Not an outcome I'd prefer, but I don't think there is any doubt that Democrats are more fired up that Republicans. Fired up like "frothing at the mouth eager to vote. Twice."
Dems didn't turn out as much when Obama wasn't on the ballot. I sense the same thing will happen with the Trump GOP.
A one in four chance is certainly significant and I'd bet is not what Nancy wants to hear. To be fair to 538, they gave the Donald, as I recall, a 25-35% shot at winning just prior to the election, which is not insignificant. 25% probability events happen all the time and are certainly not a longshot. I think the problem for most partisans is that they cannot wrap their heads around probabilities. If the Republicans keep control, I don't think it should be surprising, since I think they have a reasonable (i.e. 25%) shot.
Now to see, if the widespread misreading of the 2016 presidential election was an aberration, or represented a structural problem with polling. Last I heard generic democrat versus generic republican in a congressional seat was up about 4-5 points. That sounds like Dems take over if they only need 20 something seats to flip. While the republicans have won the overwhelming majority of special elections since the 2016 election, they have all been much closer than they probably should have. That being said, each of those elections got tons of democrat money, and tons of national press coverage from the mainstream media hoping for an upset and a sign of a 'blue wave'.
If the Republicans had developed a clear plan for what they wanted to do for healthcare to replace the failing Obamacare, they could have campaigned on that and probably won.
yes, 25% probability events do happen all the time. I get tired of people saying 538 didn't poll well leading up to the election. It's not Nate's fault people don't understand probability. They think pollsters are fortune tellers or something.
What's the chance each month of pregnancy with 2 fertile people in their prime? Not higher then 25% on any given day. Pregnancy happens all the time between 2 fertile people in their prime.
"The party in power almost always loses these mid-term elections. What's so funny?"
Republicans, cutting off their nose, to spite their face.
The Democrats are defeat-able - still - but, if successful, they will have the rose from the grave by the slimmest margin, to actually do damage to Trump where, before, they were simply ass-out. That's what's funny:
I haven't ever voted Republican. Since 1972. But I can't vote Dem anymore. This Russia crap was it for my lesser-evil. I was never going to vote for Hillary "We came. We saw. He died." Clinton but I thought I'd return when any other Dem was up. No more. The best polls to watch are graphed at Real Clear Politics. If Trump was a stock I'd buy because he's broken out to the upside. If he drops the crap and concentrates on elections he can pull the GOP up.
It's nothing new under the sun for the party in power to lose seats. It's the pendulum swinging. Considering the non-stop media onslaught against Trump, the surprise is even less so.
If the Republicans had developed a clear plan for what they wanted to do for healthcare to replace the failing Obamacare, they could have campaigned on that and probably won.
Yes, if the GOP loses the House they have no one but themselves to blame.
I wonder if losing will affect their prospects for fat lobbying jobs ?
Blogger Widmerpool said... A one in four chance is certainly significant and I'd bet is not what Nancy wants to hear. To be fair to 538, they gave the Donald, as I recall, a 25-35% shot at winning just prior to the election, which is not insignificant. 25% probability events happen all the time and are certainly not a longshot. I think the problem for most partisans is that they cannot wrap their heads around probabilities. If the Republicans keep control, I don't think it should be surprising, since I think they have a reasonable (i.e. 25%) shot.
The only disagreement I have with this is that if it was really a 25% chance to win it seems very unlikely that Donald would have won by the electoral margin that he did. While not a landslide, 306 to 232 was a very strong performance for Donald Trump. Which makes me think the real probability going into the election was closer to 55%/45%. If it was really closer to 75%/25% virtually all the probable outcomes of a Donald victory should have been narrow squeakers. I think Nate Silver put out a tweet early on into results during election night saying something like, these results don't look like a 90%/10% probability.
Then it's a 3 in 4 chance that Bill Maher gets his wish, and the economy tanks. Probably not full tanking, since Trump's policies will still be there for a while, but it will tell business owners, investors, and consumers that their fellow citizen are stupid enough to blow a booming economy and can't be trusted. That will kill a lot of jobs real quick. It happens all the time, and it's happened here more than once. Venezuela lite. I'm not eating my dogs unless they look at me like Cujo, then it's barbeque at my house.
What I got of this 75% announcement is that the Blue Wave isn't.
It wasn't long ago that the Democrat takeover of the House was a certainty, and the 75% may have been whether the Democrats takeover the Senate. That the Senate will remain a GOP majority is hardly a discussion point these days. Now it is the House that has a 1 in 4 chance of staying within the GOP.
But that's just looking at the trend. When we factor in 528's actual ability to predict the future; 75% is worse odds than they gave Hillary Clinton at this time, which was 88%. 528 only once showed a better than 50% chance Trump would win and that was immediately after the Democratic convention, and likely a heavy thumb on the scale for the pretense of impartiality. 528's final prediction for Hillary was 71% chance of winning, but she lost, and they were wrong. When an impartial poller, you can bet the real chance of Democrats winning is much lower.
Are conservatives telling pollsters the truth again? Prior to 2016 I met a lot of stealth Trump supporters. The voted for him, but if you did not know them well, you'd never have guessed.
Now i meet folks that hide being Republican just to stop hearing the vitriol.
Isn't that exactly what is historically supposed to happen anyway? So what's the news?
Most Trump voters are notoriously quiet about their support, and avoid or lie to pollsters. That's why the polls were off so much in 2016. They are more open now, but will never be anything like the left who live to be seen and heard, and to tell everyone what they think. I wish they had that enthusiasm about going to work.
if it was really a 25% chance to win it seems very unlikely that Donald would have won by the electoral margin that he did.
70,000 votes in Pennsylvania, Wisconsin and Michigan. 25% chance doesn't sound off. Pollsters are not fortune tellers. 25% chance is a pretty darn good chance. We're not talking 10 to 1 odds.
wwww - I recall an amusing NPR interview with Nate and colleagues at one of the conventions. Nate said that Trump had a 25 to 30% chance, and the uneasy host, amusingly and revealingly, asked him how often 25 to 30% events happen. Nate said, of course, 25 to 30% of the time. The host clearly wanted him to say something along the lines of never in a million years.
bagoh20 said... Then it's a 3 in 4 chance that Bill Maher gets his wish, and the economy tanks. Probably not full tanking, since Trump's policies will still be there for a while, but it will tell business owners, investors, and consumers that their fellow citizen are stupid enough to blow a booming economy and can't be trusted
********************
Just a few weeks ago our financial advisor told me and my wife that if the GOP loses the House he will go "straight to safety" with our investments, for the reasons you cite above. Rescission of Trump's tax cuts alone will put the brakes on economic expansion.
I don't think the polls can be used any longer to make predictions. If they could, Trump's odds of winning in 2016 would not have been 25% or less on election day. 538's prediction on election day 2016 was based solely on Trump's chances of winning VA, OH, FL, and IA along with the chances of him holding onto NC, GA, and AZ. If Silver had thought at all of Trump winning WI, MI, or PA, he would have put Trump's odds at 50% at least.
