 |
 |
The day of Donald Trump’s acceptance speech is a good time for another poll analysis, I figure. There are 121 new polls added since the previous analysis. A handful of them are newly-discovered older polls, but the majority are newly-released polls. Fifty one of the polls come from a 50-state + D.C. Morning Consult series that was conducted over the preceding three months. The data are newly released, but are not necessarily the most current.
I should mention that only a few polls include surveys conducted this week, and none of them fully cover the G.O.P. convention. So think of this as the pre-conventions baseline analysis. Subsequent analyses will gradually include polls taken after the Republican National Convention. Of course, the same thing will happen following the Democratic National Convention early next month.
As with the previous analysis, a batch of 100,000 simulated elections has Clinton winning 100,000 times and Trump winning 0 times. Clinton received (on average) 338 to Trump’s 200 electoral votes. In an election held now, Clinton would have a 100.0% probability of winning.
Notice, however, that Clinton’s average electoral vote total has dropped a bit from a mean of 354. We should expect at least some improvement for Trump as Republican voters accept and, in many cases, embrace him as the G.O.P. nominee.
There are two striking things about this analysis compared to the previous. First, many of the “outlier” states have “fallen in line” with expectations from recent elections: Nevada has gone from Red to Blue, Arizona from Blue to Red, Colorado from Red to Blue, Kansas from Blue to Red, Missouri from Blue to Red, Mississippi from Blue to Red and Kansas from Blue to red.
The second thing is that many states have become more polarized in the direction expected. You can look at the current map and the previous map to see this easily. For instance, many traditionally blue states have become “bluer”: e.g. OR, WA, MN, WI, MA and NJ. And many traditionally red states have gotten redder: TX, AR, LA, GA, UT, and ND are examples.
In other words, the states are falling into line as we would expect from previous elections. This suggests to me that this will be a conventional election with the battlelines coming down to the battleground states of recent elections. For the moment, Clinton leads in most of those battleground states in the East (PA, VA, NC, FL), Midwest (IA, WI, MI, OH), and West (CO, NM, NV). Of these, IA is very close, but trending toward Trump in the most recent polls, and Florida shows a slight trend in Trump’s direction.
Here is the distribution of electoral votes [FAQ] from the simulations:
Ten most probable electoral vote outcomes for Clinton (full distribution here):
- 344 electoral votes with a 5.99% probability
- 346 electoral votes with a 5.03% probability
- 347 electoral votes with a 4.88% probability
- 341 electoral votes with a 4.87% probability
- 340 electoral votes with a 4.42% probability
- 343 electoral votes with a 4.22% probability
- 338 electoral votes with a 3.51% probability
- 342 electoral votes with a 3.25% probability
- 337 electoral votes with a 3.09% probability
- 339 electoral votes with a 3.07% probability
The single most likely outcome for an election held now (or, let’s say, last weekend) is a Clinton victory with 344 EVs to 194 for Trump. That would consitute a landslide.
After 100,000 simulations:
- Clinton wins 100.0%, Trump wins 0.0%.
- Average (SE) EC votes for Clinton: 337.8 (13.0)
- Average (SE) EC votes for Trump: 200.2 (13.0)
- Median (95% CI) EC votes for Clinton: 340 (309, 362)
- Median (95% CI) EC votes for Trump: 198 (176, 229)
Each column of this table shows the electoral vote total aggregated by different criteria for the probability of winning a state (Safe=100%, Strong=90%+, Leans=60%+, Weak=50%+):
| Threshold | Safe | + Strong | + Leans | + Weak |
|---|---|---|---|---|
| Safe Clinton | 194 | |||
| Strong Clinton | 98 | 292 | ||
| Leans Clinton | 54 | 54 | 346 | |
| Weak Clinton | 0 | 0 | 0 | 346 |
| Weak Trump | 1 | 1 | 1 | 192 |
| Leans Trump | 16 | 16 | 191 | |
| Strong Trump | 41 | 175 | ||
| Safe Trump | 134 |
This table summarizes results by state. Click on the poll count to see the individual polls included for the state.
