Polls v. Results 2020 - Underestimation of Republican support in key states:
* WI: +6%
* NC: +1.1%
* FL: +4.2%
* GA: -1.3% (Underestimated Democrat support)
* PA: 0.0%
* OH: +7.2%
* MI: +1.4%
* IA: +6.2%
* AZ: +0.6%
* NV: -0.3%
* TX: +4.5%
When I was at the gym the other day, the TV had on MSNBC and they were calling the Senate rate in Ohio a dead heat. I'm not buying that. J.D. Vance should win be 10%.
It’s insane how combative everyone is in this thread. Rojo doesn’t understand something so he explained his thinking and asked for input. Why is everyone roasting him? He made no arrogant declaration that he knows for a fact that Nate Silver is wrong
I swear half of you graduated college only because of the classmates brave enough to ask “dumb” questions to your benefit, or because lecturers preemptively dumbed things down enough in expectation of all the common misunderstandings people have.
but as soon as we are out of the classroom, anyone who doesn’t understand something and dares to talk about it on the internet is “an idiot with a bad take” wtf. Not everything is an argument. Not every discussion has sides. Chill out
2016 and 2020 polls in NH underestimated Republican support by 4.4%. That's why there's a good chance that the Democrat incumbent polling at 3.4% ahead could lose, resulting in a Republican pickup of the seat.
It’s insane how combative everyone is in this thread. Rojo doesn’t understand something so he explained his thinking and asked for input. Why is everyone roasting him? He made no arrogant declaration that he knows for a fact that Nate Silver is wrong
No, his OP post is literally titled that Nate Silver isn’t smart and doesn’t understand the most basic aspects of probability. Subsequently, he descended into utter bubbaland idiocy by claiming he was more intelligent than Silver because “does he think a man can give birth to a baby?”. It’s not even like he knows Silver’s opinions on the matter, leave alone what that has to do with intelligence. The roasting was well deserved.
Have you just established that the Republicans are right and that there's election fraud? i.e., they tried to "fix" the results after being down by 6.7% in WI, but missed by 0.7%?
Maybe the polls are right, and something is going on with the Republicans.
For some reason, Statistics 101 courses in college tend to make students outside of the Math-smart disciplines believe that they conquered some genius-level subject.
derp derp independent events
derp derp gonna throw around the term Baye's Theorem in normal conversations so I sound smart
Nate Silver holds a Bachelor of Arts in Economics. He wouldn't survive 1 semester at Cal Tech.
For some reason, Statistics 101 courses in college tend to make students outside of the Math-smart disciplines believe that they conquered some genius-level subject.
derp derp independent events
derp derp gonna throw around the term Baye's Theorem in normal conversations so I sound smart
Nate Silver holds a Bachelor of Arts in Economics. He wouldn't survive 1 semester at Cal Tech.
This has been correctly explained a few times here, but here is a more in-depth example.
DD + DR + RD + RR = 100 because these encompass all possible events. Republicans win Senate is 46%, so lets write: 46 = RD + RR Republicans win House is 82%, so lets write: 82 = DR + RR Republicans win both is 46%, so we have: RR = 46. This means RD = ~0, DR = 36, DD = 18.
No contradictions here, just correlation. This is because errors in polling are correlated - Republican bias in a Senate poll often also corresponds to Republican bias in a Democratic poll. On election day the turnout of Republicans/Democrats affects both outcomes and induces correlation (people often vote for the same party for Senate and House).
By the way, at Caltech we have quarters not semesters.
Rojo, there is 100% probability that Nate Silver is smarter than you
Close to 100% that he's better at stats than me although my lone A+ at Princeton was in stats.
But overall intelligence? I'd put it way less than 100%. Does he believe a man can give birth?
PS. The whole reason I started this thread was even if he thinks the Senate is 98% correlated with the House, it would drop his percentage by 1%. He's not dropping it at all. So I guess in his mind it's at least 99% correlated.
OMG rojo. Literally no one thinks a biological male can give birth. You know this, stop trolling.
Whether people believe that gender is different than biology is not a matter of intelligence, it's opinion. All of this merely exists within our social context and our desire to classify things. We made it up, like language or economics. If you disagree that's fine, but it doesn't make someone more or less intelligent, they're just following a different social convention.
Malcom Gladwell doesn't understand why Jeffrey Toobin was fired (spoiler: he was caught playing with his little Toobin on camera in a work meeting), thinks Joe Paterno did nothing wrong, and is a Salazar apologist. "Being smarter than Gladwell" is a low bar to clear.
What did Paterno do?
He enabled a pedo to rape children. At best he made a horrible mistake in not accepting his friend was a monster an. At worst he knew his friend was a monster and helped him out.
