Election Day 2018 is quickly approaching, and for the political junkies out there, it helps to know a good bellwether for a given location. On election night, a lot of media will be leaning on exit polling and early returns to try to figure out what the end result is going to be. Sometimes, the easiest answer is from 30,000 feet.
Using historical data from the top ticket races from 2012 (Obama/Romney), 2014 (Sheheen/Haley), and 2016 (Trump/Clinton), I have compiled a table analyzing which county serves as a good barometer for what the end result will be statewide in South Carolina.. The analysis works by calculating the deviation from the overall state percentage of the vote for the two main party candidates.
For example, if the statewide result is R 52%, D 47% and one county is R 50%, D 48%, then the errors for that race are below:
Overall error: 1.5%
Without further ado, the top 5 lowest county errors in the data created from the 2012-2016 SC results are as follows:
|#||County||Comparative County Error to Statewide Result|
The simple take away from this data is to pay attention to Georgetown County! I’ve posted the PDF below of this analysis, but the fact that Georgetown fell below the 2% error threshold in all three of the top-tickets races in 2012, 2014, and 2016 is astounding! However, knowing that Georgetown County has a slight lean toward the Republican party like the overall state, this data seems sensible and believable.
In fact, just as quick evidence, Trump received 54.94% and Clinton received 40.67% of the vote statewide in South Carolina in 2016. In Georgetown County, Trump received 54.93% of the vote (0.01% error) and Clinton received 42.04% of the vote (still a not too shabby 1.37% error).
If you would like to look at better year-by-year data, or look and see where your county fits into this picture, see the two attached PDFs, one in alphabetical order and one in ascending order (smallest error first).
Remember that on election night 2018 in SC, if you already know where Georgetown is going, you probably know who the Governor is going to be.
Preview of data: