• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

PoliGAF 2016 |OT11| Well this is exciting

Status
Not open for further replies.
You pick random words out of a hat and say them?


Seriously, I don't think it's so hard. Trump's very predictable in many ways.

Ask me a debate question and I'll answer it like Trump!

Look, you want me to ask you a question? Okay, but we're not asking the right questions. Okay. We're not. We have so many questions. You know, just questions that people are asking. And, look, what we have to do is ask them. Okay. We have to ask them. We have to get tough when we ask the questions. We have to get tough. And President Obama and Crooked Hillary don't have the strength or stamina to ask the right questions. So, the question doesn't matter, folks. Doesn't matter. We're going to Make America Great again, okay. We're going to do it. We have no choice.
 
I feel this would go over very poorly.

Yep. anyone who cares about policy isn't going to be seriously considering Trump in the first place.

Trump supporters are perfectly aware he has no grasp of policy details and it doesn't matter. These people are voting based on emotional appeals to fear and bigotry.

You cannot counter an emotional appeal with a rational argument, you'll lose every time.

What clinton NEEDS to do- and is the only thing that really worked in the primaries- is hammer trump on things that destroy the emotional appeal of his campaign.

Hit him over and over again on how his business- i.e. Trump University- has harmed hard working Americans by scamming them out of their savings on bullshit degrees. Hammer him on how Trump businesses have cost small business owners by screwing them on what they're owed.

Getting lost in the weeds on policy will absolutely not work. Calling him out on being a racist will not work. But destroying the argument that he's looking out for "hard working white people" ABSOLUTELY will.
 
Morgan Freeman says Sam Wang is right you know.

I'm glad we've all concluded Trash Nate is the equivalent of Karen from mean girls.

Why are there so many field offices in NH?
 
He has the attention spam of a 18 month baby, of course he's not going to do a proper prep.

Ailes must've bailed out pretty quickly.

Either way, unless he's pretty much won the debate if the media is going with LOW EXPECTATIONS, HE'S A BUSINESSMAN FFS IT'S NOT LIKE THEY'RE RUNNING FOR THE SAME JOB FOLKS.
 
Look, you want me to ask you a question? Okay, but we're not asking the right questions. Okay. We're not. We have so many questions. You know, just questions that people are asking. And, look, what we have to do is ask them. Okay. We have to ask them. We have to get tough when we ask the questions. We have to get tough. And President Obama and Crooked Hillary don't have the strength or stamina to ask the right questions. So, the question doesn't matter, folks. Doesn't matter. We're going to Make America Great again, okay. We're going to do it. We have no choice.

It's so not hard to mimic Trump.


Honestly, the only thing the Trump impersonator needs to prep for is to pay really good attention to Trump because in the middle of these word salads he can sometimes say something awful and she can throw it back at him.
 

B-Dubs

No Scrubs
Look, you want me to ask you a question? Okay, but we're not asking the right questions. Okay. We're not. We have so many questions. You know, just questions that people are asking. And, look, what we have to do is ask them. Okay. We have to ask them. We have to get tough when we ask the questions. We have to get tough. And President Obama and Crooked Hillary don't have the strength or stamina to ask the right questions. So, the question doesn't matter, folks. Doesn't matter. We're going to Make America Great again, okay. We're going to do it. We have no choice.

Gotta trail off and change the subject, maybe mention how his buddy Carl is good at answering questions then tell a story about how the two went fishing once.
 

Boke1879

Member
Yep. anyone who cares about policy isn't going to be seriously considering Trump in the first place.

Trump supporters are perfectly aware he has no grasp of policy details and it doesn't matter. These people are voting based on emotional appeals to fear and bigotry.

You cannot counter an emotional appeal with a rational argument, you'll lose every time.

What clinton NEEDS to do- and is the only thing that really worked in the primaries- is hammer trump on things that destroy the emotional appeal of his campaign.

Hit him over and over again on how his business- i.e. Trump University- has harmed hard working Americans by scamming them out of their savings on bullshit degrees. Hammer him on how Trump businesses have cost small business owners by screwing them on what they're owed.

Getting lost in the weeds on policy will absolutely not work.

This is where I disagree. No on is trying to win over Donald Trump supporters. They are a lost cause and don't give a shit about facts.

