"Devil doll" = another fake headline from Drudge. One might think he just copies the fake headlines from e.g. thesun.co.uk, but he often jazzes them up if the originals are not click-baity enough.
Click bait. The Sun has bleeped out the doll's questionable speech, so it's impossible to judge whether the complaint is worthy or just another example of rampant insanity.
Consider this: If Trump's first year-end approval numbers are the same as Obama's, how are we to assess this in the light of 95% negative press coverage of Trump versus 95% (or more) positive — nay, worshipful — coverage of Obama from the very start of his political career?
1) Trump's poll numbers mean that given even cruelly neutral treatment by the MSM Trump would be more popular than Obama, or Clinton. He would be in Reagan territory at least.
2) The fact that he isn't is strictly due to a press corps devoted to Trump's destruction rather than to truth.
3) The press corps is singularly inept at the business of destruction when the object won't lie down and take it.
Do these numbers mean that the Network will renew Trump for another season, and that he gets to continue appointing judges and rescinding regulations ?
Israel is naming the new rail station on the Tel Aviv to Jersusem line near the Wailing wall the Trump Station. Meanwhile Obama gets new ICBM nuke installations built in Iran named for him, he bring their main developer.
Inga and her fellow leftists will be praying to Moloch nightly that tax reform backfires on the GOP and the economy goes south. They hope for this despite the fact that many millions of Americans would suffer and their own 401-Ks would take a hit. They hate Trump so much they would rather see the country fail than Trump's presidency succeed.
If tax reform does what the GOP and Trump believe it will do, Americans will prosper and the Democrats' predictions of doom will make them look like hysterical liars.
By pumping headlines lefties want to see FiveThirtyEight may be priming a second wave of delusion when the headlines don't square with reality. The Democrats’ Wave Could Turn Into A Flood...Democrats are probably favorites to win the House...No other survey taken in November or December in the year before a midterm has found the majority party in the House down by that much since at least the 1938 cycle..Republicans in worse shape right now than any other majority party at this point in the midterm cycle...Wow.
If tax reform does what the GOP and Trump believe it will do, Americans will prosper and the Democrats' predictions of doom will make them look like hysterical liars.
I like that nobody gets to pontificate over hypotheticals anymore. We're gonna get to see who's right.
"By pumping headlines lefties want to see FiveThirtyEight may be priming a second wave of delusion when the headlines don't square with reality."
Not to worry. LLR Chuck has already clearly demonstrated that he and his lefty operational allies will have no problem shifting to a Hey I Was Actually Right All Along mode after the next set of disappointments.
rehajm: "I like that nobody gets to pontificate over hypotheticals anymore. We're gonna get to see who's right."
Correct. And the entire nation will get to see in real time how much lying the lefties have been engaging in.
You can already see some lefties recognizing the danger of being so incredibly exposed on this issue and are already shifting to - sure, sure, almost everyone gets a benefit but those zillionaires really made out - talking points.
Not Inga of course. She's always the lone dead-ender hanging out on an island not realizing the war has been over for 15 years. These are the cannon-fodder foot soldiers of the left who really don't care what reality looks. They only care about what talking points they are supposed to utter today happen to be.
Which is why history must start anew each day for them, lest the cognitive dissonance overwhelms them.
In general, I don't think people understand how Silver presents probability. It might translate better if it was presented as 7 versus 3 rolls out of 10 instead of percentages.
I look at poll averages. The only single poll I give significant weight are Ann Selzer's.
Leading up to an election, I look for unknowns and volatility. Alabama was impossible to poll. State polls tend to be less reliable, because there tend to be fewer polls.
It will be harder to predict 2018 because we're looking at a bunch of state races. Some will be well polled, but others will not, so the averages, most likely, won't be accurate.
I enjoyed Inga's link. You can compare Trump to former presidents. Trump's popularity will go up with increasing peace and prosperity — like Nixon's did before he was abruptly destroyed. I hadn't realized just how graphically below average Obama was. Clearly a dud. And he started out so high.
The "Get Trump" squad is losing steam and credibility. But their one saving grace is winning the House in 2018. Mind you, this is mostly a historical trend, (opposition party picks up seats in midterms), but the Left will portray it as a vindication of all their Anti-Trump angst and effort. And then they will impeach. And then they will hope it tars him sufficiently to cause him to lose in 2020.
I am surprised with the noise from the Press Trump's rating are as good as they are. I think it's 94% negative press in the MSM for Trump per a Harvard study. Amazing he has not been impeached with that amount of attacks against him.
Found the article, only 93%. http://www.washingtonexaminer.com/byron-york-harvard-study-cnn-nbc-trump-coverage-93-percent-negative/article/2623641
"Probabilistic prediction models mean never having to say you were wrong."
