Mega-Thread: 2020 United States General Election

[Rasulis]

IA and GA are definitely on the table, too. As is Texas, theoretically, though those odds are looong.

The race for Texas for the Democrats will depend on Texas ten largest counties.

Harris alone is worth 15% of the total votes in Texas in 2016 and also in early voting 2020. Obama won Harris by 1% in 2008 and 2020. Clinton by 12% and O’Rourke by 16.7%. If Biden can win Harris by between 20 - 30%, that will be a game changer.

Then there is the turnout. Harris, Dallas, Tarrant, Bexar and Travis currently account for 49% of Texas early voting. Well ahead of 2016 where these five counties accounted for 42% of Texas total votes. Dallas voted Hillary +26%, Bexar +13% and Travis +38%. The biggest weakness is Bexar which tends to have low turnout when compared to Dallas and Travis. Even in 2020, although the county is doing better, it still lags behind Dallas and Travis.

This is where it gets interesting. Trump won Tarrant county by 8.5% in 2016. O’Rourke won Tarrant by 4,290 votes in 2018. There was an 8% shift in the parties balance of power in Tarrant between 2016 and 2018. If Biden can flip Tarrant that will be big.

Lets take a look at a couple more large Texas counties. Trump won Denton by 20% in 2016, Cruz won by only 8% (12% shift). Trump won Collin by 17% in 2016, Cruz won by 6% in 2018 (11% shift). Flipping Denton and Collin is probably a tall order. However, if Biden can close the gap, that will really help. The same with Montgomery and Williamson. If Biden can narrow the gaps to single digit that would be great.

The remaining counties, Fort Ben and El Paso are pretty solid Blue. El Paso voted for Clinton +44%. So turnout, turnout, turnout.

[Benggaul]

Another ballot box set on fire...this time in Boston:

Dozens of ballots destroyed after fire set at Boston dropbox

Authorities have charged a man for allegedly lighting a ballot box on fire and damaging dozens of ballots in Boston over the weekend, police said Monday. The Boston Election Department said 87 of the 122 ballots inside the dropbox were legible and able to be processed, but the rest were destroyed or unable to be counted.

Worldly Armand, 39, was taken into custody late Sunday after officers matched his appearance with the description of a person wanted for the alleged arson, police said. After a query by the officers, they found Armand was wanted on a warrant out of a local court for receiving stolen property. He will be arraigned in Boston Municipal Court on a charge of willful and malicious burning.

Police said smoke was coming out of a ballot box near Copley Square early Sunday morning and firefighters extinguished the fire by filling it with water. The Boston Police Department released photos to help find the culprit.

Thankfully "only" a few dozen ballots were destroyed and the local government is sending out new ballots to everyone identified as having deposited a ballot there in the time just before the damage was done, so ultimately this shouldn't have an effect on the election.

[zEmini]

I voted. First time voting democrat since Obama's first term.

[Edge-]

Another ballot box set on fire...this time in Boston:

Dozens of ballots destroyed after fire set at Boston dropbox



Thankfully "only" a few dozen ballots were destroyed and the local government is sending out new ballots to everyone identified as having deposited a ballot there in the time just before the damage was done, so ultimately this shouldn't have an effect on the election.

I'm about to drop mine and my brothers off at a local dropbox this week, nobody in my sleeply little town better feel the need for some fire. We already got enough of that shit from the wildfires, and I don't need to worry about my ballot not getting counted.

Also, fuck some of the propositions on the CA ballot. Prop 24 is a fuckin doozy of a proposition, with some great shit in there and...some really bad shit that functionally seems to undermine it. Really unsure of my vote still, and I desperately hope we can get a better written law/proposition to protect consumer data in the future.

[DarkTZeratul]

Also, fuck some of the propositions on the CA ballot.

Fuck Proposition 22 in particular.

[D Luniz]

You and I have had our share of differences, but we've always been on the same side of the equation. Faithless posters like @Donatello Trumpi aren't worth responding to, except perhaps for practice against literate adversaries. You're never going to get a "you're right", although it is nice to see him eat his own words.

