2016 Presidential Election Prediction
And the winner is......
As I write this in the afternoon on the day before the election, we’re just about twenty-nine hours from the point at which we’ll start getting election results in the U.S. Presidential Election, and Americans will begin settling down to the quadrennial ritual of watching the election results pour in and the states start being painted red or blue. Kentucky and Indiana, both of which are expected to be easy states for Republicans, will be among the first states to report, with far more important states such as Virginia, North Carolina, and others closing starting at 7pm Eastern time and then, for the most part, on an hourly basis right up until 11pm when the West Coast finishes voting. By the time it’s all over, and probably some time much earlier than that, we’ll know who the next President will be and, quite possibly, who will control the Senate starting in January. The final numbers are, of course, in the hands of the voters, but as I’ve done in the past here are my projections about what’s likely to happen. These projections are based on the polls, on Electoral College trends and the potential path to victory for both candidates, and on a little bit of guesswork regarding how I think voting is likely to go tomorrow.
Popular Vote:
Hillary Clinton — 49.5%
Donald Trump — 43.3%
Gary Johnson — 5.0%
Dr. Jill Stein — 1.2%
Other — < 1.0%
In both of the past Presidential elections, both the winner and the loser have outperformed the final RealClearPolitics average by several percentage points and I expect that the same thing will happen this time around as well. Much of Clinton’s vote will come from states that she’s guaranteed to win, such as California and New York, so it won’t necessarily help her in the Electoral College, but it will bring her closer to 50% than current numbers are suggesting. Trump will also see his numbers do better than the polls, but the benefit won’t be nearly as big and that will mean that the gap between him and Clinton will be larger than currently projected. Once again to a large degree this will happen because he is likely to receive fewer votes in solid blue, and possibly even some solid red, states than a Republican candidate might otherwise get. This seems as though it will be especially true in states like California where the Republican Party tends to be more moderate than the national party. I’m also predicting that Gary Johnson will end up getting 5% of the vote. This appears to be a larger margin than many pundits are predicting, but it still feels to me as though there is a significant portion of the population that is disgusted enough with the campaign and the two major party candidates to vote for a third-party candidate. In any case, even if he falls short of 5%, Johnson will most assuredly get a higher percentage of the vote, and a larger measure of votes, than any Libertarian Party nominee has ever gotten before. Jill Stein has been falling steadily in the polls and may continue to do so through Election Day, but I’ll peg her final nationwide number at just above one percent mostly because I think that frustration with the two major party candidates will bring some stray voters into her camp.
Electoral College:
When Donald Trump won the Republican nomination, the prospect of a true electoral disaster for the party seemed readily apparent. Many people, even including some Republicans, began predicting a General Election blowout for the Democrats. There was talk of red states turning blue, Electoral College landslides akin to what we saw in 1984 or 1988, and political disaster for the Republican Party. As time has gone on, though, it’s become clear that, while Trump was at a disadvantage in the race against Clinton and her Electoral College advantage, the political divide in the country was strong enough that Clinton would not be able to break through and win in states where Democrats have not won in a long time. As a result, though I’m predicting that Trump will lose, it’s not as bad a loss as may have been anticipated when this campaign started.
Here’s the map:
As I noted when I looked at the Electoral College late last week, Clinton walks into this election with a significant advantage and the data makes clear that she’s likely to build on that advantage as the results come in tomorrow. I expect that she will win North Carolina, Florida, Nevada, and New Hampshire, thanks in no small part to turnout among her supporters. Trump, on the other hand, will win Iowa and Ohio, which is a state that has been hotly contested but which also is well situated to be one where Trump can do well due largely to demography. Additionally, Trump will hold on to three red states that have been heavily contested this year, Georgia, Arizona, and Utah. Obviously, if any of these states are reversed than the totals will change accordingly, but I still expect that, in the end, Hillary Clinton will pull out a victory in the Electoral College just like she’ll win the popular vote.
Update: My predictions for the House and Senate are up.
