POTUS or Prisoner; The '24 Trump Campaign Fvckery thread

[Dameon Farrow]

Speak on it breh.. @FAH1223 posting 10 polls back to back to back talkin bout it’s bad news for Biden but posting senate posts on swing states showing the Dems being up

I wanna see a poll about which poperty he'll have to sell off to meet this bond by Monday. But I know he won't do that.

 

[Professor Emeritus]

I’d say 2016 and 2020 are evidence of how badly polls are off the mark. 2016 had Hillary w/ a fairly comfortably win. 2020 had Biden w/ a blowout, but dude “barely” won across 3 critical states.

538 predicted heading into the final 2016 weekend that based on all the polling, Clinton had a 70% chance of victory. That's not a "fairly comfortable win", that's closer to 50-50 than to 100%. Imagine if a doctor told you before a surgery that you had a 30% chance of death, would you feel comfortable?

If you want to put it a different way, the polling average had Clinton up in the popular vote by 3.9%. In reality, she won by 2.1%. So the polls were off by 1.8 percentage points. Are we really going to claim, "Damn, these polls are worthless!" because they were off by just 1.8%?



538 predicted heading into the final 2020 weekend that Biden had a 90% chance of winning - so Trump still had 10% chance. Biden did indeed win by the skin of his teeth, but he had three different buffer states - Trump would have had to come out on top in Arizona AND Georgia AND Wisconsin in order to beat him. It feels like the race was indeed closer than the polls indicated, but was it that big a miss? If they were really predicting a blowout, they would have given Trump a 1-2% chance, not a 10% chance.

On the state-by-state forecast, literally the only states that 538 missed on in 2020 were North Carolina and Florida. They got 48 out of 50 states correct. That's 96%.

In terms of the popular vote, the final voting average predicted that Biden would win the popular vote by 7.9%. Instead, he only won by 4.4%. So that was a 3.5% miss. That's not a good outcome at all, the polling companies want to do better than that. But they were trying to predict turnout in a literally unprecedented situation - during a pandemic, with massive mail-in voting, with a fanatical Trump base that is like nothing we've ever seen. And despite all those unique factors, they still only missed by 3.5%.

 

[Professor Emeritus]

I'm far from a pollster apologist but I feel like people really just started paying attention to polls last general election but don't really understand them.

Case in point:

Pollsters now say that they have margins of error so they don't lose face. But unless the margin is very narrow then the poll becomes as reliable as a decision in a close boxing fight.

Breh, pollsters have been reporting their margins of error for your entire lifetime. A margin of error is a statistical necessity for any sort of representative sampling, it is literally impossible to poll every single voter. No, having a margin of error does not make the poll useless, not even for close elections - you just have to realize that you're dealing with probabilities rather than with absolutes.

How good the poll is is primarily based on how large their sample was, how representative their sample was, and whether they asked the questions in a way that led people to give their true answer. And "how representative their same is" becomes the toughest question because you don't just need to be representative of the American population, you need to be representative of the actual people who will turn out to vote.

 

[Liu Kang]

Case in point:




Breh, pollsters have been reporting their margins of error for your entire lifetime. A margin of error is a statistical necessity for any sort of representative sampling, it is literally impossible to poll every single voter. No, having a margin of error does not make the poll useless, not even for close elections - you just have to realize that you're dealing with probabilities rather than with absolutes.

How good the poll is is primarily based on how large their sample was, how representative their sample was, and whether they asked the questions in a way that led people to give their true answer.

I messed up my reply, I meant "bigger margins of error". I'm already aware of what probabilities are about, don't worry.

There was a previous tweet in this thread where a pollster had something like 52/48 for Trump but then said that with the margin of error, Biden could still win. My point was that if the margin of error is big enough that a 4 point difference could still mean the losing side can win, then it's meaningless.

Either the sample is not good enough or the methodology is flawed but in both cases, the poll shouldn't be published.

 

[Pressure]

I think I’m going to change my name btw. Something way too self important.

Call
Me
Pressure
On
The
DNC

 

[T’Challa’s Heir]

538 predicted heading into the final 2016 weekend that based on all the polling, Clinton had a 70% chance of victory. That's not a "fairly comfortable win", that's closer to 50-50 than to 100%. Imagine if a doctor told you before a surgery that you had a 30% chance of death, would you feel comfortable?

If you want to put it a different way, the polling average had Clinton up in the popular vote by 3.9%. In reality, she won by 2.1%. So the polls were off by 1.8 percentage points. Are we really going to claim, "Damn, these polls are worthless!" because they were off by just 1.8%?



538 predicted heading into the final 2020 weekend that Biden had a 90% chance of winning - so Trump still had 10% chance. Biden did indeed win by the skin of his teeth, but he had three different buffer states - Trump would have had to come out on top in Arizona AND Georgia AND Wisconsin in order to beat him. It feels like the race was indeed closer than the polls indicated, but was it that big a miss? If they were really predicting a blowout, they would have given Trump a 1-2% chance, not a 10% chance.