| 1 | 0 | EC | # | Total | % | % | Clinton | Trump | |
|---|---|---|---|---|---|---|---|---|---|
| 2 | 8 | Votes | polls | Votes | Clinton | Trump | % wins | % wins | |
| AL | 9 | 1 | 1555 | 38.9 | 61.1 | 0.0 | 100.0 | ||
| AK | 3 | 1 | 435 | 37.5 | 62.5 | 0.1 | 99.9 | ||
| AZ | 11 | 5 | 4365 | 47.6 | 52.4 | 1.1 | 98.9 | ||
| AR | 6 | 3 | 2242 | 42.5 | 57.5 | 0.0 | 100.0 | ||
| CA | 55 | 9 | 13744 | 62.0 | 38.0 | 100.0 | 0.0 | ||
| CO | 9 | 5 | 3716 | 54.4 | 45.6 | 100.0 | 0.0 | ||
| CT | 7 | 2 | 1923 | 53.1 | 46.9 | 97.3 | 2.7 | ||
| DE | 3 | 1 | 958 | 53.0 | 47.0 | 90.4 | 9.6 | ||
| DC | 3 | 1 | 1131 | 76.5 | 23.5 | 100.0 | 0.0 | ||
| FL | 29 | 16 | 16838 | 51.5 | 48.5 | 99.8 | 0.2 | ||
| GA | 16 | 5 | 5062 | 49.3 | 50.7 | 24.6 | 75.4 | ||
| HI | 4 | 1 | 801 | 61.9 | 38.1 | 100.0 | 0.0 | ||
| ID | 4 | 3 | 1788 | 40.1 | 59.9 | 0.0 | 100.0 | ||
| IL | 20 | 4 | 5075 | 58.6 | 41.4 | 100.0 | 0.0 | ||
| IN | 11 | 3 | 4118 | 45.3 | 54.7 | 0.0 | 100.0 | ||
| IA | 6 | 8 | 5895 | 50.5 | 49.5 | 70.8 | 29.2 | ||
| KS | 6 | 2 | 1581 | 45.8 | 54.2 | 1.3 | 98.7 | ||
| KY | 8 | 1 | 1825 | 41.4 | 58.6 | 0.0 | 100.0 | ||
| LA | 8 | 2 | 1834 | 39.9 | 60.1 | 0.0 | 100.0 | ||
| ME | 2 | 2 | 1027 | 50.8 | 49.2 | 64.4 | 35.6 | ||
| ME1 | 1 | 1 | 201 | 59.2 | 40.8 | 97.1 | 2.9 | ||
| ME2 | 1 | 1 | 162 | 49.4 | 50.6 | 44.4 | 55.6 | ||
| MD | 10 | 1 | 1289 | 60.9 | 39.1 | 100.0 | 0.0 | ||
| MA | 11 | 2 | 2080 | 60.7 | 39.3 | 100.0 | 0.0 | ||
| MI | 16 | 5 | 3873 | 53.4 | 46.6 | 100.0 | 0.0 | ||
| MN | 10 | 2 | 1803 | 56.7 | 43.3 | 100.0 | 0.0 | ||
| MS | 6 | 1 | 1295 | 39.4 | 60.6 | 0.0 | 100.0 | ||
| MO | 10 | 3 | 3390 | 45.1 | 54.9 | 0.0 | 100.0 | ||
| MT | 3 | 1 | 1153 | 44.1 | 55.9 | 0.1 | 99.9 | ||
| NE | 2 | 1 | 1093 | 42.5 | 57.5 | 0.0 | 100.0 | ||
| NE1 | 1 | 0 | (0) | (100) | |||||
| NE2 | 1 | 0 | (0) | (100) | |||||
| NE3 | 1 | 0 | (0) | (100) | |||||
| NV | 6 | 3 | 1710 | 52.0 | 48.0 | 88.0 | 12.0 | ||
| NH | 4 | 6 | 2855 | 51.3 | 48.7 | 83.7 | 16.3 | ||
| NJ | 14 | 6 | 5285 | 55.8 | 44.2 | 100.0 | 0.0 | ||
| NM | 5 | 2 | 1368 | 53.4 | 46.6 | 95.6 | 4.4 | ||
| NY | 29 | 5 | 7727 | 60.9 | 39.1 | 100.0 | 0.0 | ||
| NC | 15 | 10 | 8188 | 50.6 | 49.4 | 75.7 | 24.3 | ||
| ND | 3 | 1 | 1226 | 44.6 | 55.4 | 0.5 | 99.5 | ||
| OH | 18 | 13 | 11065 | 50.6 | 49.4 | 82.6 | 17.4 | ||
| OK | 7 | 2 | 1253 | 40.1 | 59.9 | 0.0 | 100.0 | ||
| OR | 7 | 4 | 3291 | 55.1 | 44.9 | 100.0 | 0.0 | ||
| PA | 20 | 10 | 11038 | 51.3 | 48.7 | 97.3 | 2.7 | ||
| RI | 4 | 1 | 886 | 57.0 | 43.0 | 99.8 | 0.2 | ||
| SC | 9 | 1 | 1380 | 44.2 | 55.8 | 0.1 | 99.9 | ||
| SD | 3 | 1 | 657 | 40.9 | 59.1 | 0.0 | 100.0 | ||
| TN | 11 | 2 | 2981 | 41.5 | 58.5 | 0.0 | 100.0 | ||
| TX | 38 | 3 | 3706 | 42.3 | 57.7 | 0.0 | 100.0 | ||
| UT | 6 | 5 | 3131 | 45.7 | 54.3 | 0.0 | 100.0 | ||
| VT | 3 | 2 | 914 | 51.1 | 48.9 | 69.1 | 30.9 | ||
| VA | 13 | 7 | 6483 | 53.1 | 46.9 | 100.0 | 0.0 | ||
| WA | 12 | 2 | 2357 | 58.1 | 41.9 | 100.0 | 0.0 | ||
| WV | 5 | 2 | 2232 | 33.9 | 66.1 | 0.0 | 100.0 | ||
| WI | 10 | 7 | 4683 | 53.7 | 46.3 | 100.0 | 0.0 | ||
| WY | 3 | 1 | 690 | 29.6 | 70.4 | 0.0 | 100.0 |
Details of the methods are given in the FAQ.