You're trying to win over the undecideds and people who claim they are voting 3rd party.

She can't attack Trump every single time he opens his mouth. Like Mamba said. Everything he says will be some form of BS. You hit him on outrageous shit like "stop and frisk" or building the wall etc.

She then counters that with what she would do and outlines a policy. That's something Donald can't do and it'll be evident. When he has to follow up which the moderator will ask him to do he won't be able to because he doesn't really have a plan.
 
Yep. anyone who cares about policy isn't going to be seriously considering Trump in the first place.

Trump supporters are perfectly aware he has no grasp of policy details and it doesn't matter. These people are voting based on emotional appeals to fear and bigotry.

You cannot counter an emotional appeal with a rational argument, you'll lose every time.

What clinton NEEDS to do- and is the only thing that really worked in the primaries- is hammer trump on things that destroy the emotional appeal of his campaign.

Hit him over and over again on how his business- i.e. Trump University- has harmed hard working Americans by scamming them out of their savings on bullshit degrees. Hammer him on how Trump businesses have cost small business owners by screwing them on what they're owed.

Getting lost in the weeds on policy will absolutely not work. But destroying the argument that he's looking out for hard working white people ABSOLUTELY will.

Hillary doesn't have to convince any Trump supporters of anything.

She needs to convince the undecideds and the soft supporters to get their ass out and vote for her on election day.

Showing herself to be intelligent and rational above a buffoon is the best way.

Now, that's not to say she shouldn't hit him for the things you mentioned. I absolutely think she should. But there will be questions where that opportunity won't come up. If they're asked about Iran and the nuclear deal, there's no real chance to attack Trump's history or racism there. You have to demonstrate policy.

She's at the point now where she's convinced enough people to not vote for Trump. Now she needs to close the deal and bring the soft supporters to her.
 
It's so not hard to mimic Trump.


Honestly, the only thing the Trump impersonator needs to prep for is to pay really good attention to Trump because in the middle of these word salads he can sometimes say something awful and she can throw it back at him.
Look, what you're talking about is political correctness okay? Political correctness. And we can't be politically correct anymore. We can't. And, I say what I mean, okay? I do. And, sometimes, it's hard to hear. It's hard to hear folks. Just like it's hard to hear English in our schools anymore. But, we have to make American great again. We have to. We have no choice. EXTREME. CHOICE.
Gotta trail off and change the subject, maybe mention how his buddy Carl is good at answering questions then tell a story about how the two went fishing once.

Carl is a winner. Bigly. He wins bigly. And, also, let me read to you a poem about a snake for a few minutes. It sums it up. Okay. It sums it up.
 
Y'all should watch Designated Survivor. It was good.

And not Adam level good. Actually good. Fiance even liked it.

I wonder/fear how they will stretch the premise.

I posted about it a little while back. It's pretty solid but the most recent episode on Gender is incredible. Highly recommended.

Yeah, that episode was very good. People should listen to it. That anecdote about the 2008 NH primary blew me away. I lived through that and never even considered it.
 
How bold do you have to be

Cs_iLt0WIAAkjKM.jpg:large
 

Holmes

Member
Morgan Freeman says Sam Wang is right you know.

I'm glad we've all concluded Trash Nate is the equivalent of Karen from mean girls.

Why are there so many field offices in NH?
Maybe they thought Trump would appeal there based on how well he did in the primary and his businessman, tell it like it is attitude would resonate in such a white state. Turns out the state is pretty educated and is rejecting him, and it's the state next door that they should focus a bit more on. Well, one congressional district anyway.

Also there's a competitive Senate race and a former competitive House race, but Guinta has since kneecapped the Republicans' chances there.
 
Talking points, folks:

GOP memo: Trump won't be as polished as Clinton in debate


GOP officials are working to lower expectations for Donald Trump ahead of his faceoff with Hillary Clinton in their first presidential debate next week.

Republican National Committee (RNC) chief strategist Sean Spicer said in a memo Thursday that Trump won't come across as polished as the former secretary of State in their Monday debate.

"Trump hasn’t been running for president for 24 years, he’s spent his career as a successful businessman," Spicer said in the memo circulated to reporters Thursday, which noted a number of political and media figures describing Clinton as a strong debater.