Not strictly true. If 538's models predicted a 30% winning chance for R every election, and time after time, R always won, there are specific ways of quantifying how poor their predictions are with each successive "wrong" call, but I don't think we have enough presidential predictions for it to mean much yet (wasn't Romney about 5%?).
Almost everyone understands odds in horse racing, but, for some reason they don't seem to parse the same numbers in politics.
It might be that Silver presents his odds as percentages which induces confusion with polling percentages. Many people thought 70% chance for Hilary meant she was a "sure thing", which, if that was her opinion poll number, she would have been, but it just meant 2:1 odds, and no sane person would describe 2:1 odds on a horse race as a "sure thing"; odds of 2:1 against a horse doesn't even make him a long shot, it just makes him not the favorite.
Of course it might be that people just like yelling at someone, although why Nate Silver, of all the possible targets, I'm not sure. Much of the left were yelling at him before the election for underestimating Hilary's numbers; some commentators actually accused him of committing fraud in order to drive readers to his site. Now many on the right say "Nate Silver was wrong, never believe anything he says" when he was one of very few forecasters predicting a relatively high chance of a Trump win.
What exactly would you like instead of a probabilistic model? How would it work? I'm at a loss to understand what it would even look like. I started trying to speculate about what you're asking for, but everything I came up with just sounded snarky.
The models assign probabilities to particular outcomes. In the case of an election the possibilities are win/lose/tie. While a lower probability outcome may occur you're going to get exactly one of the outcomes but the model isn't 'wrong' just because the lower probability event occurs.
Looks 'strictly true' to me.
I started trying to speculate about what you're asking for
The key question is not where Trump stands with respect to the country as a whole. The question is where he stands with respect to enough states to give him 270 or more electoral votes.
I assumed "Probabilistic prediction models mean never having to say you were wrong" implied a criticism of said models as opposed to some notional non-probabilistic prediction model. I also assumed that you were complaining that 538's models were unfalsifiable, and I was attempting to point out that, with enough trials, you could subject them to hypothesis testing.
I stand corrected, although I now have no idea why you said it.
Be still your racing heart, Inga. No, we won't be accepting polls or anything else the mainstream media slobs have to say for a very, very long time. When your industry is predicated on integrity and you blow it...? Wellp - you lefties are going to learn the hard way that once your credibility's gone, you've got nothing left. And no, you weren't brought down by Russians, or Trump, or populism - you guys cut your own throats.
Hmmmm. Back when Obutthole was running the show a good rule of thumb was to take negative poll numbers and double them. Eg. when they claimed unemployment was 10%, you knew damn well it was closer to 20%. From what I'm seeing Trump is easily at 65% approval - maybe a smidge higher. All he has to do is keep giving the finger to the press and their drones, and keep the economy rolling - and he's in.
As a dissident conservative my big fear would be that the Donks sober up, and start throwing their stupid women like Pelosi, Hillary and Inga under the bus. I think the jig is almost up on America's flirtation with affirmative action/victim politics. The Donks are going to need candidates that can THINK.
"What exactly would you like instead of a probabilistic model? How would it work?"
The pollster would accept bets, at the odds determined by its polling (minus a reasonable vigorish, say 2.5%). I.e., they'd put their money where their mouth is.
There may be a few regulatory issues to be worked out.
Support the Althouse blog by doing your Amazon shopping going in through the Althouse Amazon link.
Amazon
I am a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for me to earn fees by linking to Amazon.com and affiliated sites.
Support this blog with PayPal
Make a 1-time donation or set up a monthly donation of any amount you choose:
46 comments:
First!
Is Drudge telling us that polls are like "Devil Dolls" that call us bitches?
After a few of those mimosas, she'll call anyone a bitch.
Can't wait to here the LLR's and Inga explain how Rasmussen is a tool of the right, and to be ignored.
"hear" not "here"
DOH!
"Devil doll" = another fake headline from Drudge. One might think he just copies the fake headlines from e.g. thesun.co.uk, but he often jazzes them up if the originals are not click-baity enough.
Click bait. The Sun has bleeped out the doll's questionable speech, so it's impossible to judge whether the complaint is worthy or just another example of rampant insanity.
On the other hand, it could be a case of just a little chromium switch (for Meade's amusement).
A first year of endless negative press and unfounded rumors vs gushing love and we end up in the same place
Consider this: If Trump's first year-end approval numbers are the same as Obama's, how are we to assess this in the light of 95% negative press coverage of Trump versus 95% (or more) positive — nay, worshipful — coverage of Obama from the very start of his political career?
I would have beaten SteveR to the punch if it were not for "conflicting edits" five times in a row.
I would have beaten SteveR to the punch if it were not for
those meddling kids!
A first year of endless negative press and unfounded rumors vs gushing love and we end up in the same place
Which proves that them Rooskies have Fabulous Secret Powers.