The media have sullied their polling with their relentless anti-Trump/anti-Republican rhetoric. I don't answer polls at all, but if I did, I would lie to them just for the amusement. I can't be a tiny minority in that regard.
"Trump is all the Republicans have got going - no one's attracted to THEM."
Yes, the right is all consumed with being attractive. You can tell by how we dominate the fields of fashion, race posturing, Hollywood, and media. It's clear by how we arrange our entire lives to avoid being seen as racist, uneducated, and how we forgive anything if it's clean and articulate, with a crisp pants crease. Yea, that's us hillbilly wingnuts, all struggling to get those flashy graduate degrees and impress everyone with our cultural superiority and wokeness. We are obsessed with everyones' opinion of us.
I just note for the record that as of this morning, the generic ballot reported on RealClearPolitics range from +4 Dem to +11 Dem. If you know anything about statistics, you know what this indicates about polling.
Blogger Earnest Prole said... Voter enthusiasm is the entire ball game in low-turnout, off-year House elections: Democrats have it and Republicans don't.
The first part is certainly correct. The second half has historically been true but I'm not sure it will be that way this year. Most polls show that democratic voters are highly energized. The special elections have all shown very high democratic turnout, along with most of the primaries. Now, in the special elections they each individually received lots of national money and national media attention, all of which will push up turnout. The best thing the Republicans can hope for is democrat fatigue from everlasting Trump Outrage Syndrome. If the Mueller probe wraps up before the election and turns out to be the big fat nothing burger that it looks like now, and the media moves on to some other "OUTRAGE", I could see that having a mass wearying effect on the democrats. Lots of them have become emotionally invested in Mueller somehow getting rid of Trump, if nothing comes from it will be a big let down and will certainly drive down turnout in November.
Tim in Vermont said... Nate's the one that didn't understand probability. He thought that each state was like an independent roll of the die, when in fact, the states all moved together based on an underlying demographic trend that should have been obvious to him.
Nate Silver precisely made this distinction between his models and others. In particular, the Clinton internal polls seemed to treat each state independently and missed the possibility of a rollover effect.
The Vault Dweller said... If it was really closer to 75%/25% virtually all the probable outcomes of a Donald victory should have been narrow squeakers.
And they were. Electoral College votes don't represent the margins in key states:
Trump won Michigan by 0.3% Trump won Wisconsin by 1% Trump won Pennsylvania by 1.2% Trump won Florida by 1.2%
Basically Trump won every squeaker that mattered.
In close states, those under a 2% margin of victory, Clinton won New Hampshire and Minnesota. Woo Hoo.
The main thing going against the Republicans is that they are happy with what they have right now, becuase it's pretty damned good - historically really good even for the people who hate them. But that satisfaction is not going to get as many to the polls as the painfully inflamed Lefties, who feel like they are having their teeth drilled without novocaine 24/7. They imagine themselves voting against Hitler. You gotta go vote with that attitude, but it's not a good thing to go shopping when you are really hungry, or make an important decision when you're having a tantrum.
You keep posting this phrase Crack, in what I assume is a desperate attempt for more attention.
I have tried to engage you in the past, but you get so hung up on your "angry Black man/stupid White people" schtick that it turned out to be a waste of time.
Do you use this blog as therapy or are you trying to sell something?
yeah, it's been pretty stable for a couple of months now. Those chances are going to go up for Democrats unless something happens. I keep expecting something interesting to happen in Korea that would make it clear that Trump is doing well there. I'm not sure what else is going to change between now and then.
That is kind of my point. He won lots of those squeakers. The way I remember the end of election night was that it came down to Trump could win if he won one of either, Pennsylvania, Michigan, or Wisconsin. And he ended winning all three. If he really only had a 25% chance it seems far more likely that he would have only won one of those. Two is even more unlikely. And three should have been even more unlikely than than that. The fact that he won with most unlikely scenario (of the outcome of those three states) makes me think the initial estimate of a 25% chance to win overall was off.
"The Democrats are defeat-able - still - but, if successful, they will have the rose from the grave by the slimmest margin, to actually do damage to Trump where, before, they were simply ass-out. That's what's funny: Stupid white people."
They were never ass-out. Their demographic advantage prevents any ass-outery. Trump identified a new way to win: boost white votes by 4% rather than trying to boost Hispanic support by 70%. It ain't pretty, but he's just playing by the rules other wrote.
asked him how often 25 to 30% events happen. Nate said, of course, 25 to 30% of the time.
LOL! I love it!
I've been taken aback how people see polls as if they are a certainty. 25% chance means a 25% chance. I love this stuff. Moneyball is one of my favourite movies.
Surely many people have played card games and understand probability? Maybe people want, or even need, to know what is going to happen in the future. Perhaps they feel disturbed that the future is not dependable, so they impose certainty?
Nate on twitter yesterday:
p.s. The chances our model gives Democrats—between 70 and 75 percent depending on what version you look at—are pretty much exactly the odds that Hillary Clinton had of winning in 2016! So hopefully everyone's learned their lesson and won't mistake that for a sure thing. 😂😂😂
"Crack, if *we* are stupid, why aren't *you* and your cohort rich?"
Who said rich correlates with intelligence? Is Ann rich? Are you? Jenny McCarthy has her dressing rooms "cleansed" for spooks and thinks vaccines caused her sons autism - but still has more money than all of us - is she smart?
Two points. First, if the generic ballot is less than 5 points in D favor, it really means R lead. Why, because they always oversample D to skew the result they want. In my experience D +5 is the critical line. D lead in generic has declined over the last year, went up slightly during June, and is now resting on the line again. If they are calling it even then it’s definitely R leading. Second, the D talk of impeachment and public disgust with the never-ending lameness if the Russia probe will bring out R voters. Don’t be fooled by the primaries, which energizes voters for different reasons. Of course more D votes in MN when they have five D running off to face an unopposed R. The people who show up in November will be motivated differently.
Republicans increase lead in Senate and keep House majority.
I'm staying in touch with Ann and Meade, but the Republican Party? No way. MLK said not to, as did my black friends, but I tried it anyway - twice.
White Republicans don't want anyone but extreme ideologues, willing to rip blacks apart to join them, when - like the French - they're proven to be merely fair weather friends, at best.
Leaving the comfort and security of black people for that is madness. Even the suggestion is cruel - especially in a country known to have ripped us apart, literally, in the past.
No, white Republicans are their own worst enemies - and everyone else's, too. I'm not trying to be the first guy on the Dodgers. I just want to live in my country as comfortably as any other American but whites won't have it.
There's no reason for me to keep trying if they can't. I'm off to the Post Office.
Don't take this too seriously. 538 did this for the World Cup. Obviously a lot of its predictions were right and others wrong. It added to the enjoyment of following the tournament but France won on the field.
I don't get why anyone still listens to these people.
I have no college degree and I'm not too good at the maths.
But I can read and know just a tiny bit of history. And history tells us that most of the time, during the president first midterm election, he will lose both the house and Senate, or if his party doesn't already control it, he will lose more seats.
This has happened so often and so regularly throughout history that it was amazing when it didn't happen to Bush. Of course, in hindsight, it's easy to see why, with 9/11 still stuck in our heads and the country afraid for our lives.