This looks very possible here. However, it is also possible that the popular vote numbers are the same as this prediction, but both Florida and North Carolina could slip away to Trump by tiny margins. But, I think the popular vote totals would have to be closer for Clinton to lose these states. 49% of the vote compared to 43% should pull along enough states to win in the electoral college.
In 2000, Al Gore proved that he could pull 500,000 more votes than Bush, and lose the election, so it would seem likely that Clinton would not lose if the margin is 49% to 43% here.
Later polls and the FBI cloud removed over the Clinton Emails should give Hillary a tiny bit of late momentum back from the momentum Trump had the last few days. She probably narrowly wins this election somehow.
I think Clinton has a shot at Ohio, but other than that I think you are right on. We shall see. Thank God this is about to be over. Sort of.
I have serious doubts that anything close to 7% of the vote is going to third parties. The third party candidates ran weak campaigns and only ended up in the polling because the pollsters presumed that they might gain some traction due to Trump’s and Clinton’s unpopularity; none of those alternatives are Perot- or Nader-like figures with their own strong bases of support that was built with a hard-fought slog.
I’ll take a guess (and it’s just a guess) that third party votes will be something closer to 4% of the total, which is still higher than usual. Some of those supposed independents will probably spoil their ballots or stay home, with the rest dividing their votes among the two majors. Clinton wins the popular vote in that scenario, but not by much.
Really close to mine, except I have Trump taking New Hampshire.
I ended up 319 to 219 Clinton.
I still do not trust Florida or Wisconsin, hopefully none of my worries matter.
@Paul Hooson: Trump had momentum? I missed that.
@Terrye Cravens: Clinton did have as high as a 9 to 12 point lead in some tracking polls for ABC and other sources, then after FBI director Comey interjected himself into the election, Clinton’s lead had dwindled as low as 1.9% at Real Clear Politics on Saturday, but by Sunday, some flow back to Clinton seems to have taken place, as well as Comey clearing her from an FBI investigation into her Emails.
If i had to put money on a final map, this one looks pretty reasonable. Perhaps it’s optimism on my part, but I expect that to the extent we have any surprises tomorrow night they are more likely to be in Clinton’s favor … OH, AZ. That being said, we should know how things are going pretty early on. If Florida is close, or especially if it looks like Trump will win there, it could be a long nervous night for a lot of people. But if Clinton wins FL before midnight, it may be the sigh of relief heard round the world.
The momentum appears to be with Clinton and the final polls do not fully include the latest update from Mr Comey. 43% for Trump seems about his glass ceiling; that plus expectation that 3rd party candidates vote will be squeezed, so quite possible for Mrs Clinton to break the 50% threshold & thus get up towards a 10% lead (that’s where I put my £2 bet at the bookies last week (@ 9/1)).
Having said that I got Brexit hopelessly wrong!
This is very similar to my prediction. I have the same electoral college distribution, but my breakdown of popular is:
Clinton 49
Trump 46
Johnson 4
Others 1
@Paul Hooson: The fact that Comey’s little stunt hurt Clinton does not mean Trump had momentum. He stayed virtually the same…she fell and then she began to climb again.
Ok, here’s the scary last minute “no toss up state” electoral map from REAL CLEAR POLITICS, just out tonight, with Clinton at a bare 272 to 266 for Trump, with Clinton losing NC, Florida and Ohio, with only NH standing between Trump winning this election, despite Trump down by about 3 points nationally to Clinton in the popular vote estimates.
Apparently, the popular vote distribution is such that while Clinton holds big leads in a few Pacific states like California, Washington, Hawaii and to a lesser degree, Oregon, many other states are very close with Trump in the game with a lot of electoral votes like Ohio and Florida.
NH is all-important according to this REAL CLEAR POLITICS analysis if this is correct here.
@Paul Hooson:
That RealClearPolitics map also shows Nevada as going red. There aren’t a lot of people who believe that outcome at this point. Turning Nevada blue means that Hilary wins without NH.
