Let's Not Pretend the Wisconsin Result was Important (Or Why there is no Surge)

Writing from Japan, I feel as serene as Mt. Fuji in the backdrop there (a classic ukiyoe by Hokusai). I remain as assured as I was 2 weeks ago about the Democratic primary. And it has little to do with my own admittedly partisan leanings and more the rather consistent pattern of this primary. I welcome bets from people who have bought into the idea of a hyped Sanders surge, because these are based on gross miscalculations.

Now let me be clear, I am working from one specific assumption well-tested as a general indicator in political science and real results: past voting behavior tends to indicate future voting behavior. That means, and the entire primary has reinforced this over and over again, that you can largely project the remaining primary states based on 2008 results and % urban + % nonwhite.

Why I am I discounting Wisconsin? Because it didn’t surprise me. I’d pegged Sanders at between a 51-56% vote range. He got 56.5%. Obama won the state 58-41 in 2008, and I didn’t expect Hillary to improve much given Wisconsin is , one an open primary state, and two largely white state with an influential tradition of progressive populism and a lot of rural Democrats and Democrats living in mid-sized cities, plus the Madison, which is a hotbed of liberal activism of all sorts. Hillary really didn’t need to do much in Wisconsin; she just needed to win Gwen Moore’s district, and improve a little over 2008. That she did, by 2.42%, from 40.76% to 43.18%. Given that in 2008 Obama and Clinton virtually tied the popular vote if you discount Michigan where Obama wasn’t on the ballot (and you should discount the Michigan results that year for this very reason), a blanket increase of 2.42% would give Hillary Clinton a majority of the primary vote in the Democratic primary and a majority of pledged delegates.

While Hillary’s struggles are even worse this year in caucus states and she has still struggled in open primaries, she has more than made up for these struggles with her strength in closed primary states (Arizona) and semi-closed primaries (Ohio), where the large majority of delegates are awarded. Wisconsin really affirms the simplest model of this primary, where contrary to common belief, there aren’t surges or downturns and momentum means nothing, is just a matter of breaking down the fundamentals of a state based on the last time Hillary Clinton was running for President. Bernie, simply put, does best in four kinds of areas: 1) Where Obama performed well in 2008, and where there are already a natural block of anti-Hillary voters who are easily convinced to vote against her again, 2) Rural areas with heavy % of white voters, 3) Areas with large numbers of conservative white Democrats who can’t vote in the Republican primary, and so vote against the Democratic establishment instead (See Eastern Oklahoma, what will probably happen in West Virginia, parts of North Florida, and so on) and 4) Places where students have an outsized electoral presence, generally smaller mid-sized counties with substantial student populations, like Fayetteville, Arkansas, and Beloit, Wisconsin.

Looking at Wisconsin, I was able to make the following table roughly outlining the principle patterns I’m seeing this primary:

Other places in the Fox Valley played out the same. Neighboring Brown County areas were almost the same as 2008.  39.77% in Outagamie versus 39.62, 39.04 versus 38.48 in Winnebago. Worrisome for Sanders is that even in Dane County his support saw a not insignificant backslide, continuing to add yet another data point that his ability to make up ground on Obama 2008 is mostly constricted to rural white counties and that his vote total declines in big urban centers. Even in the Washington caucuses, where Sanders improved over Obama’s percentage by an impressive 6 points, he under-performed Obama by 5 points in King county (Seattle).

What’s daunting about New York is that Clinton won it 57-40 in 2008, and won all but 1 county, and that was with Obama getting about 90% of the black vote in the state, a demographic that is certain to switch to Clinton, albeit not by the kinds of margins seen in the south. Clinton took over 70% of the vote in many white, rural counties, meaning 12 point collapses in her support, as occurred in Douglas (hit hard by decline in the steel industry and shipping), won’t be enough for Sanders. He’ll need to switch nearly a quarter of the electorate just to tie. What’s more, his continued inability to cut into Clinton’s 2008 support in nearly all the suburban and urban counties bodes poorly, particularly his huge drop from 2008 Obama in Milwaukee county, which, with its large blue collar white, black, and Hispanic populations, comes closest to looking like New York City, only far less unfrienly to Sanders.

