2016 Electoral Math -- Two Weeks Out, Clinton's Looking Good

[ Posted Monday, October 24th, 2016 – 18:49 PDT ]

Both Donald Trump and Hillary Clinton had a mixed week in the polls. Some states strengthened for both candidates, and some states weakened. For the most part, though, the race remained essentially unchanged.

I should point out that most of the polling has not yet reflected any shifts in public opinion from the final debate. It takes time for such shifts to show up, so by next week any such change should be a lot more apparent.

As both Clinton and Trump look to shore up their support in the states they think they have a chance in, one thing is becoming clear: Hillary Clinton is free to make a play for states previously in the Republican column, while Donald Trump has almost completely failed in his effort to flip previously-Democratic states. Trump looks like he'll pick up Iowa this time around, and possibly Ohio -- but both of these states have traditionally been swing states, not Democratic locks. Clinton, on the other hand, has locked up Virginia, and is leading in Florida, North Carolina, and Nevada (all previous swing states). But now she's also investing campaign resources in places like Arizona, Utah, Georgia, and even Texas. None of those four have been considered anything but solid Republican in a long time. Clinton looks like she'll have her best shot at Arizona, and possibly Georgia.

The fact she has been able to redraw the map while Trump has only picked up Iowa shows that the basics of the race haven't changed this particular week: Clinton is still the odds-on favorite to win. Let's take a look at the overall totals, which is how the Electoral College would vote if all the polls were correct and the election were held today. The graph measures Electoral Votes (EV), with Clinton's blue starting at the bottom and Trump's red starting from the top. Whichever line crosses the midpoint line will win.

Electoral Math By Percent

[Click on any of theses images to see larger-scale versions.]

Hillary Clinton actually lost a little ground overall this week, as Ohio went to being tied, and then flipped over to Trump's column. She also briefly lost Arizona, but had regained it by the end of the week. Georgia briefly flipped to Clinton, but then flipped back to Trump by the end of the week.

Clinton's margin over Trump fell a little this week, but not all that much. She now has 62 percent of the Electoral College votes, to Trump's 38 percent. Last week, those numbers were 65 percent and 35 percent, respectively. Clinton is doing better now than Barack Obama was doing in 2012, but not as good as Obama did in 2008, to put this into perspective.

Let's take a deeper look at both candidates' relative strengths, using (as always) both the raw data and the category definitions from Electoral-Vote.com. First, a look at Trump.

Trump Electoral Math

[Definition of terms: "Strong" means 10 percent or better in the polls,
"Weak" means five percent or better, and "Barely" is under five percent.

Donald Trump had some good news from three states this week, and some bad news from two others. He also briefly flipped Arizona, but it flipped back to Clinton soon after.

The best news Trump got this week was seeing Ohio flip from Barely Clinton to Tied and then to Barely Trump. This moved 18 EV into his overall total, which improved from 186 EV last week to now stand at 204 EV. The other good news Trump got this week was seeing Indiana and Iowa improve from being Barely Trump to now being Weak Trump.

While all this was going on, however, Utah fell from Strong Trump down to Weak, and Georgia became a true coin-toss, as it moved from Weak Trump to Barely Clinton before flipping back to Barely Trump at the end. This could force the Trump campaign to spend time and resources on states which they really should already have in their pocket by now.

Trump's Strong numbers fell from 87 EV to 81 EV with the loss of Utah, while his Weak numbers improved from 29 EV to 36 EV. In Barely, Trump also improved for the week, from 70 EV up to 87 EV. As always, though, I find it most instructive to track the "Strong Plus Weak" line on the graph, so let's take a look and see how Trump is doing when stacked up against Mitt Romney's 2012 run and John McCain in 2008.

Trump, Romney, McCain -- Strong Plus Weak

This is where you can really see how badly Trump is doing. His Strong Plus Weak number dropped down to a paltry 106 EV during the week, before slightly recovering (as Indiana firmed up) to 117 EV. John McCain, in 2008, ended his race with only 132 EV in Strong Plus Weak, which shows how horrendous this standing is for Trump. At this point in the 2008 race, John McCain had 152 EV in Strong Plus Weak, while Mitt Romney had 180 EV.

All of this points to a historic landslide for Hillary Clinton, so let's take a look at how her numbers are doing this week.