"Few are expecting the same level of polish from a verbal gunslinger whose rhetorical strength is speaking to the heart — and the gut — of the American people," he added
.

Trump and Clinton have sought to manage expectations ahead of their first bout, which is expected to draw a huge television audience.

Trump's campaign circulated a similar memo to GOP allies this week emphasizing that Trump has never run for political office before while Clinton has "extensive experience debating."

"For Clinton, high expectations stem from ample experience," the RNC strategist argued in his memo, remarking on Clinton's "years sharpening her debate reflexes and beefing up on public policy."

"Donald Trump is new to the format," he continued. Aside from the many GOP debates in which Trump participated, Spicer argued that "Trump’s lack of formal, political, one-on-one debate experience gives Clinton a significant advantage."
 
D

Deleted member 231381

Unconfirmed Member
Wang is only right that a prediction should stay stable if the fundamentals stay unchanged. If the fundamentals change, the prediction should also - otherwise how are you even generating the prediction? So it's not that 'Nate doesn't understand statistics' or whatever, it's that Wang and Nate have divergent opinions of how much different things effect the result. Wang thinks elections boil down to a few simple inputs. Nate thinks elections are affected by a wide variety of inputs. That's all. We've also discussed this difference at least once every 5 pages for like 11 straight PoliGAF threads, so we could maybe drop it now?

Y'all are getting straight torn by Nate. Yes, his punditry is poor, so ignore it, but this is bitchcrackers syndrome.
 

Emarv

Member
Wang has explained, though. Any forecast that is volatile to anything but obvious major shifts isn't very good.

Basically, the polling has been more volatile this election and Nate's model is being exposed as a result.

Wang's right. What he's arguing is that Nate's model apparently isn't that useful for predictions way in advance because it's not really predicting the future, it's interpreting the present.

Now, if you mean on the technical details, I haven't noticed him making one. But I will say this, I think the big flaw is Nate used a loess regression which turns in aggressive results. For one, I think using such a regression is incorrect in this situation and for another, his inputs into the regression are completely wrong.

Wang is an actual professor that has advanced knowledge of statistics. Silver is a very smart guy who learned stats but doesn't understand stats (if you understand what I'm saying) and often misapplies them (not just in politics, he does it in sports too!).

He was lucky that 4 years ago the model wasn't volatile largely because the incumbent input was there and the polling was better.
I feel bad always defending Silver, but I call bull on the notion that just because Wang teaches and is more decorated, he's more of an authority. Wang's background is physics, Molecular Biology and Neuroscience. Now I don't know much about Wang's background, but I did go to school for Molecular Biology and many of my friends went on to get their PhD's in those subjects. From what I gather, there's not a ton of advance probabilistics involved in the early stuff.

Now based on a cursory glance at Wang's background, the dude seems like a true genius and savant but I'd venture to guess he kind of just did election results in his spare time as it doesn't have much to do with his real scientific work. Now, while people like Silver, Enten and Cohn might not be prodigies like Wang, their backgrounds are solid and their passions are undisputed, in addition to the field being so much still in its infancy.

I just think it's silly to dismiss Silver et al as some lucky hobbyist compared to Wang. (Again, I also have serious questions about his model and demeanor this cycle)
 
Good responses across the board. Here's where I'm at:

This is where I disagree. No on is trying to win over Donald Trump supporters. They are a lost cause and don't give a shit about facts.

You're trying to win over the undecideds and people who claim they are voting 3rd party.

As far as I'm concerned, anyone who is still "undecided" at this stage in the game isn't really much different than a Trump supporter. Those people aren't "undecided." there's no rational basis for "undecided" between Clinton and Trump, Trump is CLEARLY, demonstrably unqualified. Those people ARE however still vulnerable to emotional based appeals on fear.

Not every Trump supporter is a lost cause, foaming at the mouth racist. A lot of these people are rural white voters without degrees who have never seen a minority outside of television in their lives. They're single issue voting on guns, religion, abortion, etc. A lot of these people can be convinced that Trump is a fraud. If it gets them to simply stay home entirely rather than pulling the opposite lever for clinton, I'm all for it.