So Sanders only hope to keep New York close, much less win it, is to get massive margins in upstate New York, throughout the rural to midsize urban counties of the Hudson River Valley that border Vermont, and in the southern tier, where fracking is a big issue. Winning counties like Oneida, Otsego, Tompkins, Ulster, Dutchess, Chemung and surrounding counties will be essential to him winning. The problem is that even looking at the 4 major counties of upstate New York, Sanders has some pretty big challenges ahead if we look at not just 2008, but every bit of data we’ve gotten from 2016 so far. Even tougher, rural upstate New York was Hillary’s strongest region in 2008 (65% of the vote), yet it’s the region best matched to Sanders strengths, but where he has to go against Hillary’s longstanding popularity in the region, particularly because, over her almost decade as a Senator she was well-known for being much more attentive to upstate than previous New York City-centric Senators had been, and as a resident of northern Westchester county and someone outside of the NYC political machine, she built good rapport with the region, which also includes a lot more moderate voters than elsewhere in the state.

Wisconsin, which according to 2008 CNN exit polls was 87% white, and 83% this year (which given exit polls, have a decent margin of error). New York was 70% white in the 2008 primary exit polls, and almost half of votes were cast in New York City. New York also has a smaller % of the electorate that identifies as very liberal in the 2008 CNN exit polls, with just 22% identifying as “very liberal” and 33% as “moderate” and 36% as “somewhat liberal.” Hillary lost very liberal, as she has reliably in the 2016 primaries, but in 2008 got 62% among self-identified moderates, and “somewhat liberal” primary voters. NYC is 44% white according to the 2010 census, and the closest thing we have to New York City that’s voted already is Boston, which is 52.9% white and has a much smaller Hispanic population, which in New York City is mostly Dominican and Puerto Rican, two demographics that have been among Clinton’s strongest supporters so far. And in Boston, Sanders got 41.7% of the vote after outspending Hillary massively in the Boston media market during the New Hampshire primary, and then further outspending her 3:2 in the run up to the Massachusetts campaign.

The idea, seemingly universally shared by all the Sanders supporters I know that New York is somehow in the bag for Sanders is baffling. Yeah he grew up in Brooklyn, but one, Sanders has been gone from New York since 1968 and when questioned about how the subways worked in a recent editorial, had no idea how they even worked. He’s up against a popular former 2-term Senator from the state who has the backing of the entire, extremely influential, political establishment, even outsider/Cuomo opponent and progressive leader NYC Mayor Bill de Blasio (Clinton’s 2000 Senate campaign manager), who has been a major proponent of Hillary Clinton (www.capitalnewyork.com/...). That’s not all she has:

The former Empire State senator and secretary of state unveiled her New York “Leadership Council” this morning, a collection of the state’s elected officials, activists and donors who have endorsed her for president. The Council will be charged with building a volunteer team and campaign operation to help Ms. Clinton win New York State in the Democratic primary, and boasts many big names, including Gov. Andrew Cuomo, Sen. Charles Schumer, Sen. Kirsten Gillibrand, State Comptroller Thomas DiNapoli, Attorney General Eric Schneiderman and Assembly Speaker Carl Heastie.

DiNapoli and Schneiderman are both staunch progressives and Schneiderman has quite a cult following for his aggressive prosecution of Wall Street excesses and his stalwart liberal credentials are unquestionable. Gillibrand and Schumer remain incredibly popular and influential figures in state politics. In NYC she also has David Dinkins, the first black mayor, campaigning for her, as is the current council speaker, and even other leaders in the NYC communities of color that backed Obama last time, like Congressman Hakeem Jeffries, are backing Clinton this time. Many of the most prominent gay rights activists in NYC are behind the Hillary campaign, and she recently got the backing of District Council 37, the largest union in New York City www.nydailynews.com/…

District Council 37, which includes 121,000 workers across a host of city agencies and 50,000 retirees, is promising to mobilize its members on behalf of the former Secretary of State for the April 19 New York primary.