Clinton Electoral Math

Clinton's chart certainly looks a lot healthier than Trump's, that's pretty obvious even at a glimpse. This week Clinton got good news from four states, bad news from another four, and saw three states wobble before returning to where they started.

Arizona wobbled from Barely Clinton to Barely Trump before returning to Barely Clinton at week's end. Florida wobbled the other direction, from Barely Clinton up to Weak Clinton, but then back down to Barely Clinton again. Polling in Florida is constantly going on, so it may continue to wobble in the next two weeks as well, but at least this week it was wobbling in the right direction. Georgia also briefly wobbled into Clinton's column, but then wobbled right back to Trump.

Clinton saw three states firm up from Weak to Strong this week: Minnesota, New Jersey, and Virginia. That last one is the most important, since Trump really needs it to have any decent shot at winning. Trump continues to campaign in Virginia, even though it has quite likely completely slipped from his grasp. Clinton pulled money out of the state a month ago, which is starting to look like a smart move.

Clinton's best news, however, came from another swing state that Trump has had his eyes on, as New Hampshire went from Barely Clinton up to Weak Clinton before finishing in Strong Clinton. The reactions from the first two debates are likely driving this strong movement towards Clinton.

Clinton had bad news from four states, but none of it was really all that much to be concerned about. Michigan and Oregon both slipped from Strong Clinton to only Weak Clinton, but Hillary will quite likely easily win both states in the end. Just as last week ended, one poll put Nevada into the Weak Clinton category, but it flipped right back to Barely Clinton to start this week, which is where it ended up.

The worst news for Clinton was the loss of Ohio from Barely Clinton to Tied and then to Barely Trump. Trump seems to be fireproof in Ohio, as his numbers have actually strengthened here during all the debates. Clinton can afford to lose Ohio and still win the White House, though, while Trump cannot.

Overall, Clinton's numbers slid from 352 EV to 334 EV, with the loss of Ohio's 18 EV. However, Clinton continued to improve her Strong category, adding four states while losing two others. Clinton's Strong number went from 184 EV up to 218 EV, but then fell back to finish the week at 202 EV with the loss of Michigan down to Weak. Once again, Clinton finishes the week with almost the same amount in her Strong column (202 EV) as Trump has in all his categories combined (204 EV). That's a pretty notable comparison.

Clinton's Weak numbers bounced around a bit, from 91 EV at the start up to 104 EV before falling back to 71 EV. While Clinton has been doing a great job of moving states from Weak to Strong, she hasn't been doing as well moving states from Barely into Weak. Clinton's Barely numbers also fluctuated from 77 EV down to 54 EV then back up to finish at 61 EV.

Let's take a look at how Clinton is doing in the Strong Plus Weak category, when measured against Obama's two races.

Clinton versus Obama -- Strong Plus Weak

It's kind of hard to see Clinton's line on the chart this week, because it is so close (273 EV) to the 270 reference line. Clinton continues to do better than Obama was doing in 2012, when he had only 237 EV with two weeks to go. At one point during the week, Clinton even managed to spike up above Obama's 2008 number, when she hit 298 EV (to Obama's 289 EV), but then she slipped back again to slightly trailing Obama's historic 2008 run. Two weeks out, Obama had 289 EV in 2008, which is 16 EV more than Clinton now has. Still, she's running a lot closer to Obama's 2008 line than his 2012 line, and she's got more votes than she needs to win the race in Strong Plus Weak alone. That's a pretty good place to be, with two weeks left in the campaign.


My Picks

As always, my picks are not solely based upon just polling, but also on gut feelings about how the race is going in each state. Last week I was rather optimistic, but most of these calls have been proven to be realistic. I only had to shift a few states around from last week, some of it due to more-recent polling.

So here's my map for this week, as always provided by the folks at 270toWin, where you can easily make your own map if you disagree with any of my calls. Oh, one technical note before we start, I've added a new list of data at the very bottom of the data section (scroll all the way down to see it), where I identify not only the states which haven't had anything more than internet polling, but also the states which have not been polled at all in October, meaning their data might be somewhat stale.

My Picks Map


Likely States -- Clinton

Safe Clinton (21 states, 236 EV)
I moved two states out of the Safe column for Clinton this week, and moved two others in. Polling has weakened somewhat in both Michigan and Pennsylvania, so they both dropped to Probable Clinton for now. Wisconsin, however, strengthened for Clinton, so it moves up from Probable to Safe Clinton. But the biggest news for Clinton here was New Hampshire, which moves all the way up from Leans Clinton to Safe Clinton. New Hampshire was teetering on the edge, earlier in the race, but the last few weeks have pushed it strongly over into Clinton's column.