@Paul Hooson: If Clinton loses FL/NC/OH, she will lose. That being said, I expect her to win FL, NC is leaning her way, but OH will go for Trump, IMO. Trump’s message, if you can call it that, seems to work very well in the rust belt, and OH has too low a percentage of minority voters to overcome his lead among WWC. I think Clinton has a better chance of winning AZ than OH. Overall, I think Doug’s map is good except I believe Clinton will lose the 1 electoral vote in ME.
@Senyordave:
As far as I can tell, Clinton can lose Florida, North Carolina and Ohio, and still win.
On the other hand, Trump needs to win all three of those and more, which is why his situation is hopeless. Nevada alone puts Clinton over the top.
@Pch101: The scary thing is this REAL CLEAR POLITICS estimate has Clinton narrowly losing Nevada. That leaves NH as the kingmaker with a bare 272-266 edge in Clintons favor. That’s too close for comfort if true here.
hillary 50, trump 43.
I don’t see a more hapless than 4 years ago Gary Johnson doing several times better than last time, but then, I’m not a Libertarian hoping that my party/political philosophy will suddenly become credible either. I also don’t see the Greens breaking the 1% barrier either.
@Just ‘nutha ig’rant cracker: Yeah, I see the breakdown falling more heavily for Clinton:
C 51
T 44
J 3
Stein 1 (I can see her edging 1%, but just barely)
other 1
I think people are underestimating Clinton’s support here. She’s going to significantly win the popular vote, and I’m pegging the map above but with AZ flipped, the lone ME vote flipped, and Utah in McMullin’s column. That’s Clinton 333, Trump 199, McMullin 6.
@Paul Hooson:
The only guy who has Trump with 266 is you.
.
We’re counting on EVERY Republican to vote their conscience on November 28th !!!
,
As for me, I’m an optimist, and I’ll go with the LA Times:
Clinton 352
Trump 186
http://www.latimes.com/nation/politics/trailguide/la-na-trailguide-updates-here-s-our-final-electoral-map-of-the-1478473458-htmlstory.html
OH and AZ Go Blue! … OK… a guy can dream, right ??!!?
If Clinton really manages a 6-point lead in the popular vote, I think she’ll do better electorally than you’re suggesting. (Your electoral map is plausible, but only if Clinton’s popular vote lead remains in the 2-4% range the polls are suggesting. If she’s getting close to twice as many votes, it’s going to show up in the swing states.)
@Pch101: I just don’t see large enough numbers of Black voters pulling Clinton through in NC as one problem. I’m hopeful that Hispanics and Jews can pull Clinton through in Florida, which appears to be a little more hopeful. But, Cuban voters tend Republican unlike other Hispanics. The average of polls puts Trump ahead in Florida by a fraction of 1%, making it a true toss-up. Although too close for comfort, the cycle of polls often put Hillary ahead in NH, suggesting she should win there. Ohio looks like a lost cause for Clinton, where she’s probably down by 3 to 4 points there. Nevada is another troublesome state for Clinton, where a CNN poll on the 27th had Trump ahead by 6 points, but two other polls including one by Emerson suggest a 1 or 2 point lead by Clinton, making it less than a sure deal.
Just like the well-qualified Jeb Bush failed due to “Bush fatigue”, Clinton is battling this factor, with her best hope that Trump is viewed as a worse choice to edge him out on election day.
@Paul Hooson:
Nothing gets Trump to 270, which is the only thing that matters.
@Paul Hooson: In that little town of Dixville Notch in NH. Two for her…one for Gary Johnson and one as a write in for Romney. No Trump voter to be found.
@Paul Hooson: The last poll done in OH has Clinton up by one. And in NC she has a early voting lead. No doubt the efforts of the GOP there to suppress that vote is helping them, but I am not sure it will be enough to get Trump elected in that state. She does not need a huge win, just a win.
@Liberal Capitalist: I voted Republican for years…but this time my conscience is telling me not to vote for Trump..he is a fascist. And not a real Republican at all. It shames me to have to say that, but I do believe it.
