What Went Down On Election Night 2017

By my backend math, Northam leads by 5 points in the exit polls. Take that for what it’s worth. Gillespie leads by 8 points among men, but Northam is up by 19 points among women.

Fairly or not, the Virginia gubernatorial race is being set up as a referendum on Trump and Trumpism. It’s a lot of pressure on a single race, especially since the other off-year gubernatorial race today, in New Jersey, doesn’t look likely to be competitive. But even though the stakes of the Virginia race seem unusually high and closely tied to national politics, this isn’t completely new. Viewing these off-year contests as a bellwether for bigger national questions has been pretty standard for decades now. In 2009, observers saw the Virginia race, in particular, as a test of the endurance of Obama’s new Democratic coalition. (Republicans won in both Virginia and New Jersey that year.) In 1997, a debate about federal abortion legislation became a central issue in the Virginia contest. As far back as 1980, a former Republican governor of that state invoked national politics: “Do we want to watch the chief official of our Commonwealth make common cause with Teddy Kennedy or Walter Mondale?” asked former Gov. Mills E. Godwin Jr. In other words, depicting these races in terms of national significance for the competition between parties is hardly unique to 2017. A few things have changed, however. In the past, presidents made stops to campaign: President Clinton went to New Jersey in 1997 to help Jim McGreevey’s race against the incumbent, moderate Republican Christine Todd Whitman, and President Reagan traveled to both states to stump for Republican candidates. Trump, on the other hand, hasn’t been welcome on the campaign trail in either state.

Not to be outdone by LoBiondo, Republican Texas Rep. Ted Poe just announced on Twitter that he is also retiring after this term. Poe’s district, Texas 2, is more Republican-leaning than Lobiondo’s, though. Trump won Texas 2 by 9 percentage points in 2016, after Romney won it by 27 points in 2012. If the 2016 results are truer to how Republican the district leans now, Democrats could conceivably win it in a wave. But it’s pretty red, and will likely remain so under most circumstances.

Take this with a grain of salt, but Democrats have to be loving the turnout reports they’re seeing in Northern Virginia right now. Turnout had already exceeded 2013 totals by 5 p.m. in Fairfax, Alexandria, Charlottesville and several neighboring localities. That’s a pretty extraordinary level of engagement in Democratic-leaning localities. Meanwhile, we haven’t seen similar surges in a sampling of GOP-leaning places, based on reports from Bedford and Tazewell in the Trump-leaning southwest. I know it’s hazardous to divine inferences from turnout reports, but, hey, I grew up in New Jersey and live in Virginia, so I’m feeling loose tonight.

Polls are about to close in Virginia, but before they do, here’s how the state has voted for governor in the past:

Virginia’s past gubernatorial elections

ELECTION YEAR	WINNER	PARTY	WINNING MARGIN
2013	Terry McAuliffe	D	+2.5
2009	Bob McDonnell	R	+17.4
2005	Tim Kaine	D	+5.7
2001	Mark Warner	D	+5.1
1997	Jim Gilmore	R	+13.3
1993	George Allen	R	+17.4
1989	Douglas Wilder	D	+0.4
1985	Gerald Baliles	D	+10.4
1981	Chuck Robb	D	+7.1
1977	John Dalton	R	+12.6
1973	Mills Godwin	R	+1.5
1969	Linwood Holton	R	+7.1
1965	Mills Godwin	D	+10.2
1961	Albertis Harrison	D	+27.7
1957	J. Lindsay Almond	D	+26.7
1953	Thomas Stanley	D	+10.6
1949	John Battle	D	+43.0

If Democrats win the special election for a state Senate seat in suburban Seattle, they will have control of the state House, the state Senate and the governor’s office in Washington and will be able to adopt essentially whatever laws they want. As The New York Times detailed in an excellent piece over the weekend, a win in Washington would mean that Democrats dominate essentially the entire West Coast of the U.S., since the party already has a so-called trifecta (House, Senate, governor’s office) in California and Oregon. Trifectas matter. But they really do in a time when major legislation is often stalled in the other Washington, the nation’s capital. (Republicans ostensibly have a federal-government trifecta in D.C., but they are internally divided on many issues and the U.S. Senate’s filibuster rules give Democrats there more power than the minority has in most state legislatures.) Right now, Republicans have control of the legislature and the governor’s office in 26 states, mostly in the South and West, while Democrats control just six. (Connecticut, Delaware, Hawaii and Rhode Island are the other four controlled by the Democrats.)