Consider this:
1) Trump's poll numbers mean that given even cruelly neutral treatment by the MSM Trump would be more popular than Obama, or Clinton. He would be in Reagan territory at least.
2) The fact that he isn't is strictly due to a press corps devoted to Trump's destruction rather than to truth.
3) The press corps is singularly inept at the business of destruction when the object won't lie down and take it.
Another thesis is that the media is not that influential at all and people are simply tuning it out or discounting the narrative.
So we're back to accepting poll results as valid conclusions about anything? Glad we got that straightened out.
Good Comments... :D
See Our Website
Do these numbers mean that the Network will renew Trump for another season, and that he gets to continue appointing judges and rescinding regulations ?
Since you’re back to accepting the validity of polls, here you go.
I told my wife on inauguration day, with Pussy hats marching and Anti-fa thugs rioting in the streets, that Trump will win reelection in a landside.
I stand by that prediction.
Israel is naming the new rail station on the Tel Aviv to Jersusem line near the Wailing wall the Trump Station. Meanwhile Obama gets new ICBM nuke installations built in Iran named for him, he bring their main developer.
We know it's a steep uphill climb for your brain, Inga, but it ain't that simple.
Inga: "Since you’re back to accepting the validity of polls, here you go."
Trump has NO PATH to 270 electoral votes.
None.
At all.
FiveThirtyEight told me that. Quite explicitly.
And they are never wrong. Ever.
I hope President Hillary doesn't attack conservatives too much.
And shouldn't that Tillerson guy be long gone by now?
Inga and her fellow leftists will be praying to Moloch nightly that tax reform backfires on the GOP and the economy goes south. They hope for this despite the fact that many millions of Americans would suffer and their own 401-Ks would take a hit. They hate Trump so much they would rather see the country fail than Trump's presidency succeed.
If tax reform does what the GOP and Trump believe it will do, Americans will prosper and the Democrats' predictions of doom will make them look like hysterical liars.
Fivethirtyeight.com's 2016 election prediction: Clinton 71.4% Trump 28.6%
So Inga cites another fivethirtyeight.com poll and expects it to be accurate?
FiveThirtyEight told me that. Quite explicitly.
By pumping headlines lefties want to see FiveThirtyEight may be priming a second wave of delusion when the headlines don't square with reality. The Democrats’ Wave Could Turn Into A Flood...Democrats are probably favorites to win the House...No other survey taken in November or December in the year before a midterm has found the majority party in the House down by that much since at least the 1938 cycle..Republicans in worse shape right now than any other majority party at this point in the midterm cycle...Wow.
If tax reform does what the GOP and Trump believe it will do, Americans will prosper and the Democrats' predictions of doom will make them look like hysterical liars.
I like that nobody gets to pontificate over hypotheticals anymore. We're gonna get to see who's right.
"By pumping headlines lefties want to see FiveThirtyEight may be priming a second wave of delusion when the headlines don't square with reality."
Not to worry. LLR Chuck has already clearly demonstrated that he and his lefty operational allies will have no problem shifting to a Hey I Was Actually Right All Along mode after the next set of disappointments.
I wonder how many Dems running for Congress will seek H>er endorsement, or ask H>er to campaign for them, much admired as She is.
rehajm: "I like that nobody gets to pontificate over hypotheticals anymore. We're gonna get to see who's right."
Correct. And the entire nation will get to see in real time how much lying the lefties have been engaging in.
You can already see some lefties recognizing the danger of being so incredibly exposed on this issue and are already shifting to - sure, sure, almost everyone gets a benefit but those zillionaires really made out - talking points.
Not Inga of course. She's always the lone dead-ender hanging out on an island not realizing the war has been over for 15 years. These are the cannon-fodder foot soldiers of the left who really don't care what reality looks. They only care about what talking points they are supposed to utter today happen to be.
Which is why history must start anew each day for them, lest the cognitive dissonance overwhelms them.
"FiveThirtyEight told me that. Quite explicitly."
In general, I don't think people understand how Silver presents probability. It might translate better if it was presented as 7 versus 3 rolls out of 10 instead of percentages.
I look at poll averages. The only single poll I give significant weight are Ann Selzer's.
Leading up to an election, I look for unknowns and volatility. Alabama was impossible to poll. State polls tend to be less reliable, because there tend to be fewer polls.
It will be harder to predict 2018 because we're looking at a bunch of state races. Some will be well polled, but others will not, so the averages, most likely, won't be accurate.
I enjoyed Inga's link. You can compare Trump to former presidents. Trump's popularity will go up with increasing peace and prosperity — like Nixon's did before he was abruptly destroyed. I hadn't realized just how graphically below average Obama was. Clearly a dud. And he started out so high.
The "Get Trump" squad is losing steam and credibility. But their one saving grace is winning the House in 2018. Mind you, this is mostly a historical trend, (opposition party picks up seats in midterms), but the Left will portray it as a vindication of all their Anti-Trump angst and effort. And then they will impeach. And then they will hope it tars him sufficiently to cause him to lose in 2020.
