Can Bernie Sanders win in 2016?

 Yes. It would be hard, but it is not science fiction.

An earlier post showed how many states Bernie Sanders would win in an absolute general election wipeout. In reality, such a wipeout is impossible. Consider the strongest post-1980 Republican candidacies in each major demographic group. In 1984, Ronald Reagan won 66% of the white vote. Robert Dole won 12% of the black vote. (Actually, Ford won 17% in 1976 and Reagan got 14% in 1980, but no Republican has come that close since.) George W. Bush won 44% of the Latino vote and 43% of Asian-Americans. If our hypothetical Rubio campaign managed to match all those numbers against Sanders, he would get only 55% of the vote. If he matched all those numbers and managed to get minority turnout back to 2004 levels he would get 56%. Now that is a big margin, but it is not 59%.

Moreover, no Republican is going to get 44% of the Latino vote in 2016. Maybe if immigration reform has passed, but that ship has sailed. Even Rubio would do well to match John McCain, which would give him an eight-point margin against Sanders; ten points if minority turnout collapses. 

An eight-point margin for the GOP means a ten-point swing compared to 2012, which gets us the map on the right. (Link to 2012 state margins.) This map jibes with what we know from head-to-head polling questions that pit Sanders against Scott Walker in the general election. Those state polls show Sanders losing to Walker in Colorado (8 points), Iowa (8 points), Michigan (4 points), Ohio (10 points), Pennsylvania (5 points), and Virginia (8 points). Sanders was only matched up against Rubio in a poll in Colorado, in which case he lost by 11 points. 

This is the actual floor on a Sanders candidacy. It is a daunting map, and it is one of the reasons why Martin O’Malley would probably pick up support should a tragedy befall Ms. Clinton. But it is not as daunting a map as the one in the previous post. 

And it is even less daunting than it seems if Walker is the GOP candidate. A Walker candidacy would energize black voters. (How do I know this? Well, it is not very scientific; I know a lot of black voters. Walker skeeves and angers them; Jeb! and Marco do not.) Sanders may have some trouble with black voters right now in the primary ... but that would go away against Walker in the general.

So if Sanders can match Obama’s numbers among nonwhites, then Scott Walker wins the national vote by only 52%-48%, even if he pulls in a Reaganesque landslide among white people, a full seven-point swing for the GOP.

Starting off from a four-point deficit instead of a eight-point deficit, Sanders looks even better. Combining exit poll data from here with demographic data from here lets you then compute the following Walker margins for various swing states on the assumption that Walker gets a seven-point swing from 2012, but only among white people:

	GOP	EVs	Sanders EVs
Nevada	48.2%	6	192
Colorado	51.0%	9	201
Florida	51.0%	29	230
Virginia	51.4%	13	243
Pennsylvania	54.2%	20
North Carolina	54.7%	15
Ohio	54.9%	18
Michigan		16
Minnesota		10

The GOP column gives Walker’s starting margin against Sanders. The EV column gives the state’s electoral votes. The Sanders EVs column adds those electoral votes to Sanders’ starting point of 186 if he wins the states, which are ranked from easiest to hardest. If Sanders can replicate Obama’s numbers among nonwhites, then he gets Nevada and comes within spitting distance in Colorado, Florida, and Virginia. 

If he swings CO, FL and VA, then he is two big industrial states away from victory. In that world, it is hard to imagine him losing Minnesota. That puts him one state away. 

And this assumes that white people swing hard right with Bernie on the ballot! Is that likely?

Probably not. Look at this CNN national poll. It does not look like that Walker will get a seven-point uniform swing among white people, let alone recreate Dubya’s performance among Mexican-Americans and Asian-Americans. It is true that state polls show Walker overperforming against Sanders in Colorado and Virginia relative to this analysis, but those polls were taken when Sanders’ name recognition among normal Americans was close to zero. The CNN poll implies that Sanders has since made up that ground.

Bernie Sanders would probably lose to Marco Rubio or Jeb Bush, because those two candidates have a good shot at improving on Mitt Romney’s numbers among Latinos. But Sanders would give them a real run. They would need to work hard to win.

And Bernie could beat Scott Walker. Hell, I would bet on Bernie against Walker!

The path is not trying to recreate Kerry-Edwards. Rather it is an end-run: smash through the states where demography favors the Democrats by demonizing Walker as a closet racist (Walker does himself no favors here) while holding Minnesota and Michigan. That would put him one small state away from becoming the 45th president of the United States. 

One small state. Iowa. New Hampshire. Even Wisconsin, where Walker is not the most popular at home. 

And if he gets Pennsylvania, he wins.