State governments don’t just matter in terms of setting policies for states. Increasingly, GOP-controlled state governments are either blocking or overturning policies favored by more liberal cities. And, of course, many state governments are in charge of drawing the district lines for both state and federal races (so they can gerrymander a state’s congressional seats) and setting up the voting procedures for national elections (so states can add barriers to voting or make it easier to cast a ballot). So state government decisions affect national politics. So if Democrat Manka Dhingra wins in the state Senate seat Washington (as expected) and Murphy wins New Jersey’s governor’s race (as expected), Democrats will control the government in eight states. But Republicans would have full control of a 27th state if Ed Gillespie is elected governor of Virginia (he is an underdog) and the Republicans retain control of the House of Delegates in today’s elections, as expected.

Beyond the gubernatorial races and the New York City mayoral election, here are five other races to keep an eye on:

• Atlanta mayor (Polls close at 8 p.m. Eastern): All candidates are running in a non-partisan primary to succeed Democratic Mayor Kasim Reed, who is term-limited. A Landmark Communications survey suggests that no candidate will win a majority of the vote, which means that a Dec. 5 runoff is likely. Right now it looks like Democrat Keisha Lance Bottoms, who is black, and independent Mary Norwood, who is white, will advance to the runoff. Atlanta has had black mayors since the 1970s, and whether it “should” have a black mayor is a question being debated.
• Charlotte (North Carolina) mayor (Polls close at 7:30 p.m.): The incumbent mayor, Jennifer Roberts, was defeated by Vi Lyles in the Democratic primary. Lyles is taking on Republican Kenny Smith in the general election. Although Charlotte is an overwhelmingly Democratic city, a SurveyUSA poll conducted in October had Lyles up just 41 percent to 40 percent.
• Manchester (New Hampshire) mayor (Polls close at 7 p.m.): Admit it — you missed New Hampshire politics! Republican Mayor Ted Gatsas is in major trouble against Democrat Joyce Craig. Craig finished ahead of Gatsas 53 percent to 45 percent in the September non-partisan primary. Today, they face off one-on-one. Since Gatsas was first elected in 2009, this year’s primary was the first time he didn’t finish first.
• Nassau County (New York) executive (Polls close at 9 p.m.): Republican incumbent Ed Mangano is under federal indictment on corruption charges. Not surprisingly, Republican Jack Martins and Democrat Laura Curran have made ethics reform a top priority in their race to win his office. A recent Siena College poll puts Martins ahead, but only barely (43 percent to 41 percent).
• Westchester County (New York) executive (Polls close at 9 p.m.): Republican incumbent Robert Astorino is trying to hold on against Democrat George Latimer. Astorino, who is aiming for a third term, is fighting upstream in a county in which only 31 percent of voters cast a ballot for Trump. An Applied Techonomics survey gives Latimer a 46 percent to 43 percent lead.

Regardless of the results in the New Jersey governor’s race tonight, Democrats already got some good news in the Garden State earlier today. Longtime Republican Rep. Frank LoBiondo, who represents New Jersey’s 2nd Congressional District, reportedly will not run for re-election in 2018. LoBiondo would have been a heavy favorite to win re-election, but if he’s not in the race, his seat should probably be considered a tossup. LoBiondo seemed to have a lock on the district. He had been re-elected easily every time he faced voters since first winning the seat in 1994, never getting less than 57 percent of the vote, and he won by 22 percentage points in 2016. Nevertheless, the 2nd District is best described as purple. Trump won it by just 5 points in 2016, Obama won it by 8 points in 2012, and Cory Booker lost it by 3 points in his 2014 Senate bid. Without an incumbent running, a House race in LoBiondo’s district should be competitive. If Democrats want to take control of the House in 2018, they’ll need to win districts like this one.

Quick point: It’s highly unlikely that last weekend’s Democratic dust-up over Donna Brazile will affect any of tonight’s election returns. That was a story for hardcore politics people, those who are highly likely to vote regardless of what’s going on. More typical voters probably never even knew about it.