@Terrye Cravens: Actually, the Dixville Notch vote was Clinton 4, Trump 2, Johnson 1, Mitt Romney (write-in) 1.
Nearby Millsfield went “big” for Trump: Trump 16, Clinton 4, Sanders (write-in) 1.
The RCP maps have been a bit strange for the past month or so–showing PA as a toss-up when she was leading solidly, etc. They’ve also had way more “toss up” states than any other outfit. I’m not sure why, but my guess is that either their methodology is a bit off this cycle, or they’re trying to drive traffic to the site by emphasizing a horse race.
A friend asked this morning over coffee if I was nervous about the election. I told him no, that we were basically screwed no matter who wins. Nothing to be nervous about.
It’s not just a matter of Trump and Clinton that causes that condition. It’s a matter of the people that put them in that position. Even assuming we survive the four years and whoever wins today’s rat race, and assuming that both of these two morons removed from any further governmental imposition by time prime or slime, we still have the people who believed that they would be good leadership material to contend with at the next election.
I do not believe that any people country or culture can survive that level of idiocy for that long.
@Terrye Cravens: If Clinton looks like she’s taking NC and FL early, we’ll know before midnight. Otherwise, it may be a nail-biter. My popular vote predictions are:
Clinton 50
Trump 44
Third parties 8
@Paul Hooson:
This isn’t as accurate as it used to be. The first generation remains staunchly Republican, for reasons that don’t need explaining, but 1) they’re dying out and 2) their kids/grandkids are far less driven by / concerned with policy towards Castro & Cuba. The demographic as a whole has been slowly trending blue for a while now.
Clinton 50
Trump 44
Johnson 3
Stein/Other 1
Dems win the electoral with
Clinton / Kaine 328
Trump/Pence 204
McMullin 6
I think the big imponderable here that polls do not capture is GOTV. In 2012 the polls predicted that Obama was ahead by 1/2 per cent. He in fact won by 4. I believe that the difference was superior GOTV. Now in that race the contest was betweeen great GOTV and good GOTV.
Here the contest is between great GOTV and mediocre GOTV at best. I think that has got to be worth 3 points. This election is very much an experiment as to the efficacy of GOTV, so the political scientists must be salivating to get at the data.
Sadly this is also an experiment into how well Republican voter suppression works ( fvck you very much, Justice Roberts). Hopefully GOTV overcomes this.
@Eric Florack: Even assuming we survive the four years and whoever wins today’s rat race, and assuming that both of these two morons removed from any further governmental imposition by time prime or slime, we still have the people who believed that they would be good leadership material to contend with at the next election.
Total nonsense. You might hate Clinton with a passion, but there is no question she is qualified to be POTUS, both in knowledge and temperament. Trump fails both those tests miserably. He has a laundry list of faults that are documented, and is basically a con man/scam artist. In the debates he was at times reduced to blabbering “call Sean Hannity, call Sean Hannity”. He is cruel, petty and dishonest. Ultimately, the first Clinton ad said it best. “What sort of person man mocks a person’s disability?”
Aside from all this, he was too lazy to even bother learning what he needed to know. Hence, his comment about the “cybers” in one of the debates. He doesn’t understand the whole concept of cyber-security.
@Stonetools:
It was only RCP’s average that underestimated Obama by that much (by 3.2%, specifically). Other aggregators and forecasters were relatively closer: HuffPost by 2.4, Sam Wang by 1.7, and 538 by 1.4.
Ground game can be important (there’s research suggesting it was singlehandedly responsible for Obama’s narrow win in Florida of less than 1 percentage point), but it isn’t the whole story. There are many issues with how good polls are at getting representative samples, figuring out who’s likely to vote, and so on. It’s doubtful ground game can explain as much as a 3% discrepancy between polls and final results.