Probable Clinton (2 states, 36 EV)
Wisconsin had been the only state in Probable Clinton last week, and it moved up to Safe Clinton this time around. Michigan and Pennsylvania moved down to take its place. I was a little too optimistic about these two states last week, it seems.


Likely States -- Trump

Safe Trump (15 states, 87 EV)
No change this week. All of Trump's strongest states remained right where they were this week.

Probable Trump (5 states, 39 EV)
There was a lot of movement both in and out of the Probable Trump category, however. Three states (Indiana, Iowa, and South Carolina) moved up from Leans Trump this week. Indiana and Iowa showed stronger polling for Trump, but I have to admit I'm moving South Carolina on gut instinct alone. There hasn't been any polling from South Carolina for the past month, but I just don't think Clinton has a chance there. If she did, we'd be hearing a lot more about it in the news, and more polling would be happening. So at this point, I have to say it is Probable Trump. But while three states moved into this category, three others moved out. Texas and Utah can no longer be considered even Probable Trump, so they both moved down to Leans Trump this week. Even worse for Trump, Georgia moves all the way down to Too Close To Call. So while Indiana, Iowa, and South Carolina join Alaska and Missouri in Probable Trump, the number of EV in this category fell sharply with the loss of Texas.


Tossup States

Leans Clinton (3 states, 50 EV)
New Hampshire rose out of this category this week (all the way up to Safe Clinton), but the other three states remain unchanged. Polling for Clinton in Florida, Nevada, and North Carolina all remains positive, but not positive enough to say she's locked any of them up yet (although she does seem to be getting close to doing so in Nevada).

Leans Trump (2 states plus one district, 45 EV)
This category turned completely over this week, as all three previous Leans Trump states (Indiana, Iowa, and South Carolina) all moved up to Probable Trump. Moving down to take their place were Texas and Utah. Utah will be a real three-way race, with fringe candidate Evan McMullin polling as high as the two major-party candidates in some polls. It's really anyone's guess what the outcome will be, but for now Trump still seems to be leading both Clinton and McMullin, so it has to be seen as leaning in his direction for the moment. Also, I'm moving that single district in Maine back to Leans Trump, as I dug out what polling exists for the north woods, and it all seemed pretty positive for Trump -- so it can't really be classified as Too Close To Call anymore.

Too Close To Call (3 states, 45 EV)
Arizona and Ohio stayed complete tossups this week, with Arizona flipping for one day to Trump then flipping back to Clinton, while Ohio went from Clinton to tied to Trump. Such frantic movement means it's impossible to predict precisely what will happen two weeks from now. I also had to add Georgia to the Too Close To Call category, as it swung even more between Trump and Clinton (and then back to Trump). While Ohio prides itself on being the swingiest of the swing states, who would have thought we'd be sitting here two weeks out wondering who is going to win Arizona and Georgia?


Final Tally

While a few states moved around within Hillary's Likely category (between Safe and Probable), none of them fell down to only Leans Clinton this week. And she added New Hampshire's 4 EV to the mix as well. Hillary's Safe state total fell from 258 EV down to 236 EV this week, as Michigan and Pennsylvania were replaced by Wisconsin and New Hampshire. Her Probable number jumped from 10 EV up to 36 EV, though, which means that she now has a total of 272 EV comfortably in her corner. Astute observers will have noticed that that is two more votes than she needs to win the race -- without counting any of the states leaning towards her and without any of the tossup states at all.

Donald Trump held onto his core base this week, retaining all 15 states in his Safe category. But the loss of Texas from his Probable category meant a net loss of 34 EV in this category. Utah and Georgia also moved out of being likely states for Trump, while Indiana, Iowa, and South Carolina moved in. But Texas has 38 EV, which left Trump with the same number of Likely states (Safe plus Probable) that he had last week -- 20 in all -- but with only 126 EV between them (down from 160 EV last week). This is less than half of what Clinton can count on, and it is far short of where it needs to be to have any realistic shot at victory. Trump has to pick up an enormous 144 EV from the states which are close to win the race. And that gap has been steadily growing, not shrinking, meaning Trump is moving in the wrong direction.