More than three and a half years after it began, the Flint water crisis is now on the ballot box in Michigan in the form of a recall vote on the city’s mayor, Karen Weaver. Well, sort of. Ostensibly the recall vote is over a deal that Weaver approved — against the wishes of the city council — with a Detroit-based trash hauling company that’s connected to a federal corruption probe. But the water crisis, and the deep mistrust of elected officials it sowed among city residents, is omnipresent in Flint politics. Seventeen people are running against Weaver in the winner-takes-all election. One is Arthur Woodson, a prominent activist during the water crisis who also organized the recall. Another is Scott Kincaid, who has been a city council member since 1985. The recall itself became mired in scandal after Weaver claimed that 1,200 of the signatures required to initiate the vote were invalid. A judge ultimately ruled there were enough signatures for the recall to go forward. But the water crisis has hung over everything. The crisis began when Flint temporarily switched to using the Flint River as its water source in 2014, but failed to treat the river water appropriately. The corrosive water ate through protective films on the cities pipes, causing lead to leach into the water. Children were subsequently found to have elevated blood lead levels, and an outbreak of Legionnaire’s disease caused at least 12 deaths. Numerous public officials are facing manslaughter charges as a result of the public health disaster. Earlier this year, Weaver said it would likely be 2019 before the water would be safe to drink without the use of a filter.

It’s not quite right to say that this is the first chance to make election predictions since 2016. Before tonight, there were various special elections to Congress — and there were elections in France, Germany and the United Kingdom, along with many other countries. In some of those cases, the conventional wisdom was off-kilter: It overestimated Marine Le Pen’s chances in France, but underestimated the chance that the UK election would result in a hung parliament, for example. Nonetheless, it’s been interesting to see how television pundits adapt to the post-2016 environment. Pretty much everyone on Monday morning’s “Morning Joe” panel predicted that Gillespie would would win in Virginia despite Northam’s modest lead in the polls, for instance. Not that anyone should take it all that seriously — but the segment was a bit worrisome in that it suggests that political pundits and reporters learned the wrong lessons from 2016. Contrary to the conventional wisdom, the polls weren’t that far off last year — they were about as accurate as they’d been in past elections. But they were filtered thru a lens of groupthink that was convinced Trump couldn’t possibly win — and so pundits routinely misinterpreted polls and ignored data showing a competitive race. It’s healthy to take away the lesson from 2016 that polls are not always right. And the uncertainty may be higher in some cases than others. As in last year’s presidential election, for example, there are a large number of undecided in the Virginia governor’s race — which could indicate a polling error in either direction. But that polls aren’t always right doesn’t mean that one’s gut instinct is a better way to forecast elections. On the contrary, the conventional wisdom has usually been much wronger than the polls, so much so that it’s given rise to what I’ve called the First Rule of Polling Errors, which is that polls almost always miss in the opposite direction of what pundits expect. That the “Morning Joe” panel thinks Gillespie will win might be a bullish indicator for Northam, in other words — but we’ll know more in a few hours.

States across the country are holding a bunch of special legislative elections today. In the aggregate, the results will give us another check-in on the national political environment. Two of them, however, are worth paying special attention to: the races to represent Utah’s 3rd Congressional District and Washington state’s Senate District 45. Most of the special elections so far this year have been for state office, so Utah provides us with an opportunity to see whether what we’re seeing on the state level is happening on the federal level. Provo Mayor John Curtis, a Republican, is expected to easily defeat Democrat Kathie Allen in his bid to win the seat left open by Republican Jason Chaffetz. A Dan Jones survey last month put Curtis up 46 percent to 19 percent. It’s not entirely clear what would constitute a good performance for each party given the results of the past two presidential elections in the district. Mitt Romney won Utah’s 3rd District by 59 percentage points in 2012, but Trump — who performed unusually poorly for a Republican in Utah — won it by only 24 points in 2016. Conservative independent Evan McMullin actually beat Hillary Clinton, who finished with just 23 percent of the vote, in the district. If Allen gets more than 30 percent of the vote, that would probably qualify as a fairly strong showing for a Democrat. Meanwhile on the West Coast, Democrats will almost certainly gain control of the Washington state Senate if Democrat Manka Dhingra beats Republican Jinyoung Lee Englund in Senate District 45. Dhingra got 52 percent to Englund’s 41 percent in an August primary, which will basically be re-run today but without independent Parker Harris. A Dhingra victory would mean that every single state legislative body and governorship on the West Coast was controlled by Democrats. It would also cement the West Coast as ground zero for resistance against President Trump, whom local observers believe is a drag on Englund’s campaign.