But not only is this stupid on stilts, it doesn't even mean anything. If you read Nate Silver, even he will tell you it just means the Dems have a better chance right now.
Duh.
There is only one major purpose to these polls and prognostications. Not to predict the result but to infuence the result. Right now Democrats all over the nation are sitting down and having serious discussions with their allies in silicon valley. Nice Facebook you've got there. Have you seen the polls lately, about how we are going to take over soon? Wouldn't want you to lose that whole YouTube thing because you allowed anti Democrat, pro Republican voices, er, we mean, hate speech, wink.
Good to see that Crack worked out his cognitive dissonance concerning Republicans, what took you so long? Or was it because all the Republicans here made you feel so unwelcome and blamed you for everything any black persons has done wrong, ever?
"Leaving the comfort and security of black people for that is madness"
Tribal comfort seems to be what will dictate politics and culture in New America. Like I said, Trump (and the GOP) didn't make those rules, but I guess we're all obligated to play by them.
I've yet to hear a meaningful explanation as to why polls for Hillary! were entitled to use Obama's turnout estimates. The black guys at the barber shop are staying home over voting for a sour old white woman. Yet they're going this way again. Just look at all those D's coming out against Roy Moore! It's a sign of the blue wave I tell ya!!!
Nate was less waaaay wrong! than the rest of them but he and the rest of them are in the dog house until they call a few very very right.
I just want to live in my country as comfortably as any other American but whites won't have it.
Let's see a show of white hands. Hmmm. Not seeing it, Crack. My informal survey shows nobody gives a shit how you lead your life so long as you're not hurting anybody else. I mean there are people who want to hold you back and keep you down, but they are largely Progressive and, unfortunately, also Black like you. Every racist I've ever known was a self-identified open-minded liberal, but YMMV.
I would love to bet on that at those odds. Democrat media has pounded relentlessly. Wait til campaign season kicks in and the clueless voter starts seeing GOP tv ads and hears the news about the great economy, low black unemployment, the numerous foreign policy improvements, and the insane Democrat ideas.
Actually, self-described LLR Chuck and his "Department of Black People" remark might give Crack the impression that the establishment Republicans are racist. I can't defend those country club people because I don't know any. They do squeeze into the Big Tent, but they are losing their grip on the GOP. My family is full of people who worked for a living and hated golf, but we voted conservative. And at the heart of conservatism is the original purpose of the R party: ending slavery and fulfilling the Declaration's prediction of equality for all. Most political struggles are the D party trying to make things unequal in the name of "fairness" and the R party trying to hew to original principles, once you strip away the bullshit.
..and that's why 538 is buckets of shyte- their raison d'etre is to show lefties what they want to see, skipping the lies and damned lies and going straight to statistics.
Perhaps some odd Wednesday a few weeks from now Nate will be able to demonstrate a slight narrowing in the Senate races that's easily explained by margin of error but the FiveThirtyEight headline will read Senate Republicans in Trouble!. It's a tired shtick...
For every Crack that abandons the Republicans there are ten white people who have come to realize that you have to be a fool to vote Democratic if you are a white person.
Sure you have the crazed liberals, feminazi's and virtue signaling cucks but normal white people are abandoning the Democrats in droves.
We have become a nation obsessed with identity politics and this is the fruit of that labor.
RealClearPolitics.com lists 199 districts as “leans Democrat” or further. Six have Republican incumbents. 194 are “leans Republican” or further. That leaves 42 in the “tossup” column, and 40 of them are currently held by Republicans. If the Democrats win all of the leaners (they won’t) and 19 or more of the tossups then they win the House. The Republicans need to win all of their leaners (they won’t) and hold at least 24 of 40 seats rated “tossup” to hold the House.
So, reasons why Silver is probably wrong include:
(1) We don’t know the value of incumbency in 2018. Republicans gave their constituents a tax cut, a strong economy, and real wage growth that is weighted towards the working class (in contrast to Bill Clinton’s dot-com boom, which was weighted towards the upper end of the scale). If the incumbents have been doing a good job with constituent service then that should be more than enough to see them re-elected.
(2) Republican incumbents are in the unique position of being able to run against both Pelosi and Ryan, should they choose.
(3) Related to (2), the Democrat faces they see in the news — Pelosi, Ocasio-Cortez, Sanders, Andrew Cuomo — are bat-shit crazy. Nationalizing the election could be a good strategy.
(4) Trump
(5) Reaction to lawlessness on the left, including Crooked Hillary, crooked FBI and CIA directors, Antifa, etc.
(6) The DNC is both broken and broke.
But reasons why Silver could be right include:
(1) Democrats are frothing at the mouth to vote against Trump (well they are always frothing at the mouth about something or another, but this year, like in the local 2017 elections here in Virginia, it’s Donald Trump).
(2) Thanks to Reince Priebus, Republican get-out-the-vote efforts almost matched Democrats in 2016. Mitt Romney’s niece ain’t no Reince Priebus.
I honestly don't see the Democrats meriting those kinds of odds for a takeover of the House.
For instance, they have the Michigan 11th -- a perfect bellwether House district for much of the Midwest and suburban American -- as likely going to the Democrats. (It's close, and FiveThirtyEight knows it, of course.) And my bet in the 11th is on Lena Epstein, the Trumpist Republican, holding that seat. The district was crafted by Republicans to give us about a +4 advantage.
If we do lose that seat, then it really will be a shitty election night for Trump.
"By projecting a vision of a world in which the problems of blacks are consequences of the actions of whites, either immediately or in times past, white liberals have provided a blanket excuse for shortcomings and even crimes by blacks."
Actually, the chance Republicans keep control at 538 varies depending on which of the three models you pick:
32.1% - Lite - "Keep it simple, please — give me the best forecast you can based on what local and national polls say
25.4% - Classic - I’ll take the polls, plus all the “fundamentals”: fundraising, past voting in the district, historical trends and more
31.1% - Deluxe - Gimme the works — the Classic forecasts plus experts’ ratings
Predictit is putting the odds at 36% (with a 3% margin for arbitrage). That's more in line with 538 Lite and Deluxe.
At the race level, 538 is predicting two incumbent Republicans will lose in Iowa. Are Iowans really that disgruntled with the Republican Party? Republican Governor Kim Reynolds is still expected to win.
In any case, we can argue whether we are flipping one coin or two, but the U.S. House is a coin toss, and Democrats may be setting themselves up for another tough beat.
@The Vault Dweller: And he ended winning all three. If he really only had a 25% chance it seems far more likely that he would have only won one of those. Two is even more unlikely. And three should have been even more unlikely than than that. The fact that he won with most unlikely scenario (of the outcome of those three states) makes me think the initial estimate of a 25% chance to win overall was off.
If he had 3 50/50 races--and those three were pretty close to that for him, given his margin--his chance of winning all three is actually 12.5%. (1/2)*(1/2)*(1/2). (Exactly the same as his chance of losing all three.)
And if Nate Silver had said exactly that--Trump needs 3 to win, and his chance of doing that is 12.5%, you would be saying now that the odds couldn't have been that bad if he won all three. But that's not true.
