Jump to content

2024 Presidential Election Thread - Let's keep the party going.


bolverk

Recommended Posts

1 hour ago, tx 3 putt said:

bwahahahahahahaha

 

 

Does that Russian Morris Code translate into English? I'm asking because the number of times that woman blinks makes it look like she's in distress.

And, to be clear, I hold absolutely zero sympathy for her if she's under any sort of duress.

 

  • Hook 'Em 2
  • Haha 1
Link to comment
Share on other sites

Would you have *EVER* in your life have thought anyone who was a hair's distance away from The People's House would do this sort of thing in punlic?

Remember when Republicans were freaking out because Obama wore a tan suit or they clutched pearls when Clinton answered the "boxers or briefs" question?

Messenger_creation_306BFB47-A001-4873-A155-8F1DD9929489.gif

  • Hook 'Em 2
  • Like 1
  • Rage+1 1
Link to comment
Share on other sites

18 minutes ago, BamaATL said:

Sezler is an outlier, she does her own thing, gets real data, and politics be damned.  Shockingly enough, that simple approach with no bullshit seems to be pretty accurate.  

There are others trying to emulate her, but frankly they get drowned out, partly by partisan polls commissioned and paid for by both parties (though far more from Republicans, who to boot weaponize them), and then slapdicks like Silver who has monetized the polling business.  So the real problem is, you only have a handful of people doing actual real work, and they are completely drowned out by an old asshole trying to make a buck.  The problem with herding is, all the monetized polls figure they can't be hurt if they are in the ballpark as everyone else, so now, the actually polling data is secondary to being near the pack with everyone else.  So the idea of polling aggregation is then completely fucking useless because the data is secondary.  

In short, it's a vicious circle jerk of stupidity.  In fact the circle jerk is probably better time spent, at least at the end there is a tangible result.  

I'm not a statistician but I did have to take a few higher level probability and statistics courses for my undergrad 20 years ago. From what I remember the key to all of it is the quality of input data. That's a very difficult thing to obtain, but the majority of the work and attention to detail should be put into obtain as close to a representative sample as possible without poisoning with bias. If you put garbage data in you're going to get garbage data out. 

I don't follow election polling at all so don't know how they source their data, but it seems Selzer is the closest to quality data input by 20+ years of experience sourcing data from her smallish state. Then the only adjustments needed are adjusting the sample demographics to mirror the registered voter demographics. 

If these other pollsters are taking results they don't feel are quality enough and adjusting them to mirror prior results 4 years ago that's purely invalid, garbage data. That's omitting everything that's happened in the last 4 years. Which in this case is more change in this arena than I believe has ever happened in a 4 year period in the history of the country other than the civil war. 

What we need is about 10 Selzers, one for each of the swing states, that put all their effort and a scientific attention to detail into doing it the right way in their individual states. Right now all we have is one great sample and a shit ton of biased, garbage data it appears. 

  • Hook 'Em 3
Link to comment
Share on other sites

12 minutes ago, Hermanator said:

What we need is about 10 Selzers, one for each of the swing states, that put all their effort and a scientific attention to detail into doing it the right way in their individual states. Right now all we have is one great sample and a shit ton of biased, garbage data it appears. 

Yes she only polls in Iowa so it seems she's built up quite a great data set to draw from. She apparently doesn't even hire out for polls in other states. Just sticks to Iowa.

 

  • Hook 'Em 2
Link to comment
Share on other sites

15 minutes ago, Hermanator said:

I'm not a statistician but I did have to take a few higher level probability and statistics courses for my undergrad 20 years ago. From what I remember the key to all of it is the quality of input data. That's a very difficult thing to obtain, but the majority of the work and attention to detail should be put into obtain as close to a representative sample as possible without poisoning with bias. If you put garbage data in you're going to get garbage data out. 

I don't follow election polling at all so don't know how they source their data, but it seems Selzer is the closest to quality data input by 20+ years of experience sourcing data from her smallish state. Then the only adjustments needed are adjusting the sample demographics to mirror the registered voter demographics. 

If these other pollsters are taking results they don't feel are quality enough and adjusting them to mirror prior results 4 years ago that's purely invalid, garbage data. That's omitting everything that's happened in the last 4 years. Which in this case is more change in this arena than I believe has ever happened in a 4 year period in the history of the country other than the civil war. 

What we need is about 10 Selzers, one for each of the swing states, that put all their effort and a scientific attention to detail into doing it the right way in their individual states. Right now all we have is one great sample and a shit ton of biased, garbage data it appears. 

Well some polls are doing things like assuming a 77% white vote, skewing way whiter than normal, which pulls the number towards Trump.  

Link to comment
Share on other sites

5 minutes ago, Pato del Muerto said:

Well some polls are doing things like assuming a 77% white vote, skewing way whiter than normal, which pulls the number towards Trump.  

That's inventing demographics that don't exist. That's not incompetent, it's flat out nefarious. 

It's been quite the shocking revelation that something as big as the USA presidential election with all the billions of dollars tied up in the whole thing only has a single lady in bumbfuck Iowa that conducts representative sample election polls correctly. 

  • Hook 'Em 1
  • Haha 2
Link to comment
Share on other sites

3 hours ago, Hermanator said:

I'm not a statistician but I did have to take a few higher level probability and statistics courses for my undergrad 20 years ago. From what I remember the key to all of it is the quality of input data. That's a very difficult thing to obtain, but the majority of the work and attention to detail should be put into obtain as close to a representative sample as possible without poisoning with bias. If you put garbage data in you're going to get garbage data out. 

I don't follow election polling at all so don't know how they source their data, but it seems Selzer is the closest to quality data input by 20+ years of experience sourcing data from her smallish state. Then the only adjustments needed are adjusting the sample demographics to mirror the registered voter demographics. 

If these other pollsters are taking results they don't feel are quality enough and adjusting them to mirror prior results 4 years ago that's purely invalid, garbage data. That's omitting everything that's happened in the last 4 years. Which in this case is more change in this arena than I believe has ever happened in a 4 year period in the history of the country other than the civil war. 

What we need is about 10 Selzers, one for each of the swing states, that put all their effort and a scientific attention to detail into doing it the right way in their individual states. Right now all we have is one great sample and a shit ton of biased, garbage data it appears. 

 

6ff06a70-5af2-4c9e-9ed2-4bea8955467c.png

  • Haha 7
  • Rage+1 1
Link to comment
Share on other sites

6 hours ago, BamaATL said:

Sezler is an outlier, she does her own thing, gets real data, and politics be damned.  Shockingly enough, that simple approach with no bullshit seems to be pretty accurate.  

There are others trying to emulate her, but frankly they get drowned out, partly by partisan polls commissioned and paid for by both parties (though far more from Republicans, who to boot weaponize them), and then slapdicks like Silver who has monetized the polling business.  So the real problem is, you only have a handful of people doing actual real work, and they are completely drowned out by an old asshole trying to make a buck.  The problem with herding is, all the monetized polls figure they can't be hurt if they are in the ballpark as everyone else, so now, the actually polling data is secondary to being near the pack with everyone else.  So the idea of polling aggregation is then completely fucking useless because the data is secondary.  

In short, it's a vicious circle jerk of stupidity.  In fact the circle jerk is probably better time spent, at least at the end there is a tangible result.  

I finally got around to reading the Des Moines Register article.

The BIG difference between Selzer is that she got 808 respondents from only 1038 calls. This is an UNBELIEVABLE response rate. NYT/Siena reported 2.5% response rate.

15 - 20% would be amazing, but she's getting 80% ??? Amazing if true. Leaves me wondering what makes her dial sheet so much better than everyone else's.

  • Hook 'Em 4
Link to comment
Share on other sites

7 minutes ago, quigley said:

I finally got around to reading the Des Moines Register article.

The BIG difference between Selzer is that she got 808 respondents from only 1038 calls. This is an UNBELIEVABLE response rate. NYT/Siena reported 2.5% response rate.

15 - 20% would be amazing, but she's getting 80% ??? Amazing if true. Leaves me wondering what makes her dial sheet so much better than everyone else's.

We’re in a position of hoping she’s not faking it.  Unfortunately I’ve known a lot of women in my days, and they ALL fake it. 

  • Haha 1
  • Fuck Around and Find Out 2
Link to comment
Share on other sites

5 hours ago, Hermanator said:

I'm not a statistician but I did have to take a few higher level probability and statistics courses for my undergrad 20 years ago. From what I remember the key to all of it is the quality of input data. That's a very difficult thing to obtain, but the majority of the work and attention to detail should be put into obtain as close to a representative sample as possible without poisoning with bias. If you put garbage data in you're going to get garbage data out. 

I don't follow election polling at all so don't know how they source their data, but it seems Selzer is the closest to quality data input by 20+ years of experience sourcing data from her smallish state. Then the only adjustments needed are adjusting the sample demographics to mirror the registered voter demographics. 

If these other pollsters are taking results they don't feel are quality enough and adjusting them to mirror prior results 4 years ago that's purely invalid, garbage data. That's omitting everything that's happened in the last 4 years. Which in this case is more change in this arena than I believe has ever happened in a 4 year period in the history of the country other than the civil war. 

What we need is about 10 Selzers, one for each of the swing states, that put all their effort and a scientific attention to detail into doing it the right way in their individual states. Right now all we have is one great sample and a shit ton of biased, garbage data it appears. 

100% spot on, and as it turns out if you aggregate bullshit, it turns out, it's bullshit.  

Link to comment
Share on other sites

2 minutes ago, Evil Bill Obrien said:

Off to a very inauspicious start to Election Day, in too big a hurry just fat fingered my computer password 1 too many times and am locked out...great

It's 5 minutes bro (or sometimes 15).

Do some pushups and squats for Kamala.

Link to comment
Share on other sites



×
×
  • Create New...