Gillespie has made no secret that he’s running a campaign strongly inspired by the one Trump used in 2016, employing racial appeals and trying to capitalize on cultural divisions across race and class lines. This seemed to work for Trump last year, particularly in the Upper Midwest, as I noted in this Vox post. Will it work in Virginia? One thing that stands out about Virginia is that Clinton did no worse there in 2016 than Barack Obama did in 2012 (actually, almost a percentage point better), even while she did a bit worse than Obama in most of the rest of the country. This suggests Virginia’s modest Democratic leanings are pretty stable. However, we see interesting variation at the county level. The map below shows the Republican surge from 2012 to 2016 at the county level in Virginia, with redder counties showing a stronger Republican shift. The more rural western counties saw the most Republican shift.

In my analysis, I looked at the shift toward the Republican ticket between 2012 and 2016 across counties based on their level of education, while controlling for racial composition. Clinton did well in high education Virginia counties (defined as those counties where at least 25 percent of the population has a bachelor’s degree) but substantially worse in other counties. The difference in the Republican shift between high and low education counties was large — about 7 points, roughly as large as that in Wisconsin. This suggests that Virginia is ripe for the same sort of appeals to working-class white voters that worked so well for Trump last year. It’s also possible that such appeals are already more or less baked into the electorate in the Southern state. Working-class whites have been voting very conservatively in Virginia for a long time, whereas such leanings were relatively new and exploitable in the Midwest last year.

As I just mentioned, if Democrats gain more than about 10 seats in Virginia’s 100 House of Delegates races tonight, it could be a sign that they’re on track to win a majority in the U.S. House next year. Fortunately, you only need to follow the 25 or so most vulnerable GOP seats to track what’s going on. Here’s a quick follow-at-home guide to the top contests. Democrats’ best pickup opportunities are in Northern Virginia, where Trump ran weakest last fall and two GOP delegates are retiring. But the difference between a good night and a great night for Democrats would be picking up downstate seats in places like Virginia Beach, Newport News and the Richmond suburbs.

Virginia’s Democratic delegate success meter

The 25 most vulnerable Republican seats in the House of Delegates

Pickup opportunities for Democrats
		PAST RESULTS
DISTRICT	REGION	2013 GOVERNOR	2014 SENATE	2016 PRESIDENT
2nd	Prince William County	McAuliffe +11	Warner +6	Clinton +17
31st	Prince William County	McAuliffe +3	Gillespie +2	Clinton +7
32nd	Loudoun County	McAuliffe +7	Warner +2	Clinton +19
42nd	Fairfax County	McAuliffe +6	Warner +2	Clinton +20
67th	Fairfax County	McAuliffe +8	Warner +6	Clinton +22
Toss-ups
		PAST RESULTS
DISTRICT	REGION	2013 GOVERNOR	2014 SENATE	2016 PRESIDENT
12th	Blacksburg	McAuliffe +6	Warner +8	Clinton +2
13th	Prince William County	McAuliffe +1	Gillespie +4	Clinton +14
21st	Virginia Beach	McAuliffe +4	Warner +3	Clinton +4
72nd	Henrico County	Cuccinelli +5	Gillespie +6	Clinton +4
94th	Newport News	McAuliffe +3	Warner +2	Clinton +5
Reaches
		PAST RESULTS
DISTRICT	REGION	2013 GOVERNOR	2014 SENATE	2016 PRESIDENT
10th	Loudoun County	Cuccinelli +3	Gillespie +8	Clinton +4
40th	Fairfax County	Cuccinelli +7	Gillespie +11	Clinton +8
50th	Manassas	Cuccinelli +1	Gillespie +3	Clinton +12
51st	Prince William County	Cuccinelli +1	Gillespie +7	Clinton +6
68th	Chesterfield County	Cuccinelli +2	Gillespie +4	Clinton +10
73rd	Henrico County	Cuccinelli +5	Gillespie +5	Clinton +7
85th	Virginia Beach	Cuccinelli +2	Gillespie +3	Trump +1
100th	Eastern Shore	McAuliffe +2	Warner +1	Clinton +2
Wave territory
		PAST RESULTS
DISTRICT	REGION	2013 GOVERNOR	2014 SENATE	2016 PRESIDENT
26th	Harrisonburg	Cuccinelli +17	Gillespie +20	Trump +9
27th	Chesterfield County	Cuccinelli +8	Gillespie +8	Trump +4
28th	Fredericksburg	Cuccinelli +5	Gillespie +9	Trump +4
33rd	Loudoun County	Cuccinelli +15	Gillespie +20	Trump +16
62nd	Hopewell	Cuccinelli +9	Gillespie +7	Trump +6
83rd	Virginia Beach	Cuccinelli +4	Gillespie +9	Trump +8
84th	Virginia Beach	Cuccinelli +1	Gillespie +5	Trump +5

Past election results courtesy of The Virginia Public Access Project.