The most recent analysis in this match-up can be found from this page.
You know what it’s like to step in dog shit and not know it and walk into your house. And suddenly you start smelling dog shit every where and realize what you’ve done and there’s dog shit all over the house. That’s the RepubliKKKlan convention. Dog shit at every turn until you have to puke.
[Deleted — Off topic]
Thanks Darryl! Great to see this.
Heh. Nowhere to go but up for Drumpf. May his numbers stay in the dumps indefinitely.
Trump: 100% probability of whining.
It’s certainly shaping up as a blowout, but every election makes me nervous. Polls can flip overnight. It’s still a long road to November. I’ll breathe a great big sigh of relief when it’s over (assuming Hillary wins). Then I’ll start worrying about 2020. That’s a long way away and I can’t imagine at this point what a post-Trump GOP will look like. My guess is, if he’s crushed, the GOP establishment probably will retake control of the party, or at least its party organs. Even now, the Tea Party seems to be fading.
@5. Assuming Hillary Clinton wins, I am looking to see where the thwarted rage, the need to tear it all down, goes. How will it manifest once the only politician who would listen to THEM is illegally defeated as they see it, by all the brown people, gays, the elites and the takers? What then?
I’ve given up hope on some friends and family. This has been a pretty good show. They’re going to be locked in after tonight.;;; OMG LOOK AT THAT LOGO
Well, Michael Moore says Drumpf is going to win, but I doubt he’s an expert in these matters.
http://www.realclearpolitics.c.....atrick.net
Still, as I posted earlier, elections make me nervous. I think Hillary could lose this thing if Democrats aren’t very careful. We can’t take anything for granted. Every Democrat needs to vote.
@5,
I doubt it will look anything like a blowout in terms of the electoral count. Not likely any kind of landslide. Clinton probably has more than 300 at this point. The 30 vote cusion is hers to lose. But gaining more than those 30 will depend entirely on the Trump campaign, the RNC, and their donor base. Given the mathematical, demographic, and financial deficits that the GOP is starting from, they have a ton of catching up to do in every sense. The trump campaign is still not spending anywhere near what they would be expected to at this point. Not on building org, not on locking down air time, not on data for GOTV. But they certainly don’t start from zero. They probably start from 180 to 200 electoral votes. Do they squander what they already have? Who knows? They’ve been on something of a fuckup streak of late. But I suspect they’ll slowly get their shit together in the next few weeks and start running a professional campaign by September. By the morning after election day it probably won’t look very different from many other recent Presidential elections. But I think the trend lines and a few state flips will tell an interesting story. As will the down ballot races.
Don’t worry. Regardless of the outcome the “conservatives” will continue to claim to be in charge of everything.
And Trump will sue.
His hands really do look small.
Moore’s got a point. I hope sanity prevails in the midwestern states but it could be a battle.
Hillary’s for those trade deals and Drumpf been agin’ ’em from the start.
There are no words to describe how weird this is.
NPR saw a different speech than I did. They are talking message like it meant anything. That was the least enthusiastic, coldest, loveless, mechanical, awkward “celebration” I’ve ever seen.
@11,
Moore has a very good point that given the mood of white working class voters Democrats are more vulnerable in the industrial Midwest than they might otherwise be. But outside of Indiana, the GOP is not currently well positioned. I don’t believe Walker can or will deliver for Trump. The Kochs are going in a different direction. That leaves el Douche and the RNC on their own, behind on money, short on staff and resources, and way behind schedule. To win these otherwise blue states will depend on a very smart, calculated, data driven GOTV effort. Despite Moore’s personal observations of the primary contests, nobody can point to any evidence of that kind of effort underway by Trump or the RNC at this point. Meanwhile, and thanks at least in part to Bernie, Clinton and the DNC are already months into their plan for these states.