I say this of course and I don't follow my own advice....

He really isn't. He posted some dogwhistle standard talking point bullshit, and when asked to explain so he couldn't try the standard bigot weasel tactic of claiming people were putting words in his mouth, he just bailed on posting(but not reading, I saw your logins sucka) till today.

[draynay]

Another miscalculation by Biden on the cost of his proposals.....

He should just say Mexico will pay for it, amirite?

[Edge-]

Fuck Proposition 22 in particular.

That was a hella easy "no" at least. I don't mind if there are obvious garbage ones, 24 just hits me because there's some good shit in there I like and it does make some improvements on the existing legislation, but the removal of "Devices" from the categories of data sales (just individual/family now) and the increase in the minimum sale threshhold from 50K to 100K really rubs me the wrong way. Not a fan of putting the onus on individuals to manually request opting out of data collection/sales rather than forcing online companies to ask user for permission, which I could live with were it not for some of the other issues.

In other news - https://www.cnn.com/2020/10/26/healt...vid/index.html

Utah hospitals could be days away from using a patient's age, health and other factors to decide who can remain in overcrowded intensive care units due to an onslaught of Covid-19 cases.

A group of administrators representing Utah's hospitals presented Gov. Gary Herbert with a list of "criteria they propose doctors should use if they are forced to decide which patients can stay in overcrowded intensive care units," The Salt Lake Tribune reported.

Reminder -

The "death panels" Republicans whinged and pearl clutched about in 2008 over the Affordable Care Act, which were never a thing under Obama, exist in a very real way in Trump's America. Doctors are literally deciding if grandma and/or grandpa get hospital treatment or not, and if the hospital is overcrowded then older covid victims may be shit outta luck.

Remember about the fears of "rationing health care" under a M4A plan? It's very real under the current private insurance system.

[Paranoid Android]

No, the best case scenario would be Biden winning all 50 states, cause then it would show how fucking stupid we were for letting Trump get elected in the first place and it would be a mea culpa to the world. Anything less than that, shows the country has a LOT to learn and atone for.

Yes, I just want a win. I will still be amazed if this is close. Happy but slightly disappointed if Biden doesn't get over say the 304 mark Trump won. Yes, Electoral College is stupid with winner takes all but I'm basically stating I will be perplexed if he wins states that are not full Trumpkin, even though he can.

[Redwyrm]

Hey look! Two cases of election fraud...



Oh, wait. That's just two rich white people voting by mail. NVM

[Forogil]

Your first point is a fallacy. Increasing sample size would not lower the MoE unless it is several orders of magnitude higher, i.e. 1,000 to 1,000,000 and even then the confidence interval would be lowered by one percentage point. That's the fundamental basis for statistics and probability.

Eh?

If you increase the sampling size by 1000 your margin of error should decrease by a factor of 31.6, and not just by a percentage point.
Obviously that assumes the sampling is fair, as the systematic bias is not reduced.

However, you can also decrease the margin of error (both statistical and to some extent systematic) by using data for different sub-populations (men/women, education, race, etc) in a good way.

[snorgas]

No, the best case scenario would be Biden winning all 50 states, cause then it would show how fucking stupid we were for letting Trump get elected in the first place and it would be a mea culpa to the world. Anything less than that, shows the country has a LOT to learn and atone for.

Sorry, but you lost the trust of the world when you elected a religious maniac to high office in 2000.

There is zero probability of the US being let back into the international community. You are clearly oblivious to how the rest of the world views you: It fucking hates you. Russia, China etc aren't popular but it is not even close.

The idea that you can elect another senile warmongering racist with access to nukes as an apologia for the last one makes absolutely no sense.

[cubby]

That was a hella easy "no" at least. I don't mind if there are obvious garbage ones, 24 just hits me because there's some good shit in there I like and it does make some improvements on the existing legislation, but the removal of "Devices" from the categories of data sales (just individual/family now) and the increase in the minimum sale threshhold from 50K to 100K really rubs me the wrong way. Not a fan of putting the onus on individuals to manually request opting out of data collection/sales rather than forcing online companies to ask user for permission, which I could live with were it not for some of the other issues.