And if he'd lost all three, in that situation, you'd have said "well 12.5% chance to win all three of course he lost them"--even those losing all three was exactly as likely as winning all three, nothing has changed in the math.
People do not have a a good intuition about probability. You have to work it out.
I suspect Fivethirtyeight has not factored in Trump doing 6 big Rallies a week during two months before the day. That's easy. Remember President Trump was doing 5 a day in the weeks up to the 2016 election. DRAGON ENERGY!
One thing I wonder, particularly in the House, is how many people split their vote and voted for Clinton for President, and then a Republican congressman to restrain her worst impulses, in the mistaken belief that she was a lock to win the Presidency. My recollection is that in some swing states, she actually ran ahead of the local Democratic candidate for the Senate. In Ohio, for example, she got almost 400,000 votes more than the Democratic candidate for Senate (Ted Strickland). In Wisconsin, she got . . . about 2,000 votes more than the Senate candidate (Russ Feingold). But Trump ran about 70,000 votes behind Ron Johnson. I was going to use Pennsylvania as an example too, but actually both she and Trump ran ahead of their parties' Senate candidates, so it doesn't really work.
Anyhow, I don't know the statistics on ticket-splitting in 2016, but I can easily imagine some strategic voting on the mistaken premise that Clinton was going to be the next President. That would tend to "correct" during this electoral cycle.
I'm not buying the idee fixe that Republicans are not enthusiastic and won't turn out. Many Republican voters, including yours truly, are far more supportive of Trump than they were on 11/8/2016. The constant state of siege mounted by the leftist culture warriors has inspired us to fight back. We have no respect for polls, least of all 538, so we don't take surveys.
Here's my prediction. Republicans lose seats in the House but not the majority. Republicans gain seats in the Senate. Remember, you saw it here first.
People do not have a a good intuition about probability. You have to work it out.
Off topic, but I feel like this is a real problem for lawyers who are trying to evaluate evidence (and probably for juries trying to weigh evidence). Suppose you have to establish three factors for a man to be guilty: A, B, and C. And they're all independent. And you're 75% confident each factor is met individually. That sounds like you're pretty confident! But that would imply 0.75^3 = only a 42% chance that all three requirements are met such that he's actually guilty. How many lawyers actually think through the weight of evidence in multi-factor tests logically? Very few, I would guess.
@Balfegor: How many lawyers actually think through the weight of evidence in multi-factor tests logically? Very few, I would guess.
Made even harder because in real life probabilities relevant to criminal cases are rarely known numerically and precisely; and often they're not independent.
Suppose you know a murder was committed by a) a man b) aged 42 years old c) height 5' 10" who d) lived in the same apartment complex as the victim--all of which characteristics are shared by the gentleman in the dock. But a), b) and c) are not statistically independent. And if the apartment complex was for middle-aged men only, well...
Nate can factor normal elections. How does he factor
a) insanity. The positions and rhetoric of many Democratic candidates is...offputting to normal people. So Dems will win Stupid Ignorant 90% Dem Districts like they always will...but the moderate areas are not likely to be enthusiastic for voting for a Socialist as they read about the meltdown in Venezuela.
b) Republicans and Conservatives don't talk to Nate Silver. We just don't. We don't like the media, have had our fill of push polling and I personally have hung up on three pollsters this year.
This is data that Nate Silver misses. And those people slamming the phone down? They are PASSIONATE people. They are 'crawl over broken glass' Republicans.
It is an entire data pool Nate has very little entry into.
But I could be wrong. Maybe he knows how much spin to cypher out all the hang up calls or 'hide the decline' like Anglia-gate.
Which makes his estimates less accurate suppositions.
Personally, I HOPE he is wrong, but I am open to a split Congress.
But, but, but: what about the Rasmussen poll yesterday that had black approval of Trump up to 36% from 19% three months ago. If - big if -the Dems have the House I am willing to bet that the R's end up with 57 seats in the Senate. Anything coming from the House will die in the Senate. Court appointments will continue to be made, trade treaties have nothing to do with the House and those are the next big things on the agenda.
Do you really think that all thoseTrump voters whose lives are being made better by Trump's policies are going to be stupid enough to stay home and not vote? Really?
Trump will nationalize the elections in order to up turnout. Do you think the rallies he has been holding are just for his own entertainment? What if Kavenaugh has not been confirmed; what's the impact going to be? How about when RBG is hospitalized just before court opens?
I think polling is just as abysmal now as it was in 2016.
"Off topic, but I feel like this is a real problem for lawyers who are trying to evaluate evidence (and probably for juries trying to weigh evidence). Suppose you have to establish three factors for a man to be guilty: A, B, and C. And they're all independent. And you're 75% confident each factor is met individually. That sounds like you're pretty confident! But that would imply 0.75^3 = only a 42% chance that all three requirements are met such that he's actually guilty."
-- ... That sounds like trying to determine how many wounds a gun line of dwarves will actually get through before a death block of infantry closes on them. "First, To Hit, then To Wound, then Saves."
I, for one , am much more positive about Trump and his administration and what's been accomplished than I was in 2016. In 2016 I voted against Hillary - absolutely - in 2018 I am voting for the way things are going in the country right now. I agree with Tyrone Slothrop that there are many like us who have gone from marginally positive to fully on board.
"Just a few weeks ago our financial advisor told me and my wife that if the GOP loses the House he will go "straight to safety" with our investments, "
@Stan - perhaps Americans pine for higher taxes, more illegal immigration, more mandatory homosexuality and more pandering to foreigners. But I am guessing no.
"Just a few weeks ago our financial advisor told me and my wife that if the GOP loses the House he will go "straight to safety" with our investments, "
You should find a different financial advisor.
I don't know. The Democrats would like to crash the economy before 2020. If they get the House they might get a chance to do so.
If you think they give a shit about the American people, at least the bottom 99%, I have a bridge for you.
If he had 3 50/50 races--and those three were pretty close to that for him, given his margin--his chance of winning all three is actually 12.5%. (1/2)*(1/2)*(1/2). (Exactly the same as his chance of losing all three.)
Now I might be wrong about this it has been a while since I've studied stats. But if Trump was in a position that he needed only 1 of those states to win the way to calculate the probably of that would be 1 - (probability of losing all 3 states) and if assuming each state is roughly 50% chance either way and they are independent of one another, the chance to win at least 1 state would be (1 - 12.5%) or 87.5% Now that percentage also includes the chance to win all 3 states, but as you calculated above that is just 12.5% of that total, which leaves 75% it was another combination other than winning all three. Of that 75% there are 3 combinations of him winning 1 of each state he needed to win individually and 3 combinations of him winning 2 of those states he needed to win as a combination of the three. So my point is that because he happened land in that one 12.5% segment of winning all three instead of the 75% segment of the other two combination possibilities of win suggests to me that the initial assment was off. Now I know it isn't an unheard of possibility, it is after all 1/8th. I just tick that result off as a factor in giving me my hunch that the initial polling was off.
