Wargaming the Electoral College

125 days to go.

270 Electoral College votes to win the White House.

Who wins? Who loses?

I have no idea!

And neither does anybody, really.

But with some careful wargaming of the Electoral College, maybe we can figure out where the action is and exactly what to look for on election night.

Are you ready?

OK, then. First let’s take a look at what the political pros predict, then we’ll strip the map down to the very basics before we build it back up.

Larry Sabato’s Crystal Ball

I bring up Larry Sabato only to dismiss him. It isn’t that he’s necessarily wrong to give Hillary Clinton an Obama 2008-sized blowout; it’s just way too early. Sabato seems to have taken every tossup state, colored it blue, and called it good. Deeply uninteresting stuff.

Charlie Cook isn’t quite so cocky, as you see here.

Cook Political

Me, I’m wary of any map which already shows anybody with 270, and I’m doubly wary of anything showing 300-plus. You should be, too — especially in a year with one candidate who broke all the rules on how to win a primary race, and another candidate who broke all the laws for secure communication of Top Secret information.

For something with a touch more humility, let’s go to Stu Rothenberg & Nathan Gonzales.

Rothenberg & Gonzales

Not just more humble, R&G’s map is more interesting. They’re up to seven tossup states, which tells us that Nathan & Stu understand their limitations better than those other fellas. Still though, to understand the weaknesses and strengths of Clinton and Trump, we’ve got to strip our initial map down to hardcore Reds and Blues.

Which is why on the next page you’ll see a map with more white on it than a Vox.com editorial meeting.

Real Clear Politics: No Leaners

No, the election isn’t anything like this wide open, but a map like this does serve a purpose.

What I did was take the RealClearPolitics poll-averaged map and remove all the leaners, every single one.

What does the result tell us?

Clinton holds insurmountable leads in only 10 states (plus the District of Columbia). Even so, that’s enough to get her more than halfway to 270. 57% to be exact. Trump holds an insurmountable lead in 15 states, but that doesn’t even get him to 33% of the total he’ll need to win.

(NE and ME split their EC votes by congressional district, and one of Nebraska’s CDs only leans for Trump. That’s why there’s a white dot on NE, and why Trump has 83 votes instead of 84. ME will split in other maps below.)

The other, and more important, lesson is that Clinton shows weakness in Blue strongholds of the Northeast and the Industrial Midwest, while Trump is weak in GOP strongholds of the South and the Southwest. Voters in both parties seem to have some reservations about their candidates.

It’s a mixed up, muddled up, shook up map. To get some clarity, let’s add the leaners back in one party at a time.

RealClearPolitics with GOP Leaners

This looks better on the surface for the GOP, but Trump’s weak spots endure even if we assume all the leaners stick with him. A GOP candidate who hasn’t locked up AZ, GA, and OH at this late date? Not good. Still, getting IN, MO, MS, SC, TX, and UT into the Red column shows that hope is far from lost.

Let’s get Hillary’s leaners on there now.

RealClearPolitics: All Leaners

Clinton still maintains her lead, 210 to 164. But that doesn’t mean she can sleep easily at night, hang upside-down in her cave, bathe in the blood of virgins, or whatever it is she does to recoup. If Trump’s troubles in the South & Southwest are bad, Clinton is the only Democrat in recent memory who can’t get a lock on MI & PA and their combined 36 Electoral College votes. The last Republican to win those states was George Bush in 1988, when voters mistakenly believed that they would be getting Ronald Reagan’s third term.

This is where I should mention that no Republican has ever won without OH. But MI would almost make up the loss, and PA would more than make up for it.

If our early maps had too little white on them to be informative, this map has too much. We’ll fix that with the last series of maps, including a couple of victory scenarios.

270toWin Consensus Map

270toWin’s Consensus Map aggregates “the electoral ratings as of late June from Sabato’s Crystal Ball, Cook Political Report, Rothenberg-Gonzales Political Report, NBC Political Unit, The Fix (Washington Post), ABC News, and NPR.” Furthermore, “only states rated safe by all 7 pundits are shown in the darkest shade.” Pundits may be right and they may be wrong, but the Consensus Map may do a better job than any other tool at filtering out the noise.

And yet there’s still enough noise left to keep Clinton from reaching 270 — and a path or two for Trump to get there first.

I have to admit that I wouldn’t make any predictions about CO, FL, or VA, even under the kind of duress involving an ISIS propaganda video crew, my most tender bits, and a finishing hammer. So I’m leaving those states white. But if Trump can hold the rest of the South, shore up AZ & UT, and make a successful play to split the Rust Belt — then Trump won’t need CO, FL, or VA when he takes the classiest oath of office in American History. And if he can take MI, OH, and PA, then CO, FL, and VA would seem likely to follow suit. Trump’s high end then would seem to be around 321. Or 322 if he splits off Maine’s contested congressional district, then up to 329 if Nevada proves to be a red ringer.

Trump Wins

Polling to date doesn’t justify wargaming Trump wins in MN or WI, but winning both would make up for losing in PA. Stay tuned to this space as more numbers become available.

Going back to Sabato’s map, please note that Trump’s best possible score (probably around 329) isn’t that far off from Clinton’s (347).

So if Election Night were tonight, the East Coast early returns would tell us if Hillary has the election as fixed as she has the DOJ.

East Coast Blues

Trump could sweep tossup/blue-leaning CO, IA, MI, MN, NH, OH, WI, and still lose — if Clinton wins the big Eastern Seaboard tossups. So look early to results coming in from FL, NC, PA, and VA. If Clinton wins those, then Trump has to score a series of wins in the seven states just mentioned, plus pick off either NM or NV.

I won’t even try to pick a winner yet, but I can predict with some confidence that we’ll know the winner very early — or very, very late.

ONE LAST THING…

The bettors at Intrade can’t quite seem to make up their minds, either.

(All maps created with 270toWin’s iPad app, which is highly recommended. Perfect for fiddling with on the sofa while you watch the news. If you prefer to do your wargaming in a browser tab, the website is great, too.)