Every four years the news media pay close attention to Vigo County, Indiana in hopes of finding out who the next president of the United States will be. That’s because the candidate who won that county’s vote has also won the national vote in each of the past 15 elections.
Fifteen is quite a long run! But it doesn’t guarantee that Vigo County will be right the next time. With more than 3,000 counties in the country, after all, you could find counties to fit almost any imaginable voting pattern. Maybe Vigo County’s lucky streak was just coincidence.
What’s more, if you look closely you’ll find that even during those 15 previous elections, the county wasn’t always that precise an indicator. For instance, in 2012 the county’s voters went with Barack Obama, who indeed won the election, but his national margin was 8.5 points lower than it was in that chunk of western Indiana.
So BuzzFeed News set out in search of a more accurate model. We began by analyzing the county-by-county results of every presidential election starting in 1972. What we found: For any given election, if you select the five counties that came closest to the national results over the previous four elections, then take the average spread between the Republican and Democratic nominees in those five counties, the result comes surprisingly close to the national spread between the candidates.
Do you live in a bellwether? Type or click on your county to see how closely its votes track those of the nation as a whole.
This method got us within one percentage point of the national popular vote in five of the past seven elections. But major caveat: This time around, shifting electoral preferences could alter the delicate balance that has made these five counties so accurate in the past. This is, after all, the craziest election in recent history.
That said, this year, the five bellwether counties are: