Trevor Sullivan

Strongly related to toxic masculinity and "male loneliness epidemic".

Men have forgotten how to have friendships (or maybe the culture has forgotten that friendship is masculine), and incels are pouring gas on that fire.

1. What does that have to do with Clark County Nevada?
2. That case is over TWO votes. 2. II. Dos. Zwei. The number between one and three.

Some loony ran for office and got 9 friends to say they voted for her when she only officially tallied 7.

That's not a scandal, it's a circus sideshow.

What part of "that's not what a Russian Tail looks like" don't you understand?

Elections aren't normally distributed. The original Udot analysis that invented the term "russian tail" is predicated on the fact that elections aren't normally distributed.

That's what I said? You aren't reading me, slow down.

None of the graphs have time in them.

You are the one who brought up time.

None of their graphs and none of my graphs have time on the x axis. There are no graphs anywhere with time on an x axis. You're mad about a thing that doesn't exist

Boy, sure is a good thing that there isn't a single chart in all of this that uses time on the x axis then, isn't it?

Can you at least try to make points that are whatsoever connected to the thing you're trying to criticize?

Really this is just telling me that you didn't read it, because on one reply you say "they're right because they used Udot's method" and on the other "they're right because they fitted a bell curve", when the ENTIRE POINT of Udot's method is that elections DON'T adhere to single bell curves.

Yeah, and that reason can be "the source distribution isn't a bell curve".

The Central Limit Theorem only holds when all samples are taken from the same source.

It is insane to expect that polling sites in central Vegas and a post office on top of a mountain should average to the same avg opinion

Read the site.

ETA are lying about how Udot's analysis works.

Elections are not unimodal and spikes are not tails.

This is a good question, and led to a little bit of bonus analysis! Click through to read the thread, it's not very long.

TLDR: the claim is false, and it's another instance where a smoking gun exists in the data and is easy to find but ETA didn't publish it because it disproves them.

GitHub - Trevortds/clark-county-election-analysis: Scrollytelling Essay rejecting the analysis of Election Truth Alliance. Scrollytelling Essay rejecting the analysis of Election Truth Alliance. - Trevortds/clark-county-election-analysis

I'm so flattered!

The graphics were all done with Observable Plot, with a little D3 on top for some animations.

Processing is python and raw JS.

This is a "turnaround & performance at expense of maintainability" stack. I had to work in a narrow window between busy periods, no time for frameworks

There is more Trump-Rosen than Harris-Brown cross-dropoff, but that's not surprising.

Rosen is an anti-immigration Democrat, she has real appeal to Republicans.

Brown, rabidly anti-choice carpetbagger, had no purchase with Democrats or a public who is also voting for a right to abortion 65-35.

{ "trump-rosen": 26321, "trump-brown": 427950, "trump-none": 38446, "harris-rosen": 481698, "harris-brown": 8427, "harris-none": 29866, "by_vote_type": { "early": { "trump-rosen": 7928, "trump-brown": 213594, "trump-none": 12709, "harris-rosen": 147299, "harris-brown": 2161, "harris-none": 7245 }, "mail": { "trump-rosen": 10166, "trump-brown": 137312, "trump-none": 13346, "harris-rosen": 255005, "harris-brown": 3831, "harris-none": 12619 }, "election_day": { "trump-rosen": 8227, "trump-brown": 77044, "trump-none": 12391, "harris-rosen": 79394, "harris-brown": 2435, "harris-none": 10002 } } }

The more interesting topic is whether or not there are cross-dropoff votes, like you said.

This would be another "smoking gun" that's easy to find but ETA chose not to.

Sure enough, it doesn't support the idea. Early votes, the ones that were supposedly manipulated, have the LEAST cross-dropoff!

But that's exactly what we expect from these candidates!

Trump is a cult leader, his voters love him and only him, they tick his name and then leave. High dropoff.

Harris is just a party leader, her voters don't care about her, they want (D) to win, so they mostly vote the whole (D) package.

I was gonna pass on this, but I realized I have a chance to put it to bed, so, congrats you made me do more work lol.

So, that isn't alleged by ETA specifically, though they do imply it.

What ETA points out is that Harris had a lower dropoff rate, and they wiggle their eyebrows suggestively abt it

If this result can ONLY be the outcome of hacking, then that would mean Elon Musk hacked my simulation before I even published it, how did that happen?

If the method of the hacking was that machines start changing votes after a certain number are counted, how come none of them actually did that?

They say that sure. Does the data support it? No

The only reason they noticed the 2024 data is because Republicans early voted en masse so some tabulators counted over 1000 votes.

Remember the coin flip game: the more votes you count, the more consistent the result.

Every county is not an independent coin flip. They're more like a whole bunch of coins all loosely glued together at the sides and flipping at once.

Every county is not it's own universe. One day, in one country, people came out everywhere to vote for one candidate, which resulted in many flips.

THIS HAS HAPPENED BEFORE.

In 1932 Herbert Hoover flipped 0 counties and FDR flipped hundreds.

So this "one in an octillion" event had an observed occurrence rate of 2 in 23, or just under 1 in 10.

Did Musk go back in time and hack the hand count in '32 as well?

Yes, paper audits, the largest ones in history.

Look, you're in a headspace where you have these beliefs that you won't change because any evidence that contradicts them you find some excuse to ignore

You need to get to a space where you derive your beliefs from reality, and I can't help with that

The chance of a raindrop hitting your head in any given minute of your life is like 1 in a million

If 40 raindrops fell on your head in a minute, you haven't experienced a statistical impossibility that can only be explained by a conspiracy, you've experienced rain.

This is not how probability works. You can only multiply probabilities when events are independent (with no common cause).

The behavior of voters in different counties in the same country in the same election for the same candidate are not independent.

Can you elaborate a little bit? The names are all PA counties, it seems, so what do the numbers mean?

We should want an explanation, and we should want accountability for the people who failed us. Direct that energy towards the people who actually did fuck up: party leaders, media editors, social media companies, and voters; not the boogeyman of hackers that don't exist.

I don't want to believe it either, but that's not a good enough reason alone to disbelieve it.

The results matched the exit poll data.

Audits have been done to confirm them.

Election night was a gradually building gloom, not a big shocker when official numbers dropped different from expectations.

Hackernews thread here, for technical discussion: news.ycombinator.com/item?id=4472...

@r5-to-philly.bsky.social has conducted an excellent critique of ETA's Pennsylvania analysis. Check out these two threads for more:

bsky.app/profile/r5-t...

bsky.app/profile/r5-t...

Wisconsin audit of Trump election win finds zero voting machine errors A new audit of the November election won by President Donald Trump in swing-state Wisconsin has found that not a single vote was counted incorrectly, altered or missed by tabulating machines.

Again, read the website, I go over this.

There have been audits. Wisconsin hand counted more ballots than it ever has in its history. More than the entire population of Clark County.

Zero errors in the machine count.

www.cbsnews.com/minnesota/ne...

You're spot-on about that difficulty, but there's a pretty simple workaround:

Most people are going to vote at a location near where they live, and each vote record includes the voter's precinct of residence.

So you can figure out where each machine is by looking at who used it.

There have been audits, ETA doesn't want to tell you about them because it contradicts their narrative.

Read the website.

I made a full geographically informed simulation of the county and it produced graphs identical to the "statisticaly impossible" ones ETA claims are proof of fraud.

So it definitely turned into a thing. Today I'm launching a website that digs deep into how Election Truth Alliance misleads people with bad statistics. bsky.app/profile/trev...

@skepchick.org since this started with your bsky thread and the outcome might be interesting to you.