Record Voting Despite GOP-Imposed Voter Suppression
Early voting ended Saturday with a record-breaking 3.11 million ballots cast – 113% of 2012’s early voting performance. That puts turnout to date (ballots cast as a percentage of registered voters) at 45%, up from 41% at the end of Early Voting in 2012.
The two graphs above look at the same ballot-casting data from two different perspectives, and illustrate how critical it is to explore different forms of data visualization in order to really understand complicated stories involving big data. The left panel presents the data the way pretty much everyone else in the vote-tracking business always does, plotting cumulative ballot counts versus day and helpfully including a comparison line for 2012.
The right panel plots these same data with just one line, expressing cumulative ballots cast as a percentage of 2012’s same-period performance.
Both plots tell the same top-line story, but the one at left hides a critical insight, while the plot at right makes that insight unmistakeable: during Week 1 of early voting (marked by the horizontal bar) ballot-casting plodded along at a level noticably below 2012’s rate, then suddenly soared to record-setting levels over the first few days of Week 2. What triggered that sudden change? The end of voter suppression tricks built into the state’s first week of early voting by Republican-controlled county boards of elections egged on by the NC-GOP’s executive director, Dallas Woodhouse.
Republicans and Unaffiliated Voters Over-Delivered, While Democrats Lagged Badly
Democrats wrapped up the early voting period with just under 1.3 million votes in the bag, 99% of their 2012 total (falling 19,000 votes shy of 2012). Republicans, on the other hand, improved substantially on their 2012 early voting performance, casting 124,000 extra ballots this year, for a total of 991,000 – 114% of 2012’s total. But the single most important number to pay attention to is 306,000: roughly the margin by which Democrats out-voted Republicans during early voting this year. And it’s bad news for Democrats, because in 2012 their margin over Republicans was almost half again as large (about 449,000 ballots). Democrats head into Election Day this year with a substantially smaller lead than they did in 2012 (an election they lost to Republicans).
Most national news media covering the North Carolina race this year have embarrassed themselves – and mislead their readers – by leading with the headline that Democrats were out-voting Republicans throughout the early voting period. While that’s certainly true, it’s beside the point, because North Carolina Democrats always lead Republicans during early voting (even in elections they ultimately lose). Tar Heel Democrats simply prefer to vote early, while Republicans mostly vote on Election Day. That’s why this year’s magic number – Dems’ current 306,000 ballot margin – is such bad news for the party. State history teaches that, barring some surprise, their thin margin is likely to evaporate in the first warm rays of the Carolina sun when the polls open on the morning of November 8th.
The one wild card that might take both history and us by surprise – although that doesn’t seem likely – is the astonishing numbers those hyperactive unaffiliated voters are putting up this year (UNA in the graphs above). Unaffiliateds cast nearly 810,000 ballots during early voting. That’s a jaw-dropping 142% of their 2012 performance. The extra 239,000 ballots they’ve cast this year might be more than enough to keep Dems on top through Election Day if those unaffiliated voters lean strongly Democratic.
Except they don’t. Our analysis of unaffiliateds’ voting history in this year’s primary leads to the strong conclusion that those unaffiliateds who have voted so far likely split right down the middle in their partisan sympathies: about half lean Republican, while the other half lean Democratic. If that proves correct then those exuberant unaffiliateds have done nothing whatsoever to improve Democrats’ too-narrow lead heading into Election Day.
Democratic Demographics Underperformed, While All Others Shot Upward
Dividing up Republican, Democratic and unaffiliated voters alike along age and gender lines, the only segments that did not better their early voting balloting numbers over 2012 are all Democrats: those aged 18 through 57, men (ending up at 95% of their 2012 numbers), and women – who managed a tie with their 2012 sisters (100%) but no better. As we’ve said before, the performance of Democratic women is actually quite respectable. Democratic registration has declined by nearly 5% since 2012, so Team Blue’s women have achieved a genuine turnout success story…just not enough of one to carry the party across the finish line on their shoulders alone.
White Voters, Mostly Untouched By Natural Disaster And Voter Suppression, Were In The Driver’s Seat
White voters began Early Voting strong and just kept building on that momentum, finishing at 118% of their 2012 performance with 2.2 million ballots cast. Black voters, on the other hand, labored under the twin burdens of flooding in the eastern counties, where they predominate, plus Week 1 voter suppression (drastic reductions in the number of open polling places) that seem to have hit them particularly hard – and certainly not by accident. By the end of Early Voting, African Americans had fought their way back up to 91% of their 2012 performance, casting 690,000 ballots – 68,000 ballots shy of 2012.
We have spoken too little thoughout this series regarding the impressive performance of the ‘Other’ racial category. That’s because the nature of the State Board of Elections data, combined with our desire to keep these graphs as simple as possible, render this category a hodge-podge that is less than highly informative. As used here, the Other category lumps together Asians, Native Americans, voters of two or more races, those who declined to state a race, along with those who declared themselves as ‘other.’ Importantly, this category does not specifically include Latino voters, which the Board of Elections treats as an ethnicity rather than a race, making their voting behavior needlessly difficult to track. In our analysis, Latinos are scattered among all three racial categories we employ.
While voters in our ‘Other’ category posted impressive performance relative to 2012 (finishing at 151%), with 220,000 ballots cast they were the most minor players in our analysis.
Three forces have combined so far in this election to yield a perfect storm against which Democrats must contend: notable Republican enthusiasm for early voting (which Occam’s Razor would suggest reflects equally notable enthusiasm for their ticket), voter suppression baked into the early voting plans of a minority of the state’s 100 counties (including some critical counties such as the state’s largest, Mecklenburg), and the toll of Hurricane Matthew’s devastating floods across the eastern third of the state.
Without the confluence of those forces, the failure of young Democrats to turn out in force would probably be an inconvenience at worst. But the combined toll of all those negatives leave Democrats playing an unenviable game of catch-up as we approach Election Day.
Remarkable Democratic turnout on the 8th could still win the election. But a nail-biter will be the best result the party can hope for on Tuesday night, given the headwinds it has faced this year.
After the close of early voting on Saturday, the State Board of Elections issued a self-congratulatory press release lauding this year’s record-breaking turnout while rather defensively offering that
The vast majority of North Carolina’s counties adopted early voting plans with bipartisan support.
It is true that about two thirds of county boards of elections voted unanimously to adopt their often inadequate plans this year (each county’s board comprises two Republicans and one Democrat). But it is also true that the remaining third – those failing to reach unanimous decisions, thus kicking their contested plans up to the state board for final decision – were a number that has never been higher, making 2016 the most politically hyper-partisan early voting season ever witnessed in the state. Republicans’ successful shuttering of large numbers of polling places for the first full week of early voting this year was a transparent end run around the 4th Circuit Court of Appeals’ ruling overturning the state’s “monster voter suppression law.”
It has been well established that voter fraud is essentially non-existent in America – a figment of certain pols imaginations. We wish the same could be said for state-sponsored voter suppression.