Voters in the state of New York today are answering a question they’re only asked every 20 years. Since 1977, New York’s state constitution has required a vote every 20 years on whether a constitutional convention should be held. If a majority of voters are in favor, three delegates from each state Senate district will be selected during the 2018 general election. The selected delegates will then convene in April 2019 to debate amendments to the state constitution. New Yorkers will then get to vote “yea” or “nay” on any adopted amendments. It’s basically a way for voters to bypass the state legislature, which many people disapprove of. And yet … the polls suggest that New Yorkers will vote against holding a convention.

As you can see in the chart above, voter opinion on a convention has swung wildly. Even as late as a month ago, a plurality was in favor of a convention. But the latest Siena College poll has a solid majority, 57 percent, opposing a convention. The shift probably happened in part because of an anti-convention ad campaign led by progressive groups arguing that the gathering could lead to a rollback in protections of workers and immigrants. The New York Civil Liberties Union, for example, released a position paper opposing a convention, saying that the process favors “the interests of the powerful.” Supporters of the convention, though — including the New York City Bar Association and the League of Women Voters of New York State — say it offers an opportunity to strengthen rights protected by the state constitution and reform voting procedures and registration laws. Interestingly, recent polling indicates that no real partisan divide exists among voters. Democrats, independents and Republicans are all strongly against the convention. And the last time a constitutional convention was on the ballot, in 1997, voters rejected it.

Today’s most telling races could be the ones you’ve heard the least about: All 100 seats in Virginia’s House of Delegates are up for election, and the outcomes could tell us more about the political climate heading into the 2018 midterms than any of the headline races. Let’s face it: There isn’t much suspense in the race for New Jersey’s top job. Meanwhile, the Virginia governor’s race is much closer than the state’s anti-Trump mood would suggest, in part because there’s real support for business-minded GOP nominee Ed Gillespie in Northern Virginia that didn’t exist for Trump or 2013 GOP nominee Ken Cuccinelli. But Virginia’s down-ballot races should be driven less by personality and more by voters’ attitudes toward the parties. In 2009, Republicans beat expectations and picked off eight Democrat-held seats (and gained a net of six) in the House of Delegates, foreshadowing their takeover of the U.S. House in 2010. How many seats Democrats pick up tonight could tell us a lot about their chances of retaking the U.S. House in 2018. What’s more, in part because there’s a gubernatorial race on the ballot, today’s turnout is likelier to be more similar to what we’ll see in next year’s midterms than any of the wild special elections that have taken place so far in 2017. Democrats have picked up several very red state legislative seats across the country in special elections with extremely low turnout. And in June, Republicans held onto Georgia’s 6th Congressional District when there was extremely high turnout. No one expects Democrats to pick up control of the House of Delegates; they currently occupy just 34 seats to Republicans’ 66, owing mostly to an aggressively gerrymandered map. But Hillary Clinton won 17 of the 66 House of Delegate districts that are held by Republicans on her way to winning Virginia last year, and Democrats have excellent opportunities in many of them. A few races could be poised to make national news. In the 13th District, based in the rapidly diversifying outer D.C. suburbs, Democrat Danica Roem, a local journalist and transgender woman, is locked in a tight race against GOP Delegate Bob Marshall, who is perhaps the most socially conservative legislator in Virginia and refuses to acknowledge Roem as a woman. And in the 12th District, based in Blacksburg, Democrat and former news anchor Chris Hurst, whose girlfriend, a reporter, was killed on live television in 2015, is running to unseat GOP Delegate Joseph Yost. The bottom line: Republicans would probably consider losing less than five seats a relief. If Democrats pick up between five and 10 seats, it would likely confirm that the U.S. House is in play. If Democrats pick up more than 10 seats, it would be a clue that Democrats are on track to win control of the House.