@Kylopod:
I’m listening to Clintons GOTV director on the Keepin’ It 1600 podcast and he is very happy with the GOTV efforts, especially in Florida and North Carolina. The Clinton campaign GOTV may just have erased GOP voter suppression. Dave Plouffe called in and he sees a 3-7 spread in the popular vote, probably at the high end. I’m going with his prognosis on this. Clinton 51-45 per cent on popular vote. Plouffe sees Clinton surprises in OH and AZ.
At their final campaign rallys last night Clinton had Springsteen and Trump had Ted Nugent.
That’s all you need to know.
(Which one do you think Sad-Sack Chris Christie would rather have been at?)
@C. Clavin: At their final campaign rallys last night Clinton had Springsteen and Trump had Ted Nugent.
And Jon Bon Jovi was the “warm up” for Springsteen!
@C. Clavin: Trumpkins have been saying “Clinton has to have all these big stars to get people to her rallies, Trump does it all by himself!”
Of course that’s only because the only stars he can get are D-lister has-beens.
Being from the Detroit area I grew up listening to Ted Nugent. It makes me sad that he’s turned into a raving right-wing nutjob.
@Mikey:
Sure…I love Great White Buffalo…even though it makes me cringe to think about this raving lunatic whenever I hear it.
That was what? Circa 1987?
Ha…this is awesome…
http://www.thedailybeast.com/articles/2016/11/08/latinos-to-the-rescue-but-what-s-in-it-for-us.html
@Eric Florack:
You are an ignorant racist bastard.
Go back to the slime pit you crawled out of.
Everyone should visit Florack’s website and search the “n-word” and see how many entries you come up with.
@Pch101:
I have to take this one back: Real Clear Politics now has Trump at 266.
(It had him at 241 on Monday, so this is a recent change that I had missed. Mea culpa.)
RCP uses averages, while Wang uses median points. I would give Wang the nod for the better methodology — the median is better at excluding data noise (outliers) than is the average.
It will be interesting to see whether Silver’s extensive tweaking results in more accurate predictions than the others.
My view:
President – I think the Hillary 323, Trump 215 map Doug has is what I am expecting. Popular vote 51-45-3 (Johnson) – >1 (Stein)
Senate — I think Democrats pick up 5 seats (IL, WI, PA, NH and NC, while holding NV) for 51.
House — Democrats pick up 10 to 15 seats, but Republicans still in control.
@HarvardLaw92:
This was my 538 Office Pool take about 4 weeks ago:
Clinton 49
Trump 43
Other 8
Electoral:
Clinton / Kaine 319
Trump/Pence 219
@Pch101:
RCP uses biased polling firms, like Rasmussen, without weighting them.
You really can’t trust their average.
@C. Clavin:
Real Clear Politics uses state polls to forecast electoral votes. Rasmussen doesn’t conduct many of those, so it has little to do with that result.
How I yearn for the days where it all it took was a helmet, a tank and a bad photo op to see who the weaker candidate was.
I predict a Clinton win, but closer to 300 el. votes. And the popular vote.
I only hope it’s decisive enough that the Cheeto-Jesus has no basis for contesting it…other than being a spoiled little man-child born with a silver spoon in his mouth and never having had to suffer through anything of the sort before.
@C. Clavin:
Cheeto-Jesus is already filing lawsuits today.
What a pathetic excuse for a man.
@C. Clavin:
Doh…that suit in Nevada has already been tossed.
Plus Trump got booed when voting.
Not going to be good for him, I think.
@C. Clavin: Nevada political analyst/reporter Jon Ralston’s Twitter feed is fantastic. Apparently the judge in that inane Trump lawsuit was pretty miffed by the idiocy of the whole thing.
https://twitter.com/RalstonReports
Well, you weenied those predictions but were no worse than Karl Rove, Nate Silver, & the rest of you political science hacks. You wrap yourself around statistics and previous history, you can’t see the nose in front of your face when it comes to real life experiences.
But, please do not change your occupation – we have enough incompetence in the ‘real’ work place…
The last laugh is the loudest