In a further breakdown of the Democratic Gubernatorial Primary of 2018, I’ll start by reposting the map from Part 1’s analysis:
State Rep. James Smith, the victor in this primary, dominated the electoral landscape. A few points from a look at the above map:
- Smith beat his main competitor, Marguerite Willis, even in her home county (Florence). Regardless of how she campaigned elsewhere in the state, Willis should have at least had the name identification in her home county, as her husband, Frank Willis, was formerly a mayor of Florence and a former Democratic gubernatorial candidate himself. Honestly, my guess would be that even voters in Florence County realized that Willis was not a viable candidate in the general election and voted for Smith.
- The maximum percentage for Smith came in Orangeburg County, 76%. Again, Smith’s competitors seemed to lack viability in the eyes of primary voters. However, it is interesting that the Lieutenant Governor picks made by Willis and Phil Noble, did not seem to sway Orangeburg voters. Demographically, Orangeburg is a heavily African-American populated county (the 2010 Census had this county’s population as 62.2% African-American). One might think that the Lieutenant Governor picks of State Sen. John Scott (Willis’s pick and an African-American legislator from the Midlands) and Gloria Tinubu (Noble’s pick and an African-American professor from Georgetown County) would maybe attract some voters. This concept was completely refuted, as Orangeburg County handed Smith his biggest margin.
Moving from Smith’s dominance of this election, let’s try to look at some other numbers from the Democratic primary that might be important come the general election in November. The first item of note would be the proportion of voters that turned out in the Democratic primary compared to the amount of registered voters. That map is seen below.
A quick look at this map wouldn’t reveal anything surprising. Counties that tend to go Democratic in the general election in the past decade have much higher participation in the Democratic primary. At this high level, there is at least a loose correlation in primary results with general election results. The scary thing for South Carolina Democrats is that participation is staggeringly low in counties with massive population and growth: Horry (4%), Greenville (6%), Spartanburg (4%), and York (4%). Although independent voters or voters apathetic to primary voting can change general election results, this map still shows that Democrats have a massive weakness in the Upstate of South Carolina.
I think it is clear where James Smith needs to focus his efforts if he is to have any chance. Smith will especially be assisted if John Warren loses to Henry McMaster in their Republican runoff this week on Tuesday, June 24th. This desert for Democrats will not magically trend Democratic if one of the Upstate’s own ascends to the Republican nomination. However, if McMaster wins, I think Smith can at least reduce the Republican margin of victory to a reasonable amount.
Building on these participation numbers, one last question that enters my mind is how do these numbers look over time? Are the 2018 participation numbers suggesting a positive trend in the Democratic direction for Smith and other Democratic candidates? One last map creation helps us here.
This map needs a preface first though. This map takes an average of the participation numbers from the 2006 and 2010 Democratic primaries for each county. This would be a percentage just like the numbers above. Then, this number is compared to the number on the map above. If the trend is significantly positive for the Democrats, the background is blue. If the number is a non-significant trend (not greater than a 1% change, the background is orange. If the trend is significantly negative for the Democrats, the background is red. The trend number is put on each county with a sign indicating positive (more Dem participation) or negative (less Dem participation). Also, it is important to note that just because Democratic participation went up does not mean that Republican participation went down. They both could go up when more people decide to vote in the primaries. This data simply addresses Democratic primary participation.
Why did I choose the 2006 and 2010 Democratic primaries to create the trend data against 2018? First, these are gubernatorial years where the Democratic nomination for the top office was contested. In 2014, Vincent Sheheen had no competition for the nomination. In 2002, the incumbent Democrat, Jim Hodges, had no competition. Anything previous to this seems too far out because a lot of demographic and political changes have occurred in this state since 2000.
This map is not a definite predictor for the 2018 general election, but just tries to give a high-level idea of trends. Some counties could have really important Democratic primary elections at the county level in 2006, 2010, or 2018 that Republican and moderate voters participated in at high levels. South Carolina’s open primary rules allow crossover voters. Therefore, a county-level race (like a sheriff’s race) could cause participation in the Democratic primary at an abnormal level. These crossover voters cause an aberration in the data and destroy what is a clear trend in party participation.
Thoughts from this last map in regards to James Smith’s chances:
- Democratic participation is finally headed upward in the Upstate. Don’t take too much away from this as the numbers are still absurdly low (see the second map in this article). As aforementioned, if Smith can get McMaster as a general election opponent, there is a path to victory with improving Upstate numbers.
- Was this upsurge in primary participation a long-term trend or a wave of Trump backlash? I don’t know the answer to that question honestly. I would guess that this is an actual long-term trend as demographic trends seem to favor Democrats in the areas they are improving. Regardless, Smith should benefit from either cause of this movement.
- Democratic participation is headed upward in all of the major areas of population growth. Horry, York, Charleston, Greenville, Richland all have had an increase in participation.
- In a more non-Smith note, it would really help to recruit a viable Democrat from the Upstate of South Carolina. Smith’s home turf advantage is going to be somewhat negated by the fact that Richland is already a Democratic powerhouse and that McMaster (the assumed candidate) is from the area as well. A Dick Riley would turn some votes from R to D in a meaningful area.
Stay tuned for an interactive map with precinct level results of the Democratic primary and upcoming Republican gubernatorial coverage (after the runoff).