Although much of the attention this evening will be on the gubernatorial races, there are some cities and counties that are electing mayors and executives. The biggest of those is New York City, where Mayor Bill de Blasio, a Democrat, looks like he’ll easily become the first Democrat to win re-election as the city’s mayor since Ed Koch in 1985. Recent polls have de Blasio leading his Republican opponent, Nicole Malliotakis, by more than 30 percentage points. This wasn’t necessarily predestined to be an easy re-election win for de Blasio. A year ago, his approval ratings were middling, and there were potentially strong challengers ready to run against him. But none of them jumped in. Three key factors kept those challengers on the sidelines:

• Among Democrats, de Blasio’s approval rating stayed above 50 percent, which made a primary challenge tough. And outside of a primary challenge, de Blasio was never in much danger. In New York, where Democrats have a big advantage over Republicans among registered voters, knocking off a Democrat in a general election is tough.
• As The New York Times’ Shane Goldmacher detailed, most strong challengers decided to wait until 2021. De Blasio won’t be able to run again because of term limits, and it’s generally easier to win a nomination when there is no incumbent in the race.
• De Blasio got popular at the right time. As seen in the chart below, de Blasio’s approval ratings among the general electorate climbed over the past year. That likely scared off any strong Democratic, independent or Republican challengers.

One of the major domestic policy initiatives of the Trump administration so far has been to roll back Obamacare, both through executive action and (unsuccessful) legislation. Indeed, many Republicans came out of the 2016 election believing that they had a mandate to dump the Affordable Care Act. Today in Maine, we’ll get an interesting data point as to whether that’s true. Maine is voting on Question 2, which would accept the Affordable Care Act’s Medicaid expansion. This is after Gov. Paul LePage has five times vetoed legislation that would have expanded Medicaid in the state. Opponents of the measure cast it as welfare and fought to have the ballot measure emphasize that point by removing the word “insurance.” For their part, supporters of the measure argue that it would expand health coverage in the state and bring in $500 million in federal funding. Maine’s ballot initiative is the first time that voters anywhere in the country are weighing in directly on expanding Medicaid. And Maine is an interesting test case. In 2016, Maine moved decisively toward Trump and the Republicans: Hillary Clinton won the state by only around 3 percentage points after Barack Obama had won it by more than 15 points in 2012. Some of those voters will now vote directly on a major GOP goal. There is no major top-of-the-ticket race to boost turnout. Instead, turnout will be driven partly by Question 2 and by another ballot measure, on a casino in southern Maine. Some of Maine’s municipalities also have ballot measures to weigh in on, including a proposal in Auburn and Lewiston to merge the two cities and a Portland vote on rent control. But you can bet that the White House, Congress and politicians all over the country will be watching what Maine voters do today.

We’ll get new data in a little bit, but here are the federal special-election results we’ve gotten so far in 2017.

Democrat Phil Murphy is expected to easily defeat Republican Lt. Gov. Kim Guadagno in New Jersey’s gubernatorial race. An average of the last five polls has Murphy leading by 14 percentage points, and no poll has ever shown the race in single digits.

It would be easy to dismiss a win by Murphy as insignificant for the national political environment — a Democrat winning in solid blue New Jersey isn’t exactly surprising. But remember: Current Gov. Chris Christie is a Republican. Were either Christie and/or President Trump more popular, Guadagno might have a shot. Unfortunately for her, NJ-GOV is historically unpopular, and Trump isn’t viewed much more kindly in the Garden State. Indeed, much of the focus will likely be on Virginia tonight, but New Jersey’s election is just as important for understanding which way the political winds are blowing nationally. A big win by Murphy, as the polls project, is consistent with Democrats having the wind at their backs. If he only squeaks out a victory, though — or loses — we’ll have to ask whether New Jersey was a fluke or whether Trump and Republicans are in better shape than we thought. Of course, sometimes surprises do happen. For an early sign of whether Guadagno has a chance, keep an eye on Bergen County in the northeastern corner of the state. During the past three major statewide elections (2013 governor, 2014 Senate and 2016 presidential), the margin in Bergen has consistently been within a few percentage points of the statewide margin.

The marquee matchup today is the gubernatorial race in Virginia. Democrat Ralph Northam is taking on Republican Ed Gillespie. An average of the last 10 polls shows a really tight race, with Northam holding a slight advantage over Gillespie, 48 percent to 45 percent.

As you can see, the polls suggest that the race has tightened down the stretch. The question is whether Gillespie will be able to carry his recent polling gains to victory. One thing to keep in mind as we watch the results roll in is that Northam is likely to do very well in the suburbs of Washington, D.C.