Right now I don't think there's anything that would cause Clinton to lose the presidency, the math simply isn't there for the Trump campaign. I DO think that House and Senate races are critical- and turning out support for clinton AND depressing support for Trump is absolutely vital here.

She can't attack Trump every single time he opens his mouth. Like Mamba said. Everything he says will be some form of BS. You hit him on outrageous shit like "stop and frisk" or building the wall etc.

Attacking Trump if she does it at all should be done while keeping in mind that an emotional appeal is the center of a campaign. Countering a "Make america great again" word salad with a 60 second diatribe on policy is useless. Didn't work in the primaries, and won't work now. Trump University though? Absolutely devastating to the Trump campaign and it should be hammered home as much as possible.

Now, if we're talking a PRO CLINTON argument and not an anti-Trump argument? Go policy full blast. THATS the area where she needs to give the argument that she's the most qualified- and needs to give a lot of reasons policy based and emotionally based as to why she's more than just "not trump".
 

sc0la

Unconfirmed Member
Donald Trump beat a record 16 other candidates in numerous debates. He beat the Bush Dynasty and the up and coming young crop that the GOP has glorified for years.

Hillary beat a 200 year old socialist from vermont. Barely.


Trump's expectations should be through the roof! Nice try, GOP!
Bernie the Old is not a day older than 175, comrade.
 
D

Deleted member 231381

Unconfirmed Member
Excluding Obama, are there any historic examples in which the ground game between parties was drastically lopsided?

You'd have to go quite far back in history. Ground game died for a bit in the 80s and 90s when the dominant academic thought was that national media was the important thing.
 
Wang is only right that a prediction should stay stable if the fundamentals stay unchanged. If the fundamentals change, the prediction should also - otherwise how are you even generating the prediction? So it's not that 'Nate doesn't understand statistics' or whatever, it's that Wang and Nate have divergent opinions of how much different things effect the result. Wang thinks elections boil down to a few simple inputs. Nate thinks elections are affected by a wide variety of inputs. That's all. We've also discussed this difference at least once every 5 pages for like 11 straight PoliGAF threads, so we could maybe drop it now?

Y'all are getting straight torn by Nate. Yes, his punditry is poor, so ignore it, but this is bitchcrackers syndrome.

He made 3 models, each more volatile than the last. It seems like he did so to fuel the bad punditry of his site. This isn't bitchcrackers.
 

There's a buried lede here

The one study tactic that they both share is a review of their opponent's past performances. But while aides say Clinton has been actively reviewing Trump's work, senior level Trump sources say he was given an iPad loaded with footage of old Clinton debates though it's unclear how much time he has spent watching them.

We might need to adjust the android v. apple twitter bot
 
Wang is only right that a prediction should stay stable if the fundamentals stay unchanged. If the fundamentals change, the prediction should also - otherwise how are you even generating the prediction? So it's not that 'Nate doesn't understand statistics' or whatever, it's that Wang and Nate have divergent opinions of how much different things effect the result. Wang thinks elections boil down to a few simple inputs. Nate thinks elections are affected by a wide variety of inputs. That's all. We've also discussed this difference at least once every 5 pages for like 11 straight PoliGAF threads, so we could maybe drop it now?

Y'all are getting straight torn by Nate. Yes, his punditry is poor, so ignore it, but this is bitchcrackers syndrome.

No, Wang is objectively right in that the models should move slower without a major obvious shift. Like, every statistician in the world will agree with him, I'm pretty sure.

If you need proof, just look at how Nate's Now-Cast and Polls-Plus are nearly the same right now. That makes no sense at all if the Polls-Plus is working as a full prediction rather than not. Yes, they're supposed to converge by election night but this is too soon!

But I've explained in the past that Nate doesn't actually understand statistics and it has nothing to do with his election coverage. His mea culpa article during the World Cup proved he did not understand.

Short story: He wrote a whole article apologizing for his predicting Brazil to beat Germany when Germany destroyed Brazil and basically said he had to re-do the model. In other words, he argued that an outlier proved to him his model was wrong. []This literally demonstrates he doesn't understand statistics[/i].
 

It was never going away even when she's President. Matter of fact when she IS President expect the House at the very least try and conjure up all sorts of shit on her.