In other news - https://www.cnn.com/2020/10/26/healt...vid/index.html



Reminder -

The "death panels" Republicans whinged and pearl clutched about in 2008 over the Affordable Care Act, which were never a thing under Obama, exist in a very real way in Trump's America. Doctors are literally deciding if grandma and/or grandpa get hospital treatment or not, and if the hospital is overcrowded then older covid victims may be shit outta luck.

Remember about the fears of "rationing health care" under a M4A plan? It's very real under the current private insurance system.

And the idiot Trump Cult Followers don't realize that Trump's covid treatment was all a version of M4A.

[TexasRules]

Biden holds an 8 point lead with only 8 days left, with 60 million people already voting.

I think we're gonna be fine.

I don't believe in polls, but you guys froth over them like lemmings, so ...enjoy

https://www.rasmussenreports.com/pub...se_watch_oct26

Trump 48%, Biden 47%

[Machismo]

I don't believe in polls, but you guys froth over them like lemmings, so ...enjoy

https://www.rasmussenreports.com/pub...se_watch_oct26

Trump 48%, Biden 47%

I haven't been able to get anyone to answer this.

Why do you support a guy who keeps white supremacists as senior staff?

https://www.splcenter.org/stephen-mi...eitbart-emails

[Deja Thoris]

I don't believe in polls, but you guys froth over them like lemmings, so ...enjoy

https://www.rasmussenreports.com/pub...se_watch_oct26

Trump 48%, Biden 47%

Posting something you don't believe in. How Trumpish.

[Thepersona]

I don't believe in polls, but you guys froth over them like lemmings, so ...enjoy

https://www.rasmussenreports.com/pub...se_watch_oct26

Trump 48%, Biden 47%

Ah, look the one poll that has Dotardito Up.

In real news, Reuters/Ipsos just released a new poll:

Wisconsin:

Biden 53% (+9)

Trump 44%

Michigan:

Biden 51% (+7)

Trump 44%

Pennsylvania:

Biden 50% (+5)

Trump 45%

Florida:

Biden 50% (+4)

Trump 46%

Arizona:

Biden 49% (+3)

Trump 46%

[kaelleria]

I don't believe in polls, but you guys froth over them like lemmings, so ...enjoy

https://www.rasmussenreports.com/pub...se_watch_oct26

Trump 48%, Biden 47%

Some polling sponsored by groups that are conservative (or are Rasmussen) have this race tightening... But I really don't see it in any other piece of data that's out there.

https://twitter.com/ForecasterEnten/...25749016817672

[Endus]

Eh?

If you increase the sampling size by 1000 your margin of error should decrease by a factor of 31.6, and not just by a percentage point.
Obviously that assumes the sampling is fair, as the systematic bias is not reduced.

However, you can also decrease the margin of error (both statistical and to some extent systematic) by using data for different sub-populations (men/women, education, race, etc) in a good way.

This simply is not true.

For pretty much any given sizeable population, a sample size around 1000 reliably gives you 95% confidence in the results, +/- 3%. There's some mild variations based on the actual population, but it's marginal; a population of 30,000 needs exactly 1030 people in the sample to hit this, and 300 million needs 1067.

Increasing that sample size to 2000, for a population like the USA's 250 million voters, and for the same 95% confidence level, the interval is reduced to just +/- 2.19%

About a percentage point. Slightly less.

This is basic statistical mathematics. Feel free to use any survey size calculator, like this one, to check; https://www.surveysystem.com/sscalc.htm

If you mean something other than confidence interval or level when you say "margin of error", then you're talking about something that really has nothing at all to do with sample sizes or statistical mathematics to begin with and you're really gonna have to explain yourself.

[PhaelixWW]

I don't believe in polls, but you guys froth over them like lemmings, so ...enjoy

https://www.rasmussenreports.com/pub...se_watch_oct26

Trump 48%, Biden 47%

I don't believe for a second that Trump has 23% of the Democrat and 27% of the Black votes, nor do I think his approval is at 52%.

That Rasmussen poll is just as much trash as they mostly are.