And let’s be frank: Moore doesn’t just live in a liberal bubble. He lives in a white liberal bubble. Like many people relying on anecdotal evidence his perspectives are biased by the company he keeps. If W can’t win Michigan, how the fuck does Cheeto Jesus?
Like Moore said, Drumpf said he’d put a tariff on all foreign made vehicles. That was music to Michigan ears.
Dumbya, the lapdog of the corps and the US Chamber, would never do that.
And more voters showed up to vote in Republican primaries in Michigan. Another ominous signal.
@15,
In 2012 native son Mittens loses 44 to 54 and a total vote margin of about 450k. Even W lost by 5 points and the state got a helluva lot bluer in 16 years.
You’re suggesting that tariffs alone can swing 10% of the vote. Maybe I’m a bit more cynical about professional politicians and their backers. But I can’t think of a single one in the last 25 years who wouldn’t eagerly lie about a policy position like that for a ten point swing. Including W and his pals at the US Chamber. Which tells me they know better.
Umm in a year that has had plenty of polls that showed one thing and plenty of primaries that came out very differently there is a lot of evidence this maybe garbage in and your getting garbage out. The fact the model isn’t coming up with at least one win out of 100,000 suggests the model is biased or inaccurate. I just don’t see Trump doing all that worse than Mitt did. He’s more exciting for one thing and Trump nor Pence are Mormons.
Also at this point there are at least two other candidates that you gotta throw into the mix. Jill Stein of the Green Party and that Gary Johnson Libertarian guy. While I don’t expect either of them to get a single electoral candidate I think one or both could have some very good showing in some states. Perhaps enough that it shifts Clinton and Trump position. Maybe Clinton is hurt more than Trump or vice versa. You bet both major parties and the Libertarian’s and Greens are running their numbers will all four. With both minor candidates hoping to at least qualify for Federal election funds, and both could get ten percent or more. They may do even better if they get to be on stage at the debates. Or is Hilliary going to not debate? Your polls and numbers say she should not debate Tump. She should I got this election in the bag and I’m a winner. Neither candidate is under any obligation to debate.
Mark Adams,
First, I invite you to read the FAQ for a better understanding of the analytical methods and results.
“The fact the model isn’t coming up with at least one win out of 100,000 suggests the model is biased or inaccurate.”
The model is biased in one particular way. Ties go to Trump, presuming that the House remains under GOP control. Otherwise the model is completely unbiased is totally symmetrical.
The fact that Clinton wins all 100,000 times is simply a reflection of a poor performance by Trump in state head-to-head polls released to date. BTW: I used a “current poll window” of 3 months for this analysis. If I narrow that window to 2 months, Clinton gets even more Electoral votes. But if I narrow the window to 1 month, Trump wins about 3% of the simulated elections.
What this suggests to me is that 2 and 3 months ago, Trump was doing substantially worse in polls than he has done in the past month. Not particularly surprising given the tumultuous GOP primary season.
In past election cycles, I’ve narrowed the “current poll” window after the conventions, and will do so soon. The problem now is that the rate of polling is rather slow. It will pick up.
“I just don’t see Trump doing all that worse than Mitt did.”
Currently, Trump is doing worse than Mitt. But Trump has had a very bad run of polls from March through June:
Before March Trump occasionally had up to about a 13% chance of winning. (This is from my previous analysis 2 wks ago)
Here is the final time series analysis from 2012:
Mitt never got much above 13% at any point in the last 12 months (full analysis). If we presume that Trump recovers from his summer polling slump, he will be at about the same level as Mitt.
“Also at this point there are at least two other candidates that you gotta throw into the mix. Jill Stein of the Green Party and that Gary Johnson Libertarian guy. “
When races are polled with and without third party candidates, I use the results that include 3rd party candidates. Not much evidence for systematic differences including the 3rd party candidates, though.
“Or is Hilliary going to not debate?”
The debate schedule has been announced. Bucking tradition and refusing to debate would likely have a strong negative effects on a presidential campaign.
@16 It’s very hard to beat an incumbent President, especially one who practically saved the auto industry vs. an idiot who favored allowing the industry to “go through the bankruptcy process”.
And no not just tariffs. How about these loser trade deals? They were as effective in mitigating the freefall of 2008 as Dumbya’s tax cuts.
I hope sanity prevails but it’s on shaky ground this time around.