If you have attended an election night watch party chances are that somewhere in the room was a projection of a map made available through the North Carolina State Board of Elections website showing the precinct by precinct break down of results. Whether or not those maps are accurate will likely be determined by the type of machine your county uses to tabulate early votes.
The election maps are generated by an algorithm that takes the precinct-by-precinct results and puts them into a visual format. If however, early voting data is lumped together rather than being broken down by precinct, then the algorithm will not know where to place that data, and those votes will be left off of the map. The votes are still counted towards the total results, but will simply not be reflected in the map of precincts.
How is this happening?
Digital voting machines, such as the iVotronic, had the ability to automatically break early votes down by precinct, while paper ballot machines like the M100 or the DS200 output data slightly differently. The DS200, for example, can be printed either by style (ex. early voting, in-person, or absentee) or by precinct. The M100, however, is incapable of generating a precinct-by-precinct report for early votes.
One example of where this occurred is Robeson County where, in 2016, they used the M100. The 2016 precinct results map is inaccurate as a result. Then, in 2018, Robeson County switched to the DS200, which is capable of breaking votes down by precinct, and as a result the results map accurately reflects the outcome of the election. The difference in maps can be seen below. The map on the left used the M100 in 2016 for the congressional race between Robert Pittenger and Christian Cano, while the one on the right used the DS200 for the congressional race between Mark Harris and Dan McCready in 2018.
Though the two elections had different candidates and different circumstances, the major contributing factor to the apparently dramatic shift in the way precincts voted is the fact that the map on the left does not include any early votes, while the one on the right does. This is a phenomenon that occurs over and over in counties all across the state so The Cycle took the time to figure out just how many counties were impacted. We discovered that 46 of North Carolina’s 100 counties had maps that did not include early voting results in 2018.
In 2020 all counties will be required to use paper ballots for early voting which may have the effect of increasing the number of counties with maps that do not include early voting results. Speaking to The Cycle a Union County Board of Elections official said they are hoping that in transitioning to a paper ballot tabulating machine, supplied by Hart InterCivic, they will continue to have accurate election night maps by opting to print their ballots by precinct.
How to see if this is happening in your county
The issue is not the accuracy of election night results—those are not impacted by this issue— but rather the accuracy of the map that you are watching on election night. This could be more important to campaign officials than it is for casual observers. For a campaign official, it could be very confusing to watch a precinct where you expected to perform well go to your opponent instead. To verify the accuracy of a map you can double click on a precinct to open the graph that breaks the results down by voting method. If the map includes early voting data then it will look like this:
If the map does not include early voting data then in the Absentee One-Stop slot it will show zero votes, as demonstrated below in the 2016 results from Robeson County:
Below is a list of all the North Carolina counties The Cycle was able to identify as not accurately assigning early votes to the appropriate precincts.
Counties whose maps did not include early voting data in 2018