No, but it would go away on November 9th if Hillary Clinton were to lose the election. It's not about "finding out the truth", it's about sinking a Clinton campaign or presidency.
 
Like, I don't know why you're all complaining about Nate and stuff.

Tyler has the best model. You just, you know, keep changing it after the fact until it's right. And something about facebanking.
 
D

Deleted member 231381

Unconfirmed Member
He made 3 models, each more volatile than the last. It seems like he did so to fuel the bad punditry of his site. This isn't bitchcrackers.

One of his model is a Nowcast. That doesn't compete with either of his other two models or Wang's model. It has an entirely different purpose and should be judged by entirely different standards. His other models are volatile because he includes more inputs. It's just a statistical truth that as you increase the number of inputs, you increase the accuracy at the cost of reducing the precision (i.e. increasing the volatility). There's no way around that. Accordingly, Nate has provided a high-input and lo-input version of his model. That's a common statistical practice - like people using both R squared and adjusted R squared models.

None of this is somehow poor practice or even out of the ordinary. Hence, bitchcrackers syndrome.
 
https://twitter.com/davidplouffe/status/778974010642051072

In Trump's mind he wins by merely showing up. The ratings are going to be big and he'll likely be talked about the most no matter what happens. He's a showman, that's all. That being said I'm feeling more confident that he won't attend the next debate. Hillary Clinton is a very good debater, and arguably the best debater out of the last 16 years of national elections. I think the contrast between the two candidates is going to be so stark that it'll be very hard for most of the media to play the "both sides" game. Yes some of that will happen regardless, and Trump being entertaining/funny will sway some folks...but overall you're going to see a professional and a clown.

If the media reaction is as bad as I think it can be, and if polls are as unanimous as I expect they'll be, Trump's only recourse will be to declare the process rigged and refuse to participate again unless his demands are met. In his mind he wins via default, so any perception of a loss=someone is cheating him.

The biggest danger I see for Clinton is that Trump's ceiling is so low, expectations wise. A lot of people are going to tune in to be entertained, and Trump will easily provide that. And his debate and speaking style makes it easy for him to repeat various things for 30-90 seconds effectively. If the question is on Syria for instance he can say "Syria is a mess, it was fine before Obama and Hillary ruined it. We're losing there, Russia is doing xyz. When I'm president we're going to have a winning strategy, ok. We're going to lead." He could potentially get away with that depending on what the media has to say, and depending on voter's reactions to Hillary's answers (does she dodge responsibility, does she refuse to criticize the president, etc).
 

bachikarn

Member
I feel bad always defending Silver, but I call bull on the notion that just because Wang teaches and is more decorated, he's more of an authority. Wang's background is physics, Molecular Biology and Neuroscience. Now I don't know much about Wang's background, but I did go to school for Molecular Biology and many of my friends went on to get their PhD's in those subjects. From what I gather, there's not a ton of advance probabilistics involved in the early stuff.

Now based on a cursory glance at Wang's background, the dude seems like a true genius and savant but I'd venture to guess he kind of just did election results in his spare time as it doesn't have much to do with his real scientific work. Now, while people like Silver, Enten and Cohn might not be prodigies like Wang, their backgrounds are solid and their passions are undisputed, in addition to the field being so much still in its infancy.

I just think it's silly to dismiss Silver et al as some lucky hobbyist compared to Wang. (Again, I also have serious questions about his model and demeanor this cycle)

Pretty sure Wang's research has consisted of statistical modeling. But yeah Wang has kind of done this as a hobby, but he has been doing it since 2004. Also, his hobby has actually led to a published paper on Gerry meandering. So this is definitely much more than a quick thing he has done on the side.

I'd trust Wang on the actual mathematics behind the modeling, which I believe is what Wang is critiquing Nate on. Honestly what Wang is saying to Nate is probably no different than something a journal reviewer would say to someone trying to get a paper published. For someone who is basically the face of poll based political predictions, Nate should be held to a high standard. That's not to say he sucks, but his model can be improved.
 

