Donald Trump’s performance in primaries and caucuses has created some unusual geographic patterns. What sort of candidate dominates in both Alabama (where Trump won 43 percent of the vote on Super Tuesday) and Massachusetts (where he got 49 percent)? Why was Trump so strong in Hawaii — and so weak in Kansas? The answers to these questions are complicated. Trump attracts voters for multiple reasons: Economic anxiety and racial resentment are important factors in his success, but hardly the only ones.
Let me propose One Simple Trick that makes the geographic patterns at least a bit clearer: Trump’s strengths and weaknesses are easier to understand if you consider how many Democrats and Republicans turned out in each primary. Republican voters are a small minority of the overall electorate in states like Massachusetts and Hawaii, so their support for Trump isn’t a good indication of how those states might behave in November.
Take Massachusetts, for example. Trump’s 49 percent — his highest fraction in any state to date — was on light turnout: Only about 630,000 voters participated in the Republican primary, compared with 1.2 million for the Democratic one. Thus, Trump won only about 17 percent of the overall vote among Bay Staters who turned out that day.
By contrast, while Trump’s performance in Ohio might seem poor at first glance — he got 36 percent of the vote and lost to John Kasich — it’s better once you consider that Republicans turned out substantially more voters than Democrats. As a share of the combined Democratic and Republican primary turnout, Trump got 22 percent of the vote in Ohio — a fair bit better than he did in Massachusetts.
Here are those figures for all states where both Democrats and Republicans have voted so far.1
|TRUMP’S SHARE OF PRIMARY/CAUCUS VOTE|
This calculation makes it clearer that Trump’s strengths are mostly in the South. Of Trump’s top seven states so far by his share of the combined primary or caucus vote, five or six are in the South, depending on how you classify Missouri.
New England looks like a poor region for Trump, by contrast. His share of the combined primary or caucus vote was slightly below his national average in Massachusetts and New Hampshire and well behind it in Vermont and Maine.
The industrial Midwest has been about average for Trump. The Ohio result, as I mentioned, is better for Trump once you consider the very high Republican turnout there (and that he lost to the state’s governor). But his performances in Michigan and Illinois rate as middling by this metric even though Trump won both states. The Great Plains states have been a poor region for Trump, while the West has been a mixed bag. Trump’s big win in Nevada is less impressive once you consider that Democratic turnout outpaced GOP turnout. But his results from Arizona hold up well.
This method leaves a lot of things to be desired. If you were being more exacting, you’d want to adjust for whether each party held an open or closed primary in each state, how many opponents Trump faced at various stages of the race, and other factors.
But as the calendar turns toward bluer states, be wary of making extrapolations from Trump’s performance in the primaries to how he might perform in the general election. Overall, Trump is deeply unpopular with general election voters and will have a lot of work to do to repair his image should he become the Republican nominee. The race can and will change, and Hillary Clinton shouldn’t take a lot for granted. But Trump is more likely to “transform the electoral map” by turning red states blue, rather than the other way around.
Listen to the latest episode of the FiveThirtyEight elections podcast.