In a state like New York organization matters a lot, and the Working Families Party really has no impact on the Democratic primary. The WFP was working for Obama and pushing him hard in 2008, along with a host of influential people of color, and he got 40% of the vote. This time Hillary is coming into New York against a less well-organized campaign and stronger backing from unions and communities of color than she had in New York back in 2008. Clinton started her campaign in NYC and it is the headquarters of her campaign, whereas Sanders didn’t have anything resembling a campaign apparatus in the state until the start of 2016. Unlike the last seven or eight states, the Clinton campaign is not going to write off New York or let themselves get outspent by a large margin here. And as Bernie Sanders discovered in Illinois, unions carry a lot more weight in big urban areas than they do anywhere else, and Hillary’s network of union allies in NYC is real and its strong and has proven records of voter mobilization.

Of course, a big segment of New York’s white liberals (Sanders base) work for Wall Street and have seen Sanders campaign spend the entire campaign making inaccurate attacks demeaning Clinton’s campaign from accepting these individual donations, particularly in New York County (Manhattan). NYC accounts for between 50-55% of the state’s electorate. Take Kings County, which cast a little over 12% of all votes in the past primary. Kings County is 42.8% white, and Obama held Clinton to a 1 point win there in 2008. In Milwaukee county, which is 18% more white and gave Obama 62% of the vote in 2008, Sanders lost nearly 15% over Obama's margin. In a similar region, over in Massachusetts, despite losing 6% off her 2008 margin in the state, Hillary actually flipped Suffolk County (Boston), which is again the closest example we have so far to New York City, as the only part of the huge east coast megalopolis that has voted so far. In fact, she gained nearly 10 points in the county over 2008 and her margin there saved the state for her. Between Milwaukee and Boston, and even taking the smaller increase in vote % (in a whiter and less Hispanic county), Hillary getting nearly 60% in Kings County would essentially neutralize Sanders gains in the entire Hudson river valley even if he was jumping 20% over Obama's 2008 performance there.

Outside of NYC and rural upstate, the other two big regions, urban upstate and Long Island, don’t look good for Sanders. Erie and Monroe are two big counties that have substantial nonwhite populations and in the case of Erie, a more moderate Democratic primary electorate. Sanders has really struggled in areas with lots of government workers which also suggest Clinton will hold out well in Albany, and Clinton did so strongly in Niagara it’s unlikely to change calculus very much. What's interesting is the degree to which Sanders will be able to cut into these populous upstate counties at all. Because he hasn’t over-performed Obama in many populous counties in primaries; the Mahoning area in Ohio for sure, some of Massachusetts, and well, that’s about it. Even in Oklahoma where Clinton lost about 13 points compared to 2008, she still flipped Oklahoma county (OKC), which was the only county Obama won in the state, and in Tulsa she only lost four point off her 2008 margins. Past data points don’t suggest its very likely that Sanders can improve substantially on Obama in populous, urban counties that backed Hillary strongly in 2008, because well, he just hasn't done it, he again primarily over-performs Obama in midsize urban areas and rural areas that have high white populations, which will likely mean big wins areas like Oneida, Broome, Ulster, Saratoga, Chemung, Tompkins, and other similar sized counties. I’m unsure how well he’ll do in the northern tier around St. Lawrence, given Hillary took over 70% of the vote there in 2008, but he could well narrowly flip it if he improved by a similar margin as he did in parts of Vermont and rural Oklahoma.

The last piece ofthe puzzle is, of course, Nassau and Suffolk counties that make up Long Island and which both gave Clinton 62% of the vote in 2008. Suffolk is similar to some of the areas of Massachusetts that Clinton held up well in (a Democratic primary electorate that is a mix of nonwhite voters, women, students, blue collar union workers, and rich liberals). While Nassau's large Jewish population might suggest a possible pool of strength for Bernie, he hasn’t actually been doing that well among Jewish voters. At least, Palm Beach County had 256,000 Jewish residents according to a 2005 study, most of whom probably had roots or grew up in the Long Island and the NYC area, and well, Sanders took 27.2% of the vote there, which was actually slightly better than he did in the rest of South Florida. Florida is an interesting canary in the coal mind for several New York demos, aside from white Jewish Democrats, there are also older white democrats (big factor in NYC), and Puerto Rican and Dominican voters, both of which backed Clinton overwhelmingly in Florida, the Puerto Rican demographic giving her upwards of 70% of the vote. Interestingly enough, Nassau and Queens will be the first counties with large Asian American populations, possibly giving us insight into whether that demographic is breaking strongly towards one candidate or the other in advance of California.

Basically, at this point in the race, I can project with high confidence, that Hillary Clinton will win between 56-59% of the vote in New York, and that it’s more likely the percent will be above 59% than it will be below it. The fact that it's a closed primary and Sanders campaign hasn’t been on its game about making sure its supporters know they need to change registration to Democrat and do so before the deadline, bodes poorly for his campaign which has struggled when it has to, you know, win over actual registered Democrats, the partisan, loyal voting base of the Democratic party that often seems incredibly underpresented on the Daily Kos bubble. Even in Michigan Sanders lost self-described Democrats 57-43 and only won because of a 70+ percent showing among non-Democrats.

And that, New York, should remove any lingering questions about whether Sanders has some crazy path to victory in the Democratic primary. (And let’s say Sanders truly gets the upset of the century and wins NY 52-48. That’s not enough. It actually makes the percentage he has to get in future primaries go up even more, to around 58%, and that’s not including places like Maryland, Washington D.C. and Puerto Rico where Hillary is simply 100% certain to rack up big wins, and in fact Sanders may not end up with a single delegate in D.C.’s super late primary, so if it comes down to D.C., Hillary will win it, but this is again, just in the astonishing string of upsets and strong performances that are unlikely scenario). And none of this is about me hating the primary, hating the campaigning, fundraising, the great work being done to keep democratic issues in the spotlight. My acerbic tone is more a response to others tones, and also the way Sanders supporters are always talking like each state they win is a shocking surprise. When Hillary won Arizona, or North Carolina I wasn’t beside myself like “Oh my, What a surprise. Look at how we’re surging!” In fact, the very model in here that I used for Wisconsin and am using for New York accurately projected her % range in Arizona as well. It gets tiresome, a month and a half of seeing Sanders supporters always breathlessly talking about how the surge is coming (“trust me, the surge is coming, it’s coming alright, the surge is coming, and it’s going to be great, but first, let me tell you about the top 1%) when there’s been no such real movement and Sanders fast growth has been pre-defined from the very beginning, while Hillary’s core group of support has largely been crystallized and unmoved since January if not before. Like the smarmy, complicated, and factually inaccurate takes that Seth Abramson seems to make for HuffPo on the daily (which then get disseminated here) that try to cherrypick polls, compare polls from different pollsters to create trendlines, mis-relay exit poll data outside of important contexts, and lately, talk about early voting versus election day voting like there’s some essential surge its covering up (never mind that 70% of Arizona voters vote early, and the large majority in the ten days prior to the election, not back in February when early voting, and those that did vote then were mostly old people and partisans on both sides). New York, is, in a sense, the newest and most important test for my assumptions which explain all the variation in the primary so far, and also explain why it’s been unwinnable for Sanders for quite sometime barring an earth-shattering revelation or fuck up on Clinton’s part. I find factually deceptive or arguments that fail a number of basic smell tests (including Occam’s razor), irritating, as its quite possible to back Sanders without presenting shoddy arguments for why he's definitely going to win.

Anyway, thanks for reading.