For reference, Ted Cruz was the Senate candidate in both 2012 and 2024.

Here's 2020 for reference (a little data is missing but you can see the general trend)

Hi R5, good point on the graph. The original one you replied to was wrong about president/senate candidates being split-popular across counties, but follow-ups across election years and swing/non-swing states show some odd patterns.

bsky.app/profile/ecoe...

Good point, that is insane. Someone on Reddit pointed out the weird voter stats in PA too. Haven't crunched the numbers myself yet but it looks interesting so far.

www.reddit.com/r/somethingi...

Oh I'm referring to how all the dots/counties fall within a very narrow band (i.e. all 67 PA counties prefer Pres candidate between 0-5% in 2024, vs. at least a 10% spread and noticeable outliers in prev yrs).

I'm still trying to figure out if this lends credibility to the claim of flipped Harris to Trump votes or something else; it's just odd (to me) to see 67 and 83 counties having near-perfect margins for president-senate differences, when this isn't the case in past elections.

It's actually pretty common for a presidential candidate to be more or less popular than the senatorial candidate across all counties (every colored dot positive or negative). What's odd here is how closely each county is aligned in their preferences in swing states, but not in non-swing states.

See this thread for more details: www.reddit.com/r/somethingi...

Please see this post (bsky.app/profile/ecoe...) for most recent updates (includes AZ, MI, NV, PA, MD, NJ, UT, and WY).

The above graph has been retracted. The original claim of relative popularity between candidates is wrong, but some questions remain around degree of differences across counties.

This graph has been retracted. Original claim of relative popularity between candidates is wrong, but some questions remain around degree of differences across counties. Please see this post (bsky.app/profile/ecoe...) for most recent updates.