NeoXChaos

Member
https://twitter.com/davidplouffe/status/778974010642051072

In Trump's mind he wins by merely showing up. The ratings are going to be big and he'll likely be talked about the most no matter what happens. He's a showman, that's all. That being said I'm feeling more confident that he won't attend the next debate. Hillary Clinton is a very good debater, and arguably the best debater out of the last 16 years of national elections. I think the contrast between the two candidates is going to be so stark that it'll be very hard for most of the media to play the "both sides" game. Yes some of that will happen regardless, and Trump being entertaining/funny will sway some folks...but overall you're going to see a professional and a clown.

If the media reaction is as bad as I think it can be, and if polls are as unanimous as I expect they'll be, Trump's only recourse will be to declare the process rigged and refuse to participate again unless his demands are met. In his mind he wins via default, so any perception of a loss=someone is cheating him.

The biggest danger I see for Clinton is that Trump's ceiling is so low, expectations wise. A lot of people are going to tune in to be entertained, and Trump will easily provide that. And his debate and speaking style makes it easy for him to repeat various things for 30-90 seconds effectively. If the question is on Syria for instance he can say "Syria is a mess, it was fine before Obama and Hillary ruined it. We're losing there, Russia is doing xyz. When I'm president we're going to have a winning strategy, ok. We're going to lead." He could potentially get away with that depending on what the media has to say, and depending on voter's reactions to Hillary's answers (does she dodge responsibility, does she refuse to criticize the president, etc).

That's why he needs to be asked for specifics and not allow him to word salad his way out of answers. You know with numbers.
 
I feel bad always defending Silver, but I call bull on the notion that just because Wang teaches and is more decorated, he's more of an authority. Wang's background is physics, Molecular Biology and Neuroscience. Now I don't know much about Wang's background, but I did go to school for Molecular Biology and many of my friends went on to get their PhD's in those subjects. From what I gather, there's not a ton of advance probabilistics involved in the early stuff.

Now based on a cursory glance at Wang's background, the dude seems like a true genius and savant but I'd venture to guess he kind of just did election results in his spare time as it doesn't have much to do with his real scientific work. Now, while people like Silver, Enten and Cohn might not be prodigies like Wang, their backgrounds are solid and their passions are undisputed, in addition to the field being so much still in its infancy.

I just think it's silly to dismiss Silver et al as some lucky hobbyist compared to Wang. (Again, I also have serious questions about his model and demeanor this cycle)

I don't think Silver is generally lucky (just lucky that the model 4 years ago wasn't exposed for its flaws). He's obviously smart and he's obviously on the right track in building a forecasting model. But there's statistical reasoning problems and things I wouldn't expect someone like Wang to mess up on.

And I really like Cohn as well, and I've disagreed here and there with him (for instance his argument regarding missing white voters earlier). But unlike Silver, Cohn seems to be humble and stuff.

My main beef with Silver, again, has been his movement into punditry (after hating on it) and his clickbait bullshit. That, and 538 has been a massive disappointment in terms of sports analysis (there's so much better out there and I was hoping his site would be great for it).


Anyway, Silver having a model is good. In fact, it wouldn't be half-bad to aggregate all the models too! So the more inputs, possibly the better.

I just find it odd that some people here are questioning people who think his model is weaker than previously thought. I, myself, argued months ago that the loess regression makes no sense and is misapplied in the model and wasn't aware of it 4 years ago. It has nothing to do with me wanting to pick a model that is better. Votematic isn't doing great for Hillary right now but I think it's a bit better than Silver's, from what I can tell. The inputs have been shitty, though.


If nothing major happens and Hillary comfortably wins the election (300+ EV and 4+ PV), can we then agree that Silver's model was pretty poor this election cycle because in mid-September it was well off the mark?
 
One of his model is a Nowcast. That doesn't compete with either of his other two models or Wang's model. It has an entirely different purpose and should be judged by entirely different standards. His other models are volatile because he includes more inputs. It's just a statistical truth that as you increase the number of inputs, you increase the accuracy at the cost of reducing the precision (i.e. increasing the volatility). There's no way around that. Accordingly, Nate has provided a high-input and lo-input version of his model. That's a common statistical practice - like people used both R squared and adjusted R squared models.

None of this is somehow poor practice or even out of the ordinary. Hence, bitchcrackers syndrome.

