seen from Saudi Arabia
seen from France
seen from United States
seen from United States
seen from Türkiye
seen from Malaysia
seen from United Kingdom
seen from T1
seen from United States
seen from United States
seen from United Kingdom
seen from Australia
seen from Singapore
seen from Australia
seen from United States
seen from Türkiye
seen from Australia
seen from Australia
seen from Australia
seen from Australia
Just a little election season reminder to y'all:
When the polls are making you feel discouraged about Kamala's chances of winning, remember that the majority of these polls are taken BY PHONE. The groups most likely to vote for her are the ones who are young enough/smart enough to NOT answer their phones when it's a number they don't recognize.
New Yorkers line up to vote in the presidential election, November 2, 1936.
Photo: General Photographic Agency/Getty Images/U.S. News & World Report
HEY I'M MAKING A DYSKINGDOM.
Turns out there are a few other Dyskordes when I searched but I found myself the most in the results. So I'm starting the Dyskingdom. I will make a poll for an election for a DysKing (gender neutral). Any other Dyskordes are welcome, as well as anyone else who'd like to join. Long live the Dyskingdom!
Who shall be the DysKing of the Dyskingdom?
Me, Dyskorde aka almostangrything
My mother (please no)
YOUR mother
a rat (me in disguise)
an actual rat
moots are automatically part of the Dyskingdom
@lavenderwhirls @yumnasfunblogsaysfreesudan @sleepdeprivedsimp234 @alchemicalwerewolf @b3nnyh4rr1s
@snailcheeserulz @gayoticbeing @usernose @sleepinggoddess222 @sleepy-boything-shit
@doomwolfe @autistic-fuckwad @bettathanyou @logan-bear-bear @omghiiiiii
I'm gonna need everyone to get real savvy and real skeptical real fast about political polls and election models in the next 4 months. News organizations are starting to grasp for polling numbers to understand how Kamala Harris will match up against Trump, and I need you all to remember and repeat this maxim:
Electoral opinion surveys and preference polls are not useful without precedents.
The media is going to turn to them because they are desperate for narratives and we are going to be starved of information up until votes get counted.
Do not be misled; the media are effectively running to soothsayers. Both if polls show Kamala doing well, and if they don't.
Here's the thing you need to understand about election polls in the US. Predicting a winner in a presidential election is extremely nontrivial, because of the Electoral College. Outcomes depend on margins in a handful of states; this is why you see organizations like 538 talking about the "Path to 270" and the make-or-break states for individual candidates. A huge fraction of American voters live in California, and their preferences largely don't matter statistically, since California always votes blue. So a straight random poll of Americans will always show the Democrat ahead, because so many Americans live in California and New York (yes this is also why the Electoral College sucks).
To counter this trend, pollsters will use statistical models to adjust the polling sample, trying to guess at who will vote in which state, and rebalancing their polling sample (often by applying statistical weights) to try to get a more accurate sense of how the race will actually break.
So right off the bat, that introduces a serious degree of uncertainty. Good pollsters account for that in the poll's margin of error (not that the media ever let a margin of error get in the way of a narrative), but estimating the margin of error due to statistical weights itself relies on statistical models. Pollsters look at past elections and the variability in turnout relative to polling results (since there are polls that try to measure turnout specifically) to build a model for how uncertain their turnout predictions are.
And that is the key point here. A presidential election opinion poll rests entirely on a prediction of future voter behavior based on past behavior.
That works fine in normal elections with normal candidates. This was not a normal election before Biden dropped out, and it certainly is not one now. No major party's candidate and presumptive nominee has dropped out after the main nominating contest. Not since the advent of the modern party system with primaries and caucuses that are open to the average voter. No major party has fielded a campaign for the presidency during a standard election year in only 15 weeks. Ever.
We do not know what impact that will have on voter turnout, in any demographic or in any region. We have no precedent for this.
So if someone tries to sell you a narrative about how the election is going? They are lying to you in order to sell you a narrative, either for propaganda reasons (a campaign) or financial reasons (a for-profit media enterprise).
I know it is tempting to grasp at every straw of information to understand how the campaign is going. Trust me, I have 3 weather forecast apps and 4 weather radar apps on my phone. I know how compelling it is to want to know what's happening and what will happen.
The best we can do is have conversations with the people in our lives who are not convinced they want to vote for Kamala Harris or aren't convinced they want to vote, and honestly and earnestly get them to come out to vote. Doesn't matter what state you vote in. Don't let historical precedents that do not apply dictate your actions. Have those conversations.
And for fuck's sake, ignore the media pundits losing their minds trying to fill the airwaves with predictions. They don't know what's going to happen any more than you do.
My fellow (USA) Americans: Did you vote in this election?
Fuck yeah I voted!
Well, I was gonna…but I didn’t actually vote
No I didn’t vote intentionally, but I’m eligible to vote
I am NOT an eligible voter but would vote if I could
I am NOT an eligible voter and would NOT vote if I could
Something else
See results
My get out the vote message in gifs:
Resources for voting:
The first presidential election in which every state picked their electors via statewide popular vote was 1868. Before that, at least one state had the legislature pick the winner (South Carolina was the last holdout). Back then, only white men over 21 could vote, but the electorate saw significant boosts in 1920 with the passage of the 19th amendment (white women) and 1972 following the civil rights movement (black men and women) and the passage of the 26th amendment (18 year olds).