This post, unlike the previous one, has no “counterfactual history” tag. This could happen. Bernie Sanders is no George McGovern. He isn’t even a Walter Mondale.

Sanders is a serious candidate for the 2016 general.

Should you so desire, vote for him in the primary with confidence.

Comments

[James Davis Nicoll]

So, what's the reaction over on Fox if he wins? Mass hysteria? Wasp-killer laced Flavouraid? Mass emigration to Alberta?

[Noel Maurer]

Shareholders celebrate! Four years of an actual socialist in the Oval Office? Ratings would go through the roof! Happy happy days!

Or so I imagine.

[James Davis Nicoll]

People up here stopped freaking out about the NDP taking over Alberta pretty fast but I guess the cultures are a bit different.

[Logan]

New Mexico doesn't stay Blue in a scenario of Bernie replicating Obama's performance among non-white voters? Huh.

[Noel Maurer]

That's my fault, Logan! You're right, New Mexico should be blue.

Obama's margin was 9.9% in 2012. That put it right under the cut-off for states that the Democrats would lose under a 10-point swing. So it didn't belong in the initial map. I didn't recalculate numbers for New Mexico because unlike the other swing states, there were no exit poll data.

But that's an oversight, since there is no way that a seven-point swing among white voter only could cause Sanders to lose the state.

Thank you!

[Noel Maurer]

James: I suspect that the general wonder at the election of a self-proclaimed socialist would last about as long in America as it did in Alberta. But among Fox News producers and the viewers they court? That's a different story! Their outrage could get amped up all over again.

AFAIK, all you guys have is the National Post. Weak tea compared to our right-wing noise machine! Hooah! U-S-A!

[shah8]

cosign on the marked Black antipathy to Walker, such that it's remarkable.

[Noel Maurer]

James: coincidentally, one of the energy writers at Vox went outside his normal field and wrote what seems to be a pretty good piece about the right-wing media in the United States.http://www.vox.com/2015/7/30/9074761/conservative-media-republican-party

I wonder why there's no equivalent in Canada? As I jokingly mentioned, the National Post is really nothing by comparison.

Shah8: It really is remarkable, isn't it? In part, it's about policy. But it's also about the way he relishes the policy. Consider the announcement speech at the below link. Go to minute 13:55. Listen to what he has to say. Then watch what happens at minute 14:03.

http://www.c-span.org/video/?327000-1/governor-scott-walker-rwi-presidential-campaign-announcement

Yes. He really did that.

[Noel Maurer]

Logan: I have an out about New Mexico: the ticket is Walker-Martinez!

The truth is that red New Mexico is an oversight, but thinking about it, Martinez or Sandoval are obvious Veep picks for almost any GOP candidate. Or is that logic mistaken?

[McDevite]

Noel: Not sure Martinez could carry New Mexico in a Presidential year, but who knows.

Anyway, on paper they're good generic VP picks, but neither may clear basic vetting or end up getting asked, and in at least one case, could easily pass on it if offered. And neither of them may end up making sense for Walker.

Martinez has declined in the past, citing her sister's long-term health care needs. That's good cover, but it may also be that she knows she wouldn't clear vet (small state, medium number of scandals and budget issues would suggest that she's--and her team--aren't ready for prime time.)

Sandoval makes sense on paper, but I think he'd pass for two reasons. First and foremost, I think he's better off running for President in his own right as the savior of the GOP, and could do that in attempting to unseat Hillary's second term than risking following Walker. Second, he's close to the line on veto points for: taxes, being pro-choice curious, and mildly gay friendly. So, while he'd come back clean from vetting, he may not want the offer.

Finally, it's been a while since a sitting governor picked another sitting governor at all.

Focusing on the post-reform nominating system, governors tend to pick Senators (sometimes, Congressmen), to cover their flank.

Walker's optimum pick is probably someone like Kelly Ayotte. She's the third amigo of the Graham-McCain gang, she's seen as a GOP moderate, and she gives foreign policy cred to the ticket. The "fuck it, we're gonna win this thing" is probably Deb Fisher or Tim Scott.

[Ken N]

I've actually met Sandoval. I agree with McDevite's analysis and I'm a Democrat. This guy is just biding his time until 2020. He could absolutely win the Presidency. He has a great story, no scandals, and a lot of governing wins.

[Noel Maurer]

These are two extremely informative comments. Thank you!

Sandoval looks like someone to watch very carefully. His state has basically turned blue, as far as I can tell; I'm not sure that it's really swingy anymore. If that's right, it makes the man doubly important to watch.

Hmm. Is Nevada still purple in presidential elections?