Five Flips in Six Chances: What Pinellas County Measures

Between 2000 and 2024 there were seven presidential elections, and therefore six chances for any county to change which party it backed. Pinellas County, Florida — St. Petersburg and Clearwater, on the western shore of Tampa Bay — took five of them. It went Republican in 2004, Democratic in 2008, Republican in 2016, Democratic in 2020, and Republican again in 2024. The only handoff it declined was 2012.

Pinellas is the extreme case, but it is not a lonely one. In the file behind this piece, 353 counties changed sides at least twice across those six opportunities. That is more churn than the national numbers would seem to allow. Over the same seven elections the country's two-party margin ranged from D+7.4 in 2008 to R+2.5 in 2004 — a total span of just under ten points, with five of the seven results landing inside a six-point band.

The instinct is to read a serial flipper as a fickle place, a county that cannot make up its mind. The arithmetic points somewhere else. A county changes sides because the two coalitions inside it are close to the same size, which means an ordinary national swing is enough to carry it across the line. Serial flipping is a symptom of balance. The counties that moved furthest since 2000 mostly stopped flipping — they crossed the line once, kept going, and are now nobody's target.

What a flip actually measures

We use lean throughout: a county's two-party margin minus the national two-party margin. A lean of R+3 means the county voted three points more Republican than the country did, whatever the country did. Lean is the durable part of a county's politics; the winner's name on the map is the volatile part, because the map is binary and the margin underneath it is continuous.

Duval County, Florida is the cleanest demonstration in the file. It flipped Democratic in 2020 and Republican in 2024, and its 2024 lean is D+0.0 — it voted the national number to the decimal. Duval did not change its mind between 2020 and 2024. The country did, and Duval, sitting exactly on the seam, changed color along with it. Three more of the fourteen largest serial flippers finished 2024 within a point of the national result: Riverside at D+0.2, San Bernardino at R+0.7, Williamson at R+1.0.

A five-time flipper is not a county that keeps changing its mind. It is a county where both coalitions can actually win.

This is also why the county map exaggerates in one direction and hides things in the other. A two-point shift in a county that sits on the line repaints the entire county. A fifteen-point shift in a county that sits twenty points from the line changes nothing you can see.

The county that has flipped five times

Two things are happening in that table at once. Pinellas is close — no result in seven elections is wider than D+8.3 or R+5.2, a total range of 13.5 points across a quarter century. And Pinellas is drifting: its lean moves from D+3.4 in 2000 to R+3.7 in 2024, a seven-point rightward shift relative to the country. In 2004 it voted Republican while standing 2.4 points to the left of the nation. In 2024 it voted Republican while standing 3.7 points to the right of it. The flip count alone would tell you neither of those things.

Balanced is not the same as indecisive

Kent County, Maryland — rural, on the Eastern Shore — is the purest case of balance without movement. Its widest result in seven elections is R+6.7, and five of the seven fell within three points. Its lean has sat between R+1.3 and R+7.1 in every single cycle; it has never once been to the left of the country. Kent has not changed. It flips because it is parked a few points right of a national margin that keeps sliding back and forth underneath it. Four flips, no reinvention.

Butte County, California and Tippecanoe County, Indiana show the other combination: flipping and moving at the same time. Both opened at exactly R+17.0 in 2000, a coincidence we will not attempt to explain. Butte, which contains Chico, ends 2024 at a lean of R+1.6 — sixteen points leftward relative to the country. Tippecanoe, which contains Lafayette and West Lafayette, ends at D+1.4, a nineteen-point relative move, and in 2024 it voted Republican while sitting to the left of the nation. Both are college-anchored counties in otherwise rural surroundings, and both crossed the line four times on the way through.

Blue Earth and Nicollet counties, adjacent in south-central Minnesota, flipped in the same four years as each other — 2004, 2016, 2020, 2024. Neighboring counties that share a coalition tend to share a threshold, which is worth remembering when a single county result gets read as a local verdict. Note also that five of these six counties voted Republican in 2000 and all six voted Republican in 2024. Between those bookends they changed sides more than twenty times combined.

The counties that stopped flipping

Three counties on that list have already left the competitive zone. Salt Lake County flipped three times between 2008 and 2016 and now sits at D+12.1. Chester County, in Philadelphia's western suburbs, flipped three times over the same stretch and now sits at D+16.1. Macomb County, north of Detroit, flipped three times and went the other way, to R+12.4. Their flip counts are historical records, not live traits. Barring a reversal of direction, none of them is decided by a two-point national swing any more.

• Flipped Democratic in 2020, Republican in 2024: Maricopa, Tarrant, Williamson, Duval — four Sun Belt counties, all finishing 2024 between R+3.7 and D+0.0.
• Flipped Democratic in 2008, held through 2020, Republican in 2024: Riverside, San Bernardino, Hillsborough, Fresno — one flip in and one flip out, sixteen years apart.
• Flipped, then left: Salt Lake (D+12.1), Chester (D+16.1), Macomb (R+12.4).

What to watch, with the usual caveats

If the 2024 leans hold — and leans do drift, as Pinellas and Tippecanoe both demonstrate — the counties worth watching are not the ones with the most dramatic histories but the ones sitting closest to zero. Duval at D+0.0, Riverside at D+0.2, San Bernardino at R+0.7 and Williamson at R+1.0 would each change hands on a national swing of roughly a point. Kent County, Maryland has stayed within eight points of the national margin for seven straight elections. Nothing here forecasts an outcome; it identifies which places are positioned to move first if the country moves at all.

The larger point is about how to read the map. A county that has flipped five times has not been unstable. It has been an unusually honest instrument — a place where the margin is small enough that the national tide shows up in the result rather than being absorbed by a comfortable majority. The 353 are not a story about voters who cannot decide. They are a map of where the line between two roughly equal coalitions currently runs, and Pinellas County has spent twenty-four years sitting directly on top of it.