Trump won:
Essex County, Vermont
Hidalgo County, New Mexico
Juneau County, Wisconsin
Ottawa County, Ohio
Sawyer County, Wisconsin
Valencia County, New Mexico
Vigo County, Indiana
Washington County, Maine
Westmoreland County, Virginia
Wood County, Ohio
Calhoun County, South Carolina
Colleton County, South Carolina
Coös County, New Hampshire
Northampton County, Pennsylvania
Sargent County, North Dakota
Val Verde County, Texas
Most of these places are wrong at one time or another scattered throughout the last 100 years, never all at once.
The chances that these specific 16 counties were wrong? Slim.
The chances that any 16 or more bellwether counties were wrong? Much higher.
With large data sets you can always find patterns that fit the data, but that doesn't mean that those patterns drive the data or will predict the future.
16 out of 24 from what I can see. Not a huge dataset.
Also, very irregular that all would be wrong at once, that’s the point.
Finally, who’s to say there wasn’t also fuckery in the other 8?
The data set is all counties that have voted in the presidential election over the last century.
Intuitively, since elections are roughly a 50/50 split I'd expect any set of 24 random counties to have on average 12 who voted in each direction. Bellwether status isn't some intrinsic property of a county.
I get what you’re trying to say...
I disagree with how you shluff off bellwethers to quickly, I don’t think you understand my point.
Also, quick fact: In 2016 Trump won 2,626 counties and Hilldog won 487.
The country has been slowly transformed....
Even so, not very likely.......
irregularities found in those that didn't lose bellweather status like delware county