When I see someone in his position throw out multiple models, I make one of a few assumptions:

1. He's incompetent.
2. He has no confidence in his abilities.

...or the one I think is most likely,
3. He wants to give himself options for pundry. If he wakes up and wants to write a super reactive hot take, he can grab from the Nowcast. If he wants to write a far looking contrarian piece, he can grab from Polls Plus or Polls only. It's a pretty good way to use data to run a punditry site.

It might be common for people working on campaigns to run multiple models, but could you point me to another polling aggregate company that does so? Upshot, PEC, Sabato... I can't think of anyone else that does so. I don't think it's crazy to say that 538 signed its deal with ESPN to innovate on how statistics are used in political coverage. The whole site seems to be built on the idea of statistics-based punditry.

I think it's perfectly acceptable to call that out and question how that aim might affect his model.
 
Okay, regarding the 3 models.


The Now-Cast model is garbage model meant for website clicks. Nothing more, nothing less.


Now, I actually DO like that he has a polls-only model. He's testing his actual model against it.

Think of the polls-only model as a control group in a scientific study. He's comparing his model to a polls-only model to see how it stacks up. That's good application!

The problem is that his polls-plus model has been very volatile and very re-active to the polls, so what is he going to learn when the election is over?

If the election ends up being super close, he won't know if his polls-plus did better or worse than the polls-only.

And if the election isn't close, will he be able to figure out what was causing the bad volatility (other than polls, of course)? Probably not.
 
The Hecks sounds like a nice family.

https://www.buzzfeed.com/christophe...acist-commen?utm_term=.dqr4R10AjG#.lyRE28De0B

The son of Republican Rep. Joe Heck made several sharply racist and sexist comments and upvoted inflammatory memes on Reddit, according to a BuzzFeed News review of his account.

Rep. Heck, who is running for US Senate in Nevada, issued an apology on Thursday after BuzzFeed News notified his campaign of his son’s actions. His son, Joseph Heck III, who goes by Joey, also apologized, and said he was entering counseling. Joey Heck’s comments have since been removed from the website.

In 2013, BuzzFeed News previously reported on Joey Heck’s online comments, including saying in a tweet that President Obama was only good at “spear chucking and rock skipping.” Rep. Heck issued an apology for his son’s tweets at the time. Joey Heck is now 19 and in college.

In his statement, Joey Heck wrote that his parents “raised me to be responsible and respectful and to love my neighbor.”

“I’ve let my family down and failed to show a proper level of respect for others,” his statement said. “As a 15-year-old, it was easy to get caught up in and manipulated by social media. As an 19-year-old, I have no excuse. I know better. This is my mistake. I own it and I sincerely regret it. I am truly sorry.”

The Heck family, in a statement to BuzzFeed News, said, “While it is difficult to convey the magnitude of our disappointment in our son Joey’s behavior as well as our concern by the content of his social media account, our love for him does not waver. Our family accepts his apology and fully supports his decision to seek counseling. We will stand by Joey as he works through these issues as any parent would do, and his sisters are fully committed to provide support and guidance.”

The comments and upvotes on Reddit include graphic language.

Using the username Joeyj424, Joey Heck posted inflammatory and misogynistic comments on Reddit during the past year. As recently as two months ago, a message (called “flair” by Reddit users) appeared next to his username on all his comments that said, “I wish Hillary would violate me as much as she did federal law.”

Joey Heck also commented more than once on the rape allegations against Bill Cosby. In one post a year ago, he commented on a photo comparing the comedian, accused by dozens of women of sexual assault and by some of drugging them, to champion boxers. He wrote, “Up there with the best of them, good for you Billy”. In another comment on a New York Magazine article cover that featured 35 women who accused Cosby of assaulting them, “Number 9 didn’t even need the drugs her eyes are already all sorts of fucked up.”

Heck’s son also upvoted, or endorsed, a number of racist and anti-Semitic pictures and memes. Reddit’s algorithm uses upvotes to determine what gets promoted on the site. One image he upvoted was a racist depiction of a group of black children with the caption, “God made the little niggers, He made them in the night, He made them in a hurry, and forgot to make them white!”

He also upvoted a meme of a white person with a disability in a wheelchair, which read, “Just another retard who thinks hes a nigger I hate when white kids throw up gang signs”. In another picture he upvoted, a penguin is shown in front of an oven expressing happiness that “my jews are almost done!”