@Ken B, I know where you're coming from. In mathematics and statistics I've heard it called the "Chinese Emperor's Nose Fallacy." You ask hundreds, maybe thousands, of people to estimate the length of the Chinese Emperor's nose. You collect the data, average the results, and calculate an estimated length for his nose to as many digits as your computer supports.
But, you see, no one is actually allowed to look at the face of the Chinese Emperor, so you are averaging hundreds or thousands or wild guesses, and the result is a wild guess, but to three or four or more significant digits.
Oh sorry slight follow up, in my above post we were talking the chance of him winning all 3 states assuming a roughly 50% chance of him winning, and that percentage was 12.5% Now if the reality of the probabilities were closer to the polls where it was 90% Hillary 10% Trump then the probability of him winning all three of those states was .1%. And that is my main point in that because of the margin of his victory in the electoral college I assume the probability of him winning is much closer to a 50% 50% split than 90/10 and 75/25 predictions we had seen from other pollsters.
@Vault Dweller, the problem you are having is that the three events are not independent nor are they uncorrelated. Let’s look at the conditional probabilities. For the sake of argument, lef’s accept that the a priori probabilities of winning Pennsylvania, Wisconsin, and Michigan are each 10%. Now let’s look at conditional probabilities. Would not you say that the probability of winning Pennsylvania given that Trump won Michigan is much, much higher? Maybe closer to 80-90%? And the probability of winning Michigan goes up if Trump has won Pennsylvania, though certainly not as dramatically.
But my biggest problem with Nate Silver comes when he looks at historical data as part of his analysis, because I think we are living in times that are fairly unprecedented. For instance, one of the reasons that the party holding the presidency loses seats in Congress two years later is that the new president’s coattails pull in some weak candidates that lose their first reelection campaign. But the GOP actually lost a few seats in the House in 2016, so it’s unclear — at best — what Trump’s coattails were like. Put another way, a Congress Critter who won in 2014 and was re-elected in 2016 is not as vulnerable in 2018 as someone who won for the first time in 2016 and is facing his or her first reelection in 2018.
"Tribal comfort seems to be what will dictate politics and culture in New America."
Blacks left the South looking for the law. Whites made sure we never found it. You can't expect any trust after that.
"Trump (and the GOP) didn't make those rules, but I guess we're all obligated to play by them."
See how you lied, to yourself, there? If whites made sure there was no law for blacks leaving the South - and Trump first came to public notice in a trial for discriminating against blacks in his buildings - how didn't whites make "those rules" we are "all obligated to play by" now? I'll tell you how:
You let yourselves off the hook for making the rules others live by.
It's hard to deal with people who are so dishonest, it carries over even to themselves.
@THe Valut Dweller:nd that is my main point in that because of the margin of his victory in the electoral college I assume the probability of him winning is much closer to a 50% 50% split than 90/10 and 75/25 predictions we had seen from other pollsters.
But we did the math. Probability does not work that way. When we started with 3 50/50 races we got a 12.5% chance of winning all three. Winning all three was PERFECTLY consistent with the odds being 50/50 on each one. Things that have a 12.5% chance of happening, happen all time. 12.5% of the time.
You have a 1/6 (17%) chance of rolling a 6 in one try. If you do it, and you roll a six, you are saying that you now have evidence that the odds must really have been higher than 1/6 or you couldn't possible have done it in one try. That's not how probability works, and casinos and state lotteries and people who run sports betting could not possibly make money, if probability worked the way you argue here that it does.
Click here to enter Amazon through the Althouse Portal.
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:
123 comments:
538 predicted Hillary would win in 2016 all the way up to Election Day.
All serving under President Hillary. Most qualified candidate EVER!
(Yes, I'm skeptical of the Polling results)
Yeah, we shall see. I no longer take Nate Silver as gospel.
Not an outcome I'd prefer, but I don't think there is any doubt that Democrats are more fired up that Republicans. Fired up like "frothing at the mouth eager to vote. Twice."
Dems didn't turn out as much when Obama wasn't on the ballot. I sense the same thing will happen with the Trump GOP.
If Republicans lose, I'm going to LAUGH my ass off.
I'm changing my political party in a half hour.
Bye, Republicans! It was fun while it lasted!
A one in four chance is certainly significant and I'd bet is not what Nancy wants to hear. To be fair to 538, they gave the Donald, as I recall, a 25-35% shot at winning just prior to the election, which is not insignificant. 25% probability events happen all the time and are certainly not a longshot. I think the problem for most partisans is that they cannot wrap their heads around probabilities. If the Republicans keep control, I don't think it should be surprising, since I think they have a reasonable (i.e. 25%) shot.
Now to see, if the widespread misreading of the 2016 presidential election was an aberration, or represented a structural problem with polling. Last I heard generic democrat versus generic republican in a congressional seat was up about 4-5 points. That sounds like Dems take over if they only need 20 something seats to flip. While the republicans have won the overwhelming majority of special elections since the 2016 election, they have all been much closer than they probably should have. That being said, each of those elections got tons of democrat money, and tons of national press coverage from the mainstream media hoping for an upset and a sign of a 'blue wave'.
If the Republicans had developed a clear plan for what they wanted to do for healthcare to replace the failing Obamacare, they could have campaigned on that and probably won.
"If Republicans lose, I'm going to LAUGH my ass off."
The party in power almost always loses these mid-term elections. What's so funny?
1 in 4. About the same chance that Trump had of winning on election night in 2016. Not bad odds.
25% probability events happen all the time
yes, 25% probability events do happen all the time. I get tired of people saying 538 didn't poll well leading up to the election. It's not Nate's fault people don't understand probability. They think pollsters are fortune tellers or something.
What's the chance each month of pregnancy with 2 fertile people in their prime? Not higher then 25% on any given day. Pregnancy happens all the time between 2 fertile people in their prime.
CJinPA said...
"The party in power almost always loses these mid-term elections. What's so funny?"
Republicans, cutting off their nose, to spite their face.
The Democrats are defeat-able - still - but, if successful, they will have the rose from the grave by the slimmest margin, to actually do damage to Trump where, before, they were simply ass-out. That's what's funny:
Stupid white people.
I haven't ever voted Republican. Since 1972. But I can't vote Dem anymore. This Russia crap was it for my lesser-evil. I was never going to vote for Hillary "We came. We saw. He died." Clinton but I thought I'd return when any other Dem was up. No more. The best polls to watch are graphed at Real Clear Politics. If Trump was a stock I'd buy because he's broken out to the upside. If he drops the crap and concentrates on elections he can pull the GOP up.
It's nothing new under the sun for the party in power to lose seats. It's the pendulum swinging. Considering the non-stop media onslaught against Trump, the surprise is even less so.
"Assuming that all of our explicit and implicit assumptions are correct."
If the Republicans had developed a clear plan for what they wanted to do for healthcare to replace the failing Obamacare, they could have campaigned on that and probably won.
Yes, if the GOP loses the House they have no one but themselves to blame.
I wonder if losing will affect their prospects for fat lobbying jobs ?
Blogger Widmerpool said...