And all will be at peace again.
Well, at least that's the plan.
I am surprised with the noise from the Press Trump's rating are as good as they are. I think it's 94% negative press in the MSM for Trump per a Harvard study. Amazing he has not been impeached with that amount of attacks against him.
Found the article, only 93%.
http://www.washingtonexaminer.com/byron-york-harvard-study-cnn-nbc-trump-coverage-93-percent-negative/article/2623641
In general, I don't think people understand how Silver presents probability.
They ask Hillary and then tone it down for modesty's sake.
Probabilistic prediction models mean never having to say you were wrong.
Hillary said she didn't know why she wasn't up by 50 points, but she meant 100 in her little mind.
"Probabilistic prediction models mean never having to say you were wrong."
Not strictly true. If 538's models predicted a 30% winning chance for R every election, and time after time, R always won, there are specific ways of quantifying how poor their predictions are with each successive "wrong" call, but I don't think we have enough presidential predictions for it to mean much yet (wasn't Romney about 5%?).
Almost everyone understands odds in horse racing, but, for some reason they don't seem to parse the same numbers in politics.
It might be that Silver presents his odds as percentages which induces confusion with polling percentages. Many people thought 70% chance for Hilary meant she was a "sure thing", which, if that was her opinion poll number, she would have been, but it just meant 2:1 odds, and no sane person would describe 2:1 odds on a horse race as a "sure thing"; odds of 2:1 against a horse doesn't even make him a long shot, it just makes him not the favorite.
Of course it might be that people just like yelling at someone, although why Nate Silver, of all the possible targets, I'm not sure. Much of the left were yelling at him before the election for underestimating Hilary's numbers; some commentators actually accused him of committing fraud in order to drive readers to his site. Now many on the right say "Nate Silver was wrong, never believe anything he says" when he was one of very few forecasters predicting a relatively high chance of a Trump win.
What exactly would you like instead of a probabilistic model? How would it work? I'm at a loss to understand what it would even look like. I started trying to speculate about what you're asking for, but everything I came up with just sounded snarky.
Not strictly true.
The models assign probabilities to particular outcomes. In the case of an election the possibilities are win/lose/tie. While a lower probability outcome may occur you're going to get exactly one of the outcomes but the model isn't 'wrong' just because the lower probability event occurs.
Looks 'strictly true' to me.
I started trying to speculate about what you're asking for
I never asked for anything.
Again, at what point will Trump's approval/disapproval ratings mean Hillary! becomes President?
I hadn't realized just how graphically below average Obama was.
It wasn't just the graphs.
The key question is not where Trump stands with respect to the country as a whole. The question is where he stands with respect to enough states to give him 270 or more electoral votes.
I assumed "Probabilistic prediction models mean never having to say you were wrong" implied a criticism of said models as opposed to some notional non-probabilistic prediction model. I also assumed that you were complaining that 538's models were unfalsifiable, and I was attempting to point out that, with enough trials, you could subject them to hypothesis testing.
I stand corrected, although I now have no idea why you said it.
Drudge uas too many links yo wapo and nyt and other unreliable sources. Use Whatfinger now.
The importance of polls is in convincing the GOP in Congress to keep promises.
The next year will be interesting because the GOPe must want to be re-elected except those who are quitting to become lobbyists.
Even the NeverTrumpers are going to start seeing reality, except of course the Bill Kristols and George Wills, who are not running for office.
Be still your racing heart, Inga. No, we won't be accepting polls or anything else the mainstream media slobs have to say for a very, very long time. When your industry is predicated on integrity and you blow it...? Wellp - you lefties are going to learn the hard way that once your credibility's gone, you've got nothing left.
And no, you weren't brought down by Russians, or Trump, or populism - you guys cut your own throats.
Hmmmm. Back when Obutthole was running the show a good rule of thumb was to take negative poll numbers and double them. Eg. when they claimed unemployment was 10%, you knew damn well it was closer to 20%. From what I'm seeing Trump is easily at 65% approval - maybe a smidge higher. All he has to do is keep giving the finger to the press and their drones, and keep the economy rolling - and he's in.
As a dissident conservative my big fear would be that the Donks sober up, and start throwing their stupid women like Pelosi, Hillary and Inga under the bus. I think the jig is almost up on America's flirtation with affirmative action/victim politics. The Donks are going to need candidates that can THINK.
"What exactly would you like instead of a probabilistic model? How would it work?"
The pollster would accept bets, at the odds determined by its polling (minus a reasonable vigorish, say 2.5%). I.e., they'd put their money where their mouth is.
There may be a few regulatory issues to be worked out.
CNN is investigating big white trucks. Without using the license plate. Because Trump.
Post a Comment