The gap between the two candidates (in their Likely column) continues to grow, as well. Clinton now has 146 EV more than Trump, when counting states that are reliably going to vote for either candidate. The trendlines are getting worse for Trump, and he's running out of days to do anything about it.

Eight states are in the Tossup grouping this week -- three for Clinton, two for Trump, and three that are truly Too Close To Call. As noted, Clinton already has enough EV in her Likely states to win, but if you add in the states leaning towards her, you get an even more impressive 322 EV victory. This is the same total as last week, in fact. If Trump picks up both states leaning his way, he improves to 171 EV, still almost 100 EV away from victory. Adding in all the Too Close To Call states only gives Trump 216 EV -- also unchanged from last week. What has changed is that even if Trump swept all his Lean states, all the Too Close To Call states, and Clinton's Lean states, he'd only have 266 EV -- four short of the goal.

There's only one thing that could save Trump now, and that is if public opinion shifted massively towards him because of the final debate. This doesn't seem very likely to happen, but you never know. Absent a major post-debate movement towards Trump, though, it looks like this election is pretty much over. Because the margin of victory is still so close for Clinton, though, I still cannot make the claim that she has locked the race up entirely (which, indeed, I was doing for Barack Obama, eight years ago). But she's pretty close to that point, and I really don't see that changing all that much in the next two weeks. The smart money now would be to bet on Clinton to win -- if anyone's even still offering odds on it, at this point.


[Electoral Vote Data:]
(State electoral votes are in parenthesis following each state's name. Washington D.C. is counted as a state, for a total of 51.)

Hillary Clinton Likely Easy Wins -- 23 States -- 272 Electoral Votes:

Safe States -- 21 States -- 236 Electoral Votes
California (55), Colorado (9), Connecticut (7), Delaware (3), Hawaii (4), Illinois (20), Maine (3), Maryland (10), Massachusetts (11), Minnesota (10), New Hampshire (4), New Jersey (14), New Mexico (5), New York (29), Oregon (7), Rhode Island (4), Vermont (3), Virginia (13), Washington (12), Washington D.C. (3), Wisconsin (10)

Probable States -- 2 States -- 36 Electoral Votes
Michigan (16), Pennsylvania (20)


Donald Trump Likely Easy Wins -- 20 States -- 126 Electoral Votes:

Safe States -- 15 States -- 87 Electoral Votes
Alabama (9), Arkansas (6), Idaho (4), Kansas (6), Kentucky (8), Louisiana (8), Mississippi (6), Montana (3), Nebraska (5), North Dakota (3), Oklahoma (7), South Dakota (3), Tennessee (11), West Virginia (5), Wyoming (3)

Probable States -- 5 States -- 39 Electoral Votes
Alaska (3), Indiana (11), Iowa (6), Missouri (10), South Carolina (9)


Tossup States -- 8 States -- 140 Electoral Votes:

Tossup States Leaning Clinton -- 3 States -- 50 Electoral Votes
Florida (29), Nevada (6), North Carolina (15)

Tossup States Leaning Trump -- 2 States (plus one district) -- 45 Electoral Votes
Maine (1), Texas (38), Utah (6)

Too Close To Call -- 3 States -- 45 Electoral Votes
Arizona (11), Georgia (16), Ohio (18)


Polling data weaknesses:

Unlike in 2008 and 2012, polling data does exist for all 51 states (adding in Washington D.C.). What follows are two lists of states: one where the polling data is rather suspect, since the only polls which have been conducted were all conducted only on the internet; and one where no data has been available during October. The internet-only list stayed the same this week, but we're adding the second list of states which have not yet been polled in October, to put their data in the proper perspective.

Internet-only polling (with dates last polled) -- 7 States

Alabama (9/1), Hawaii (9/1), Kentucky (9/1), Mississippi (9/1), South Dakota (9/1), Washington D.C. (5/31), West Virginia (9/1)

No polls since September (with dates last polled) -- 8 States

Arkansas (9/17), Connecticut (9/5), Delaware (9/28), Maryland (9/30), Nebraska (9/27), North Dakota (9/17), South Carolina (9/26), Wyoming (9/11)


-- Chris Weigant


Cross-posted at The Huffington Post

Follow Chris on Twitter: @ChrisWeigant