A one in four chance is certainly significant and I'd bet is not what Nancy wants to hear. To be fair to 538, they gave the Donald, as I recall, a 25-35% shot at winning just prior to the election, which is not insignificant. 25% probability events happen all the time and are certainly not a longshot. I think the problem for most partisans is that they cannot wrap their heads around probabilities. If the Republicans keep control, I don't think it should be surprising, since I think they have a reasonable (i.e. 25%) shot.
The only disagreement I have with this is that if it was really a 25% chance to win it seems very unlikely that Donald would have won by the electoral margin that he did. While not a landslide, 306 to 232 was a very strong performance for Donald Trump. Which makes me think the real probability going into the election was closer to 55%/45%. If it was really closer to 75%/25% virtually all the probable outcomes of a Donald victory should have been narrow squeakers. I think Nate Silver put out a tweet early on into results during election night saying something like, these results don't look like a 90%/10% probability.
President Hillary Clinton was not available for comment.
Whatever. The Supreme Court was Job 1 and that's handled, and could get handled-er.
Then it's a 3 in 4 chance that Bill Maher gets his wish, and the economy tanks. Probably not full tanking, since Trump's policies will still be there for a while, but it will tell business owners, investors, and consumers that their fellow citizen are stupid enough to blow a booming economy and can't be trusted. That will kill a lot of jobs real quick. It happens all the time, and it's happened here more than once. Venezuela lite. I'm not eating my dogs unless they look at me like Cujo, then it's barbeque at my house.
What I got of this 75% announcement is that the Blue Wave isn't.
It wasn't long ago that the Democrat takeover of the House was a certainty, and the 75% may have been whether the Democrats takeover the Senate. That the Senate will remain a GOP majority is hardly a discussion point these days. Now it is the House that has a 1 in 4 chance of staying within the GOP.
But that's just looking at the trend. When we factor in 528's actual ability to predict the future; 75% is worse odds than they gave Hillary Clinton at this time, which was 88%. 528 only once showed a better than 50% chance Trump would win and that was immediately after the Democratic convention, and likely a heavy thumb on the scale for the pretense of impartiality. 528's final prediction for Hillary was 71% chance of winning, but she lost, and they were wrong. When an impartial poller, you can bet the real chance of Democrats winning is much lower.
Here's my prediction.
Democrats have a 35.4379% chance of regaining control of Congress.
Stay home if you don't want to be a loser like your friends.
It's official then: Republicans Keep the House!
Republicans, unnecessarily losing to a bunch of NewAgers and BLM assholes, are going to be waaaay more fun to watch than those idiots losing to Trump:
Trump is all the Republicans have got going - no one's attracted to THEM.
And Republicans melt down as easily as Democrats.
Are conservatives telling pollsters the truth again? Prior to 2016 I met a lot of stealth Trump supporters. The voted for him, but if you did not know them well, you'd never have guessed.
Now i meet folks that hide being Republican just to stop hearing the vitriol.
The odds are significant over a large number of trials.
Isn't that exactly what is historically supposed to happen anyway? So what's the news?
Most Trump voters are notoriously quiet about their support, and avoid or lie to pollsters. That's why the polls were off so much in 2016. They are more open now, but will never be anything like the left who live to be seen and heard, and to tell everyone what they think. I wish they had that enthusiasm about going to work.
if it was really a 25% chance to win it seems very unlikely that Donald would have won by the electoral margin that he did.
70,000 votes in Pennsylvania, Wisconsin and Michigan. 25% chance doesn't sound off. Pollsters are not fortune tellers. 25% chance is a pretty darn good chance. We're not talking 10 to 1 odds.
Trump is all the Republicans have got going - no one's attracted to THEM.
This is almost certainly true.
And Republicans melt down as easily as Democrats.
I doubt this one. The left generally is more emotionally unstable than the right.
wwww - I recall an amusing NPR interview with Nate and colleagues at one of the conventions. Nate said that Trump had a 25 to 30% chance, and the uneasy host, amusingly and revealingly, asked him how often 25 to 30% events happen. Nate said, of course, 25 to 30% of the time. The host clearly wanted him to say something along the lines of never in a million years.
bagoh20 said...
Then it's a 3 in 4 chance that Bill Maher gets his wish, and the economy tanks. Probably not full tanking, since Trump's policies will still be there for a while, but it will tell business owners, investors, and consumers that their fellow citizen are stupid enough to blow a booming economy and can't be trusted
********************
Just a few weeks ago our financial advisor told me and my wife that if the GOP loses the House he will go "straight to safety" with our investments, for the reasons you cite above. Rescission of Trump's tax cuts alone will put the brakes on economic expansion.
Voter enthusiasm is the entire ball game in low-turnout, off-year House elections: Democrats have it and Republicans don't.
The Crack Emcee said.. I'm changing my political party in a half hour.
Will you Please take Chuck with you?
I don't think the polls can be used any longer to make predictions. If they could, Trump's odds of winning in 2016 would not have been 25% or less on election day. 538's prediction on election day 2016 was based solely on Trump's chances of winning VA, OH, FL, and IA along with the chances of him holding onto NC, GA, and AZ. If Silver had thought at all of Trump winning WI, MI, or PA, he would have put Trump's odds at 50% at least.
The media have sullied their polling with their relentless anti-Trump/anti-Republican rhetoric. I don't answer polls at all, but if I did, I would lie to them just for the amusement. I can't be a tiny minority in that regard.
"Trump is all the Republicans have got going - no one's attracted to THEM."
Yes, the right is all consumed with being attractive. You can tell by how we dominate the fields of fashion, race posturing, Hollywood, and media. It's clear by how we arrange our entire lives to avoid being seen as racist, uneducated, and how we forgive anything if it's clean and articulate, with a crisp pants crease. Yea, that's us hillbilly wingnuts, all struggling to get those flashy graduate degrees and impress everyone with our cultural superiority and wokeness. We are obsessed with everyones' opinion of us.
============NOT=============
I just note for the record that as of this morning, the generic ballot reported on RealClearPolitics range from +4 Dem to +11 Dem. If you know anything about statistics, you know what this indicates about polling.
You can take this one to the bank. 538 has never been wrong, and never will be wrong. NEVER!
Blogger Earnest Prole said...
Voter enthusiasm is the entire ball game in low-turnout, off-year House elections: Democrats have it and Republicans don't.
The first part is certainly correct. The second half has historically been true but I'm not sure it will be that way this year. Most polls show that democratic voters are highly energized. The special elections have all shown very high democratic turnout, along with most of the primaries. Now, in the special elections they each individually received lots of national money and national media attention, all of which will push up turnout. The best thing the Republicans can hope for is democrat fatigue from everlasting Trump Outrage Syndrome. If the Mueller probe wraps up before the election and turns out to be the big fat nothing burger that it looks like now, and the media moves on to some other "OUTRAGE", I could see that having a mass wearying effect on the democrats. Lots of them have become emotionally invested in Mueller somehow getting rid of Trump, if nothing comes from it will be a big let down and will certainly drive down turnout in November.
The website's map titled Our Forecast for Every District is done very well.
Look at each gray-colored spot.
Tim in Vermont said...
Nate's the one that didn't understand probability. He thought that each state was like an independent roll of the die, when in fact, the states all moved together based on an underlying demographic trend that should have been obvious to him.