A lifetime of being a pretty decent guy will forever have this clid over it...
Polls typically underestimate Republican support in Pennsylvania by about 6%, for example. If Fetterman were ahead on average by around 4%, that wouldn't be anything close to a comfortable lead, would practically be a wash.
Your own table above shows 0.0% underestimate in PA. Please explain.
The fact that there is both 46% chance that the Rs take senate and both house and senate means that there is a 0% chance the Rs take the senate while not taking the house. It’s not complicated.
Rojo doesn’t understand something so he explained his thinking and asked for input. Why is everyone roasting him? He made no arrogant declaration that he knows for a fact that Nate Silver is wrong
You make is sound like rojo's posts are along the lines of "These predictions are weird to me, what's going on here?" or even "Is Nate Silver wrong?"
Instead, we have:
I thought Nate Silver was supposed to be smart. It appears he doesn't understand the most basic aspects of probability
[Silver] thinks there's a 100% chance the GOP takes the hosue [sic] but won't admit it.
Close to 100% that [Silver is] better at stats than me although my lone A+ at Princeton was in stats. But overall intelligence? I'd put it way less than 100%. Does he believe a man can give birth?
If you suggest that someone is an incompetent liar and start bragging about your grades in school, you really have to be right. If not, you deserve to be roasted. Nevermind that this has been rojo's MO for decades...
For some reason, Statistics 101 courses in college tend to make students outside of the Math-smart disciplines believe that they conquered some genius-level subject.
derp derp independent events
derp derp gonna throw around the term Baye's Theorem in normal conversations so I sound smart
Nate Silver holds a Bachelor of Arts in Economics. He wouldn't survive 1 semester at Cal Tech.
This has been correctly explained a few times here, but here is a more in-depth example.
DD + DR + RD + RR = 100 because these encompass all possible events. Republicans win Senate is 46%, so lets write: 46 = RD + RR Republicans win House is 82%, so lets write: 82 = DR + RR Republicans win both is 46%, so we have: RR = 46. This means RD = ~0, DR = 36, DD = 18.
No contradictions here, just correlation. This is because errors in polling are correlated - Republican bias in a Senate poll often also corresponds to Republican bias in a Democratic poll. On election day the turnout of Republicans/Democrats affects both outcomes and induces correlation (people often vote for the same party for Senate and House).
By the way, at Caltech we have quarters not semesters.
Wut?
There is a 46% chance GOP wins the Senate, and 82% that they win the House per Nate Silver. He then goes on to claim that the chance they win both houses is 46%.
This is nonsensical.
For your analysis to work, "RD" (GOP wins senate and Dems win House) has to be 0%.
Which ought to tell you your conclusion is ridiculous.
The fact that there is both 46% chance that the Rs take senate and both house and senate means that there is a 0% chance the Rs take the senate while not taking the house. It’s not complicated.
Which is obviously ridiculous. No chance at all GOP takes the senate but not house?
DD + DR + RD + RR = 100 because these encompass all possible events. Republicans win Senate is 46%, so lets write: 46 = RD + RR Republicans win House is 82%, so lets write: 82 = DR + RR Republicans win both is 46%, so we have: RR = 46. This means RD = ~0, DR = 36, DD = 18.
No contradictions here, just correlation. This is because errors in polling are correlated - Republican bias in a Senate poll often also corresponds to Republican bias in a Democratic poll. On election day the turnout of Republicans/Democrats affects both outcomes and induces correlation (people often vote for the same party for Senate and House).
By the way, at Caltech we have quarters not semesters.
Wut?
There is a 46% chance GOP wins the Senate, and 82% that they win the House per Nate Silver. He then goes on to claim that the chance they win both houses is 46%.
This is nonsensical.
For your analysis to work, "RD" (GOP wins senate and Dems win House) has to be 0%.
Which ought to tell you your conclusion is ridiculous.
there's an implied rounding to the nearest whole number here. Maybe odds of winning the Senate are 46.499999% and odds of winning both are 45.5000000001%. Given how correlated the two events are it doesn't seem at all odd that the outcome of GOP winning Senate but not the House is well under 1%.
There has also been quite a bit in the new about polling error as of recent years. In general there are commentaries about how right leaning citizens may be less likely to be involved in polls and thus the polls are flawed towards left support.
If you consider how much the past 2016 adminstrative team has pushed Americans from thinking elections are legitimate, it seems possible that people wont pick up the phone from a university, wapo, npr, others to say they like a conservative candidate.
this goes back to people pointing out that nate silver's models are only as accurate as the data they input from polling organizations.
