(MAP) 2018 Primary Electorate by County (COMPLETE*)

[Adam Griffin]

A work in progress: just like I did this for the 2016 presidential primary, I'm going to compile a county-by-county map for 2018. For now, though, here's a partial map + the statewide two-way results:

Exceptions In Results Calculations:

NY & UT: no county-level data and/or extensive primary on GOP & DEM sides, respectively. Figures are based on most recent active voter registration statistics by county

SD: No contested DEM primary statewide or in most counties; statewide figures based on GOP voter turnout versus DEM & NPA turnout (NPA voters were only eligible to pull a DEM primary ballot, making GOP vs DEM/NPA voters the most realistic comparison)

VA: No DEM primary statewide and no local elections; projection made by analyzing contested primaries in both parties for applicable 7 CDs and then comparing to statewide share of both Trump and Clinton electorate in these 7 CDs; applied comparable swing to remainder of electorate for a projection of how electorate would have voted assuming contested primaries statewide for both parties

[Adam Griffin]

Added 5 states' county results to the map.

[Adam Griffin]

IL & ID added.

[Adam Griffin]

What are you using as your data source? I'm making a similar map myself that averages the statewide contests (expect North Carolina) totals, R vs D. I can see at least one different county from what you have.

Complete NYT results in situations where possible (i.e. top-ballot contests where all candidates are shown) and state election data (whether it's top-ballot results or actual primary turnout figures; some don't offer the latter) in situations where NYT truncates the figures to the top three candidates. It's entirely possible that really close results could be diff depending on the exact ballot counted, and just as possible I flipped a result by accident.

[Adam Griffin]

Updated (sans NC/KY).

[Adam Griffin]

Added several new states. Waiting on CA to fully finish and SD's results are weird because of the Democratic/Non-Partisan situation, but I'll get them added in the near future.

[Adam Griffin]

Updated with last night's completed results.

As of now, 26 states have voted, with Democrats winning 13 and Republicans winning 13. Kentucky is almost certainly majority-D by this measurement (though we won't have breakdowns apparently for another 2 months).

And as far as Virginia's figures go, it's a projection built on last night's partial results and 2016 results. This explains how the figure was calculated.

[Adam Griffin]

With 90% of Maine's precincts now reporting, I've went ahead and hastily added the results to the map. There are many counties where the margins are close; it wouldn't surprise me if there are some D counties that ultimately go R and vice-versa. Once the results are complete, I'll update the map.

[Adam Griffin]

What's up with that  uber red Texas county just southeast of the panhandle? Thats a super Republican area.

Jimrtex has the explanation for it somewhere. Until recently, the parties held the primaries on their own behalf. If I recall correctly, now, it's because whoever holds the county elections office can (or could) choose which party primaries get to happen (because they're in charge of them).

EDIT: here's the explanation.

[Adam Griffin]

OK and CO county-level results added; MD statewide figures added.

[Adam Griffin]

And just for fun, a swing map between 2016's primary electorate and 2018's. Granted, it's not necessarily an apples-to-apples comparison in several states (primary versus caucus, for example), but not completely futile for a large number of states:

[Adam Griffin]

Dems definitely lost ME-02, bodes poorly for Golden.

Eh, it's a lot closer than you might think. Adding together all the votes from all the whole counties in ME-2 gets you to around 45k votes for each party (well, at least in the GOP-GOV and DEM-HOUSE contests; haven't done an apples-to-apples comparison).

Several of those ME-2 counties that leaned in favor of the GOP were tiny victories. I still don't think the final numbers are available by county yet (from the looks of ME's crappy website)

And why can't they publish the raw turnout figures for CDs where incumbents are uncontested? This is so weird to me...not having candidates on the ballot when there are party primaries, just because those candidates are not being challenged.

[Adam Griffin]

With ~99% of California's votes finally counted, I've added the results to the county map (+ updated the statewide figures, which shifted considerably).

[Adam Griffin]

How did Democrats get a large share of the primary vote in Oklahoma?

Same reason as states like WV/KY/NC (closed primary systems; lots of ancestral Democrats in rural areas - in this case, Little Dixie - remaining local dominance of Democrats and a general laziness to update voter registration).

[Adam Griffin]

Not sure why I didn't add MD yesterday, but it's added now.

[Adam Griffin]

Bumping; partial results of TN now available.

[Adam Griffin]

MO & KS mostly complete; added.

[Adam Griffin]

Washington's tentative results added (along with all other counties reporting thus far from all of tonight's contests).

[Adam Griffin]

All current results added (HI/MN/WI/CT/VT).

[Adam Griffin]

AK/WY figures added.

[Adam Griffin]

Went ahead and did AK's breakdowns by HD: can find it in the OP and here.

[Adam Griffin]

Went ahead and did AK's breakdowns by HD: can find it in the OP and here.

Dems in 11/40? Ouch

FWIW, AK's primary electorate was 65% GOP, 35% DEM in 2014 (63/37 this year); I haven't looked at the HD figures for that. A majority of voters are unaffiliated, which may create a psychological impact on primary participation (even though DEM primary is fully-open & GOP primary is semi-closed) since a good share of any winning Democratic coalition in AK will be independents & unaffiliateds.

[Adam Griffin]

Went ahead and did AK's breakdowns by HD: can find it in the OP and here.

Dems in 11/40? Ouch

FWIW, AK's primary electorate was 65% GOP, 35% DEM in 2014 (63/37 this year); I haven't looked at the HD figures for that. A majority of voters are unaffiliated, which may create a psychological impact on primary participation (even though DEM primary is fully-open & GOP primary is semi-closed) since a good share of any winning Democratic coalition in AK will be independents & unaffiliateds.

FTR, I went and looked at primary totals by HD for 2014. I probably need to see if I can find actual turnout totals for '14 by HD; all I saw at first glance on AK's website was the results by HD for each party's legislative primary. In that, 13 HDs went uncontested (3 w/ no ADL candidate & 10 w/ no GOP candidate; these are the ones with darkest red/blue shades).

Assuming that each uncontested district reflects a district where the unopposed party was the majority electorate, then Dems won 11/40 in the 2014 primary as well - but Democrats in 2018 swapped 2 district wins in Anchorage (HDs 17 & 23) for 1 in the panhandle (HD 34) and 1 near Fairbanks (HD 4).

[Adam Griffin]

Hey Griff. I was wondering if you're using the Senate or Gubernatorial Primary results to make this? Or is there a ballot count you're using?

Some states publish the actual turnout for each party's primary/ballot by county: where these are available, I use them.

Otherwise, I use the top contest on the ballot where both parties have available results. Some states don't have senatorial races this year; some don't have gubernatorial races. In a few states, the top contest is only on the ballot for one party (i.e. the state literally doesn't even put an uncontested race on the ballot), meaning you have to go to the next contest or project the outcome, as in the case of VA.

[Adam Griffin]

FL added.