A year ago, Joey Heck commented on a picture of a sex toy in a basket of children’s toys, saying, “I’m deeply ashamed of you for this. You’re a failure as a father. 3 years old and that’s all she can take?”

In a number of comments, Heck’s son refers to people with disabilities by the slur “potatoes.” He called a woman who appeared in a video with her autistic son a “Hot ass potato farmer.”

And in a comment on a meme of a dancing black child wearing a diaper, with the words, “Guess Who’s Getting Confederate Flag Shirts,” Joey Heck posted, “Great, what’s next? Giving them food too? Greedy bastards”. In another comment, Heck wrote that he was “impressed” by a meme of a Muslim man being sodomized by a pig.

Also, apparently Catherine Cortez Masto isn't Mexican????

http://www.reviewjournal.com/politics/election-2016/former-joe-heck-aide-s-tweet-causes-stir

A former campaign aide to GOP Senate candidate U.S. Rep. Joe Heck started a ruckus on Twitter on Thursday, giving an opening to Heck’s opponent, Democratic candidate Catherine Cortez Masto.

“Catherine is about as Mexican is me,” tweeted Tom McAllister, a former political director on Heck’s Senate campaign. “It’s only relevant when applying for scholarships … or running for #NVSEN.”
 

Cybit

FGC Waterboy
From what I understand Iowa was not one of the earliest target states for Clinton. Mobilization there will ramp up in coming weeks. I think the utter lack of any registration machine or GOTV focus on Trump's end will make up for it.

To kinda go back to Cybit's point earlier...if Cruz was the nominee he'd win Iowa easily. Trump is fucking this up in epic fashion.

Sad thing is, Clinton ignored Iowa (relatively speaking) in 2008 as well. >_< She got some buckeye fans or something she's angry against? :D

As for Silver vs Wang argument 10000; you're probably not going to be able to find a winner regardless after the election - considering almost all sites have had Clinton above 50% to win the entire time in their forecast, you can't exactly claim victory. If Trump shits the bed or Clinton shits the bed in the debate and causes a massive swing, everyone's forecasts will shift appropriately before the election to show said massive swing.
 

NeoXChaos

Member
I hope Good Nate does not turn into bad Nate by 2020. Nate Cohn does not get into pundtry. He's down to earth and humble. Silver on the other handle is snarky and prideful atleast in this cycle he's been that way.
 
Sad thing is, Clinton ignored Iowa (relatively speaking) in 2008 as well. >_< She got some buckeye fans or something she's angry against? :D

As for Silver vs Wang argument 10000; you're probably not going to be able to find a winner regardless after the election - considering almost all sites have had Clinton above 50% to win the entire time in their forecast, you can't exactly claim victory. If Trump shits the bed or Clinton shits the bed in the debate and causes a massive swing, everyone's forecasts will shift appropriately before the election to show said massive swing.

No, you can determine a winner under some circumstances.

If the election is about the same as 4 years ago, then Silver's model was wrong in some way. Because you'd have to explain why, with similar polling data as 4 years ago, his model gives Trump's chances a lot higher now than Romney's peak.
 
Okay, regarding the 3 models.


The Now-Cast model is garbage model meant for website clicks. Nothing more, nothing less.


Now, I actually DO like that he has a polls-only model. He's testing his actual model against it.

Think of the polls-only model as a control group in a scientific study. He's comparing his model to a polls-only model to see how it stacks up. That's good application!

The problem is that his polls-plus model has been very volatile and very re-active to the polls, so what is he going to learn when the election is over?

If the election ends up being super close, he won't know if his polls-plus did better or worse than the polls-only.

And if the election isn't close, will he be able to figure out what was causing the bad volatility (other than polls, of course)? Probably not.

Yeah, the problem is that if you throw out the obviously garbo Nowcast, then you are left in a "Never bring two clocks to sea" situation.

If he wants to run other models as a control, he can just retroactively do that with the data after the whole thing is done. I just don't see the value in doing it publicly and from what I can tell, neither do any of his peers.
 
Okay, after seeing the news of the Overwatch porn parody...Think there's going to be a porn parody of this year's election? Who'll play Donald, and who'll play
_Putin
?
 
Status
Not open for further replies.
Top Bottom