On the state-by-state forecast, literally the only states that 538 missed on in 2020 were North Carolina and Florida. They got 48 out of 50 states correct. That's 96%.

In terms of the popular vote, the final voting average predicted that Biden would win the popular vote by 7.9%. Instead, he only won by 4.4%. So that was a 3.5% miss. That's not a good outcome at all, the polling companies want to do better than that. But they were trying to predict turnout in a literally unprecedented situation - during a pandemic, with massive mail-in voting, with a fanatical Trump base that is like nothing we've ever seen. And despite all those unique factors, they still only missed by 3.5%.

Did the polls have Hillary Clinton winning in 2016? I’m aware that this is kind of moving the goal posts from what I initially said. But are you willing to acknowledge Trump’s win in 2016 involved every poll being wrong in his favor? If so, how much stock should we be putting polls.

Same deal for 2020. The polls had Biden up
By a ton, then he barely won across the crucial states.

Then 2022 had the GOP w/ a red wave that turned into a super slim majority.

1) how many polls need to miss before we re-evaluate

2) how come bad polls only seem to be held against Dems

 

[Professor Emeritus]

There was a previous tweet in this thread where a pollster had something like 52/48 for Trump but then said that with the margin of error, Biden could still win. My point was that if the margin of error is big enough that a 4 point difference could still mean the losing side can win, then it's meaningless.

Either the sample is not good enough or the methodology is flawed but in both cases, the poll shouldn't be published.

So you think that every time a poll shows a close race, they shouldn't publish the poll?

Having a margin of error larger than the margin of the poll doesn't make the poll useless, it all comes down to probabilities. The margin of error gives you the 95% confidence level. So if the race is so close that it is small than that level, you know that you have less than 95% confidence that a repeat poll would produce the same winner, but it still might be 80%, 70%, etc.....which is more information than nothing.

 

[Pressure]

I ain’t gonna lie. A guy said he was voting for Trump because we have to drill more oil and limit illegal immigration.

When shown that Trump forced republicans to abandon their own immigration bill and the US is producing more oil under Biden he just said it wasn’t true and we don’t know what we’re talking about.

Narrator:

He doesn’t care about those things

 

[Hood Critic]

I think I’m going to change my name btw. Something way too self important.

Call
Me
Pressure
On
The
DNC

Careful, you starting to sound like Negro Nap

 

[Pressure]

Careful, you starting to sound like Negro Nap

I was just following the Dankster.

Man said he changed his name so we would take him seriously

Breh done had a whole Kanye 3 album release of name changes.

College dropout —the dankster
Late registration— Rhakim
Graduation— professor emeritus

 

[the cac mamba]

 

[the cac mamba]

I ain’t gonna lie. A guy said he was voting for Trump because we have to drill more oil and limit illegal immigration.

When shown that Trump forced republicans to abandon their own immigration bill and the US is producing more oil under Biden he just said it wasn’t true and we don’t know what we’re talking about.

Narrator:

He doesn’t care about those things

the Dems are weak on messaging, on both issues they pander to the wrong people

Biden should obviously be bragging about oil production, but he's worried about... what? are we really worried about the kind of voters that vote based on climate change? those people hate trump they aren't staying home in november. and whether biden brags or not...
he's still beating trump's oil production biden acting ashamed of it isn't, like, offsetting environmental impact

and the idiocy of pandering on illegal immigration, i've voiced enough opinion on that

 

[Robbie3000]

I bet the conservative court never saw this coming when they ruled on Citizens United case.

 

[Json]

the Dems are weak on messaging, on both issues they pander to the wrong people

Biden should obviously be bragging about oil production, but he's worried about... what? are we really worried about the kind of voters that vote based on climate change? those people hate trump they aren't staying home in november. and whether biden brags or not...
he's still beating trump's oil production biden acting ashamed of it isn't, like, offsetting environmental impact

and the idiocy of pandering on illegal immigration, i've voiced enough opinion on that

Nobody is gaining or losing votes on oil proeduction. I doubt 90% of people even know someone who works on a oil platform or derrick.


Same reason bush jr didn’t lose because of the Iraq war lie. Most people don’t have family in the military to care.

The reason we have so few independents is people either vote for their team or they don’t vote. The ” sportification“ of politics means people will call a foul if it helps their team and say it’s cheating if it goes against them. The only thing unique about Trump is that people are being forced to vote against him because his cult of personality won’t let him rot alone at Mar-a-Largo and move on.

 

[the cac mamba]

Nobody is gaining or losing votes on oil proeduction. I doubt 90% of people even know someone who works on a oil platform or derrick

no, of course not. but votes are being gained and lost on the dumbed down version, which is "TRUMP = CHEAP GAS. HIGH GAS PRICES BAD"

if it doesn't matter, why is the GOP running on "unleashing american energy"? i dont understand why the fukk Biden can't just show some balls, and blunt this talking point. what exactly is your reason for why biden shouldnt do this?