The reason that’s important is because historically that section of the state reports last. Both Gillespie in the 2014 Senate race and Trump in the 2016 presidential race led late into the evening before being overtaken as the Washington suburbs reported their results. It’ll be key to look at past electoral performances in the areas that report first and see if Gillespie or Northam is outperforming his baseline. That’s exactly what we’ll be doing as the results roll in.

Virginia, the state for lovers, is also the state that’s getting most of my attention tonight. Democratic gubernatorial candidate Ralph Northam and Republican Ed Gillespie are locked in a fascinating race that has taken on the broader dynamics of U.S. politics in 2017 in several ways. How things turn out tonight could lend us some insight as we head into the 2018 midterms. Northam faced a surprise primary challenge from Tom Perriello, a former representative in the U.S. House who billed himself as a progressive populist, ready to take on Trump with a heavy dash of economic messaging. Perriello lost, of course, but put forth an interesting strategy of focusing on the state’s rural, whiter areas, where he ended up outperforming Northam. Northam will need to maintain his previously demonstrated strength in Virginia’s urban centers but is also looking to court white voters who might have voted for Obama in the past but cast a ballot for Trump in 2016. Northam’s late-in-the campaign comments that he would support a ban on sanctuary cities in Virginia suggests that he might be looking to strike a more moderate tone for these voters. Democrats planning for 2018 will be sifting through the Virginia results to see whether Northam managed to make inroads with this group. Gillespie, who won the Republican primary by a much slimmer margin than expected, will also need to make a play for white, rural voters — the kind who voted for Trump in Virginia. Gillespie’s narrow victory over Corey Stewart, who ran an anti-immigrant campaign, might explain some of Gillespie’s campaign tone; he has run an ad that features Latino men with facial tattoos and link immigrants to crime. “It feels like my campaign, doesn’t it?” Stewart said to The Washington Post. “I feel vindicated by it. What is it that they say? Imitation is the best form of flattery.” Will the racially tinged Stewart-Trump tactics pay off? Or will they do Gillespie more harm than good by hurting him among Virginia’s many college-educated voters? We’ll get some answers tonight.

It’s Tuesday, Nov. 7, 2017 — election night in America. And FiveThirtyEight will be here all evening to cover the results for you. Our eyes are mostly focused on the gubernatorial races in Virginia and New Jersey. In Virginia, where polls close at 7 p.m., the race has tightened down the stretch run. Most polls still show the Democrat Ralph Northam ahead of Republican Ed Gillespie, but Northam’s lead averaged 3 to 4 percentage points in the final round of surveys — down from a mid-to-high-single-digit lead at earlier points in the race. As our regular readers know, a 3- or 4-point lead is a long way from being safe, so while Northam is the favorite, the outcome is uncertain. A clear Northam win, a narrow Northam win or a narrow Gillespie win are all distinct possibilities based on the polls. A clear win for Gillespie would be more of a surprise — and would leave pollsters with a lot of explaining to do. Polls close at 8 p.m. in New Jersey, where Democrat Phil Murphy will become the state’s next governor, unless there’s a polling error of nearly unprecedented magnitude. Murphy leads by about 14 points in the polling average, and his Republican opponent, Kim Guadagno, has the disadvantage of trying to succeed Chris Christie, New Jersey’s exceptionally unpopular outgoing governor. We’ll also be looking at New York City, where Democrat Bill de Blasio is all but certain to win a second term as mayor — polls close at 9 p.m. — along with a variety of other races, such as a ballot initiative on Medicaid in Maine, a special congressional election in Utah (Republicans are heavily favored to retain the seat), and a state Senate seat in Washington state that will determine whether Democrats take control of the state legislature. Virginia and New Jersey are also holding legislative and other elections. What all of this means for the midterm elections in 2018 is difficult to say. Gubernatorial elections and other statewide elections have only a moderate correlation with congressional races, and the Virginia gubernatorial race in particular has not been all that good a political bellwether. (Democrats won Virginia’s governorship in 2013 before their disastrous 2014 midterm, for instance.) With that said, the political climate is seemingly very good for Democrats, based on measures like the generic congressional ballot, President Trump’s approval rating, and the large number of retirements among Republican members of Congress. While the results of individual races may vary, one would expect Democrats to have at least a pretty good night under those circumstances; to the extent they don’t, it will be reasonable to ask what went wrong.