The total number of eligible voters grows with the overall population of the country, and looking at the data since 1972 we can predict that there will be about 244 million eligible voters this year.
But eligibility is only half the equation. Actual voter turnout never even comes close to 100%. Since 1972, it has never gone above 66.6% (in 2020), averaging 57.2%
57.2% of 244 million voters would mean we'd see 140 million votes cast in November, but that's actually way lower than we would expect. Let's ignore turnout percentage (which fluctuates wildly between 50% and 60%) and look instead at the actual number of votes cast:
There were over 158 million votes cast in 2020, and given that we saw record turnout of 66.6% due to a number of simultaneous crises, it makes perfect sense that the total number of votes cast in 2024 would fall, but not all the way down to 140 million. Between 1984 and 1988, the popular vote dropped by 1 million. Between 1992 and 1996, the popular vote dropped by 8 million. Between 2008 and 2012, the popular vote dropped by 2 million. No polls seem to indicate an 18 million dropoff in voter turnout this year; that would be unprecedented. According to the line of best fit, it is much more likely that we'll see ballpark 150 million votes cast, probably closer to 151 million. That would indicate a voter turnout of around 62%, which is higher than average but more realistic.
Since 1972, the two major parties combine to win an average of 95.5% of the popular vote per election. Third party candidates do well in waves, with considerable showings (over 5%) in 1980, 1992, 1996, and 2016. Since Kennedy dropped out of the race and neither the Libertarians nor the Greens are making waves as spoilers this year, we can assume that the Democrats and Republicans will have a better than average showing. Looking at the trend lines for major and third party percentages, it would show Kamala Harris receiving 51.7% of the vote (about 78 million), Donald Trump receiving 45.0% (about 68 million), and third party candidates combining for 3.3% (about 5 million).
To scale:
I do not believe that the actual final results will be anywhere close to this prediction. I think that these extrapolations show Kamala Harris overperforming by a considerable amount. She will almost certainly win the popular vote, I don't doubt that, but I think she will win a plurality instead of a majority (under 50%). Donald Trump is not wildly more or less popular than he was the last two times, in which he received 46.1% and 46.8% of the vote, so I do not think it is unreasonable to assume he will receive a minimum of 47% this year, given demographic shifts among suburban voters, black men and Hispanics. He may even perform better than that, but not better than Harris. I would bet my life on that fact. Democrats have won the popular vote in 7 of the last 8 elections, and polls do not show Trump magically coming up from behind on his third try.
When we exclude years with exceptionally high third party turnout, the major parties average 98.3% instead of 95.5% since 1972, but we can't just pick and choose what data to include. At this point, I'm deviating from the hard numbers and am operating more on vibes. Third party turnout will be lower than in 2016 (5.7%), but probably higher than in 2020 (1.8%). If we take the 3.3% figure from the trend lines, that would mean Harris and Trump would receive a combined 96.7% (146 million votes). If we assume Trump has a floor of 47% support (about 71 million votes), then the best Harris could do is 49.7% (about 75 million votes). There is no conceivable scenario where Trump wins the popular vote, so his ceiling (and Harris's floor) is something like 48.35% (73 million votes each).
The only president to win two non-consecutive terms was Grover Cleveland, but he actually won the popular vote all three times he ran, 1884 (48.8%), 1888 (48.6%, lost the electoral college), and 1892 (46.0%). Trump is the exact opposite, having NEVER won the popular vote but still won the electoral college regardless. Franklin Roosevelt won four times in a row, 1932 (57.4%), 1936 (60.8%), 1940 (54.7%), and 1944 (53.4%). There was a major dropoff between his second and third bid, even though he was ridiculously popular. Trump is NOT ridiculously popular, but I don't think he's going to perform worse this time. Cleveland performed worse because there was a significant third party challenger, and FDR was the first sitting incumbent to run for a third term (at a time when WW2 was ramping up and voters were afraid of America joining). Trump does not have either of these disadvantages, so while it's possible he could perform worse than in 2020, I think Harris has so much baggage attached to her name (real and imaginary) that Trump will do better than he did against Clinton or Biden. A lot of voters are fed up with him, but that doesn't mean they will support Harris.
I have no real data to confirm these confidence intervals, but this what I would expect next month
47.0% to 48.3% for Trump (71 million to 73 million), I'd say 48.0% (72.5 million)
48.4% to 49.7% for Harris (73 million to 75 million), I'd say 48.7% (73.5 million)
Now, the electoral college is anybody's guess! Who fuckin knows?
Hey guess what, trump is up in the polls, despite everything. We really need to change that especially in the swing states. DO NOT VOTE 3RD PARTY. DO NOT SIT OUT. IT IS THE DIFFERENCE BETWEEN A FASCIST IN OFFICE AND NOT HAVIG THAT. Take it from a fool who voted for trump naively in 16, and against him in 20. We DO NOT WANT HIM BACK!
