective. We all saw how important that Michigan win was to change the conversation.
The only poll in the race with recent trendlines, Q-poll, shows modest momentum in Sanders’ direction:
|
CLINTON |
SANDERS |
Quinnipiac |
3/2-7 |
52 |
43 |
2/16-20 |
55 |
40 |
PPP |
3/4-6 |
56 |
35 |
1/12-14
|
53 |
37 |
CNN |
3/2-6 |
63 |
33 |
Given those numbers, it’s hard to give the edge to Sanders, but I’m doing it anyway because 1) Michigan, and 2) he has no choice but to win here.
In total, 792 691 delegates will be allocated Tuesday. Given that Sanders already trails the delegate count 760-546, or 58-42 percent, this day is genuinely a do-or-die night for him.
REPUBLICANS
The national picture:
It’s been a while since I looked at the three GOP buckets. Let’s see if they’ve held up after the recent set of primaries and caucuses:
|
TRUMP |
RELIGIOUS RIGHT |
ESTABLISHMENT |
Undecided |
Mar 10 |
39.9 |
29.7 |
29.2 |
6 |
Feb 23 |
37.7 |
26.5 |
30 |
5.6 |
Jan 28 |
37.4 |
26.5 |
23 |
8.7 |
Jan 21 |
37.1 |
29.1 |
24.3 |
6.8 |
Jan 13 |
37 |
29.6 |
24.6 |
5.4 |
Jan 7 |
36.9 |
29.5 |
25.8 |
3.7 |
Dec 6 |
34.7 |
29.1 |
25.6 |
7.8 |
Nov 4 |
28.4 |
34.1 |
27.3 |
7.4 |
Oct 4 |
28.1 |
27.6 |
30.4 |
6.9 |
Sep 7 |
31.8 |
27.7 |
23.5 |
6.9 |
Aug 3 |
24.5 |
20.6 |
36.7 |
8.4 |
After being stuck in the 37 percent range, Donald Trump finally busted through to nearly 40 percent. The religious right bucket still hasn’t cracked 30 percent except for that brief period where people thought Ben Carson was smart and reasonable. Then they realized he wasn’t, and that was that.
Meanwhile, the establishment bucket is still hovering around 30 percent, all which means that no one has a clear majority hurling toward the convention. But those winner-take-all states are sure looking good for Trump now, since he can scoop up scores of delegates without actually having to win over a majority of his party.
Next Tuesday is certainly looking good for him:
Florida (99 delegates, winner take all): Trump 39, Rubio 26, Cruz 17, Kasich 10
If Marco Rubio dropped out and the establishment vote rallied around John Kasich, there would be a real chance to best Trump. Hence, expect even more pressure on the little one to GTFO. I mean, come on …
Illinois (69, winner take all): Trump 32, Cruz 22, Rubio 22, Kasich 18
I get the value of proportional allocation of delegates, but there is something satisfying about a system in which second-place losers can’t claim victory.
Missouri (52, proportional): No polling
Trump won Michigan easily and is leading Illinois comfortably, so I’m guessing cultural similarities (economically disenfranchised white conservatives) will lead to a similar outcome here.
North Carolina (72, proportional): Trump 33, Cruz 23, Rubio 15, Kasich 9
Ohio (66, winner take all): Kasich 35, Trump 30, Cruz 17, Rubio 7
Most recent Ohio polling has shown Trump actually in the lead, but always within the margin of error. If the polling is correct, this one might be a late night. But what happens if Kasich loses his home state, and Rubio loses Florida? It has to mean that the establishment wing gets bupkus this cycle, right? What does it matter if they rally around this loser or that loser?
Trump’s current delegate lead stands at 464 to Cruz’s 362, Kasich 54. It won’t be anywhere near that close after next Tuesday. Trump needs 1,237 delegates for an outright majority.
UPDATE: Sanders won Michigan in large part by winning independents by landslide numbers (71-28) even as Clinton won self-identified Democrats (58-40). It’s a trend that we’ve seen in most contests thus far. So whether a state has an open or closed primary is actually quite relevant in projecting who has the edge. So for next Tuesday:
Florida: Closed
Illinois: Open
Missouri: Open
North Carolina: Open
Ohio: Open
To be clear, North Carolina and Ohio have “modified” primaries, in which you have to declare party allegiance at the ballot box. That shouldn’t prevent any independent from voting. So Florida is that much tougher for Sanders, but a big chunk of his base shouldn’t be prevented from voting in the other four states.