Nate Silver precisely made this distinction between his models and others. In particular, the Clinton internal polls seemed to treat each state independently and missed the possibility of a rollover effect.
Silver took a lot of flak from the left in 2016 for giving Trump a non 0% chance. Where he admitted failure was in 538's early estimates, where with minimal data, he followed the conventional herd.
* * *
The Vault Dweller said...
If it was really closer to 75%/25% virtually all the probable outcomes of a Donald victory should have been narrow squeakers.
And they were. Electoral College votes don't represent the margins in key states:
Trump won Michigan by 0.3%
Trump won Wisconsin by 1%
Trump won Pennsylvania by 1.2%
Trump won Florida by 1.2%
Basically Trump won every squeaker that mattered.
In close states, those under a 2% margin of victory, Clinton won New Hampshire and Minnesota. Woo Hoo.
The main thing going against the Republicans is that they are happy with what they have right now, becuase it's pretty damned good - historically really good even for the people who hate them. But that satisfaction is not going to get as many to the polls as the painfully inflamed Lefties, who feel like they are having their teeth drilled without novocaine 24/7. They imagine themselves voting against Hitler. You gotta go vote with that attitude, but it's not a good thing to go shopping when you are really hungry, or make an important decision when you're having a tantrum.
"Stupid white people."
You keep posting this phrase Crack, in what I assume is a desperate attempt for more attention.
I have tried to engage you in the past, but you get so hung up on your "angry Black man/stupid White people" schtick that it turned out to be a waste of time.
Do you use this blog as therapy or are you trying to sell something?
I think the polls are going to be wrong once again. People just have a need to run with the crowd except in the privacy of the voting booth.
Don't believe it unless they're taking bets at those odds.
"Stupid white people"---Crack
**************************************
Crack, if *we* are stupid, why aren't *you* and your cohort rich?
yeah, it's been pretty stable for a couple of months now. Those chances are going to go up for Democrats unless something happens. I keep expecting something interesting to happen in Korea that would make it clear that Trump is doing well there. I'm not sure what else is going to change between now and then.
Basically Trump won every squeaker that mattered.
That is kind of my point. He won lots of those squeakers. The way I remember the end of election night was that it came down to Trump could win if he won one of either, Pennsylvania, Michigan, or Wisconsin. And he ended winning all three. If he really only had a 25% chance it seems far more likely that he would have only won one of those. Two is even more unlikely. And three should have been even more unlikely than than that. The fact that he won with most unlikely scenario (of the outcome of those three states) makes me think the initial estimate of a 25% chance to win overall was off.
The Crack Emcee said..
"The Democrats are defeat-able - still - but, if successful, they will have the rose from the grave by the slimmest margin, to actually do damage to Trump where, before, they were simply ass-out. That's what's funny: Stupid white people."
They were never ass-out. Their demographic advantage prevents any ass-outery. Trump identified a new way to win: boost white votes by 4% rather than trying to boost Hispanic support by 70%.
It ain't pretty, but he's just playing by the rules other wrote.
Don't leave.
Francisco D said...
"You keep posting this phrase Crack, in what I assume....."
Stupid white people.
Do you respond to telephone polls? I don't.
I am not represented in these numbers. Are you?
asked him how often 25 to 30% events happen. Nate said, of course, 25 to 30% of the time.
LOL! I love it!
I've been taken aback how people see polls as if they are a certainty. 25% chance means a 25% chance. I love this stuff. Moneyball is one of my favourite movies.
Surely many people have played card games and understand probability? Maybe people want, or even need, to know what is going to happen in the future. Perhaps they feel disturbed that the future is not dependable, so they impose certainty?
Nate on twitter yesterday:
p.s. The chances our model gives Democrats—between 70 and 75 percent depending on what version you look at—are pretty much exactly the odds that Hillary Clinton had of winning in 2016! So hopefully everyone's learned their lesson and won't mistake that for a sure thing. 😂😂😂
Jay Elink said...
"Crack, if *we* are stupid, why aren't *you* and your cohort rich?"
Who said rich correlates with intelligence? Is Ann rich? Are you? Jenny McCarthy has her dressing rooms "cleansed" for spooks and thinks vaccines caused her sons autism - but still has more money than all of us - is she smart?
Stupid white people.
"Stupid White People"
Crack is at least half right. About half of white people are fucking stupid. They are called "Democrats".
538 the day of the 2016 election for "Chance of Winning the Presidency": Hillary Clinton, 71.4% -- Donald Trump, 28.6%
Unless...respondents are lying to them.
What are Nate Silver's odds for the Senate?? I say 75% GOP maintains the majority.
Two points. First, if the generic ballot is less than 5 points in D favor, it really means R lead. Why, because they always oversample D to skew the result they want. In my experience D +5 is the critical line. D lead in generic has declined over the last year, went up slightly during June, and is now resting on the line again. If they are calling it even then it’s definitely R leading. Second, the D talk of impeachment and public disgust with the never-ending lameness if the Russia probe will bring out R voters. Don’t be fooled by the primaries, which energizes voters for different reasons. Of course more D votes in MN when they have five D running off to face an unopposed R. The people who show up in November will be motivated differently.
Republicans increase lead in Senate and keep House majority.
CJinPA said...
"Don't leave."
I'm staying in touch with Ann and Meade, but the Republican Party? No way. MLK said not to, as did my black friends, but I tried it anyway - twice.
White Republicans don't want anyone but extreme ideologues, willing to rip blacks apart to join them, when - like the French - they're proven to be merely fair weather friends, at best.
Leaving the comfort and security of black people for that is madness. Even the suggestion is cruel - especially in a country known to have ripped us apart, literally, in the past.
No, white Republicans are their own worst enemies - and everyone else's, too. I'm not trying to be the first guy on the Dodgers. I just want to live in my country as comfortably as any other American but whites won't have it.
There's no reason for me to keep trying if they can't. I'm off to the Post Office.
Don't take this too seriously. 538 did this for the World Cup. Obviously a lot of its predictions were right and others wrong. It added to the enjoyment of following the tournament but France won on the field.
I don't get why anyone still listens to these people.
I have no college degree and I'm not too good at the maths.
But I can read and know just a tiny bit of history. And history tells us that most of the time, during the president first midterm election, he will lose both the house and Senate, or if his party doesn't already control it, he will lose more seats.
This has happened so often and so regularly throughout history that it was amazing when it didn't happen to Bush. Of course, in hindsight, it's easy to see why, with 9/11 still stuck in our heads and the country afraid for our lives.
But not only is this stupid on stilts, it doesn't even mean anything. If you read Nate Silver, even he will tell you it just means the Dems have a better chance right now.
Duh.
There is only one major purpose to these polls and prognostications. Not to predict the result but to infuence the result. Right now Democrats all over the nation are sitting down and having serious discussions with their allies in silicon valley. Nice Facebook you've got there. Have you seen the polls lately, about how we are going to take over soon? Wouldn't want you to lose that whole YouTube thing because you allowed anti Democrat, pro Republican voices, er, we mean, hate speech, wink.
Well, I said Obama won win twice... and he did.
Good to see that Crack worked out his cognitive dissonance concerning Republicans, what took you so long? Or was it because all the Republicans here made you feel so unwelcome and blamed you for everything any black persons has done wrong, ever?
On the Right, the Republicans have a Big Tent - big enough white racists camp out there.
On the Left, the Democrats hold mystical events - so openminded white's brains fall out.
It's just a fucked-up situation all around.
The Crack Emcee said...
"Leaving the comfort and security of black people for that is madness"
Tribal comfort seems to be what will dictate politics and culture in New America. Like I said, Trump (and the GOP) didn't make those rules, but I guess we're all obligated to play by them.
Crack, most Democrats are not into the mystical stuff, at least not here in Wisconsin.
Probabilistic prediction models mean never having to say you were wrong.
I've yet to hear a meaningful explanation as to why polls for Hillary! were entitled to use Obama's turnout estimates. The black guys at the barber shop are staying home over voting for a sour old white woman. Yet they're going this way again. Just look at all those D's coming out against Roy Moore! It's a sign of the blue wave I tell ya!!!
Nate was less waaaay wrong! than the rest of them but he and the rest of them are in the dog house until they call a few very very right.
I just want to live in my country as comfortably as any other American but whites won't have it.
Let's see a show of white hands. Hmmm. Not seeing it, Crack. My informal survey shows nobody gives a shit how you lead your life so long as you're not hurting anybody else. I mean there are people who want to hold you back and keep you down, but they are largely Progressive and, unfortunately, also Black like you. Every racist I've ever known was a self-identified open-minded liberal, but YMMV.
Congress has the worst negatives of all, but Congressional races routinely go to the incumbent, has 538 taken this into account?
-sw
I would love to bet on that at those odds. Democrat media has pounded relentlessly. Wait til campaign season kicks in and the clueless voter starts seeing GOP tv ads and hears the news about the great economy, low black unemployment, the numerous foreign policy improvements, and the insane Democrat ideas.
Polling right now is like deciding after the plaintiff's case without hearing from the defense.
Actually, self-described LLR Chuck and his "Department of Black People" remark might give Crack the impression that the establishment Republicans are racist. I can't defend those country club people because I don't know any. They do squeeze into the Big Tent, but they are losing their grip on the GOP. My family is full of people who worked for a living and hated golf, but we voted conservative. And at the heart of conservatism is the original purpose of the R party: ending slavery and fulfilling the Declaration's prediction of equality for all. Most political struggles are the D party trying to make things unequal in the name of "fairness" and the R party trying to hew to original principles, once you strip away the bullshit.
What are Nate Silver's odds for the Senate??
..and that's why 538 is buckets of shyte- their raison d'etre is to show lefties what they want to see, skipping the lies and damned lies and going straight to statistics.
Perhaps some odd Wednesday a few weeks from now Nate will be able to demonstrate a slight narrowing in the Senate races that's easily explained by margin of error but the FiveThirtyEight headline will read Senate Republicans in Trouble!. It's a tired shtick...
For every Crack that abandons the Republicans there are ten white people who have come to realize that you have to be a fool to vote Democratic if you are a white person.
Sure you have the crazed liberals, feminazi's and virtue signaling cucks but normal white people are abandoning the Democrats in droves.
We have become a nation obsessed with identity politics and this is the fruit of that labor.
Leaving the comfort and security of black people for that is madness.
Yes, ask the 72 black people who enjoyed comfort and security in Chicago last weekend.
Ask why Reverend Wright, after decades of ranting at white people, moved to Hinsdale among all those whites when he retired.
Good luck.
RealClearPolitics.com lists 199 districts as “leans Democrat” or further. Six have Republican incumbents. 194 are “leans Republican” or further. That leaves 42 in the “tossup” column, and 40 of them are currently held by Republicans. If the Democrats win all of the leaners (they won’t) and 19 or more of the tossups then they win the House. The Republicans need to win all of their leaners (they won’t) and hold at least 24 of 40 seats rated “tossup” to hold the House.
So, reasons why Silver is probably wrong include:
(1) We don’t know the value of incumbency in 2018. Republicans gave their constituents a tax cut, a strong economy, and real wage growth that is weighted towards the working class (in contrast to Bill Clinton’s dot-com boom, which was weighted towards the upper end of the scale). If the incumbents have been doing a good job with constituent service then that should be more than enough to see them re-elected.
(2) Republican incumbents are in the unique position of being able to run against both Pelosi and Ryan, should they choose.
(3) Related to (2), the Democrat faces they see in the news — Pelosi, Ocasio-Cortez, Sanders, Andrew Cuomo — are bat-shit crazy. Nationalizing the election could be a good strategy.
(4) Trump
(5) Reaction to lawlessness on the left, including Crooked Hillary, crooked FBI and CIA directors, Antifa, etc.
(6) The DNC is both broken and broke.
But reasons why Silver could be right include:
(1) Democrats are frothing at the mouth to vote against Trump (well they are always frothing at the mouth about something or another, but this year, like in the local 2017 elections here in Virginia, it’s Donald Trump).
(2) Thanks to Reince Priebus, Republican get-out-the-vote efforts almost matched Democrats in 2016. Mitt Romney’s niece ain’t no Reince Priebus.
98% chance that the People Who Have Partaken of the Red Liquid from the Dark Sarcophagus will be in charge.
Is it worth fighting for that other 2%, or just give in?
I honestly don't see the Democrats meriting those kinds of odds for a takeover of the House.
For instance, they have the Michigan 11th -- a perfect bellwether House district for much of the Midwest and suburban American -- as likely going to the Democrats. (It's close, and FiveThirtyEight knows it, of course.) And my bet in the 11th is on Lena Epstein, the Trumpist Republican, holding that seat. The district was crafted by Republicans to give us about a +4 advantage.
If we do lose that seat, then it really will be a shitty election night for Trump.
From Thomas Sowell,
"By projecting a vision of a world in which the problems of blacks are consequences of the actions of whites, either immediately or in times past, white liberals have provided a blanket excuse for shortcomings and even crimes by blacks."
meanwhile cnn is pushing to release the names of the jurors in the manafort case, they cover only somethings with a pillow.
@Crack, we’ll save a chair for you, if you want to come back. But I don’t think we’ll miss you while you’re gone. Sorry.
Cracker Racist magically appears on a news item that shows his beloved Dhimmicrats are coming back to power. Aint that swell!
Oh, and Crack, you might want to read Whistling Vivaldi, by Claude Steele.
Blogspot really needs a "twit" filter.
Actually, the chance Republicans keep control at 538 varies depending on which of the three models you pick:
32.1% - Lite - "Keep it simple, please — give me the best forecast you can based on what local and national polls say
25.4% - Classic - I’ll take the polls, plus all the “fundamentals”: fundraising, past voting in the district, historical trends and more
31.1% - Deluxe - Gimme the works — the Classic forecasts plus experts’ ratings
Predictit is putting the odds at 36% (with a 3% margin for arbitrage). That's more in line with 538 Lite and Deluxe.
At the race level, 538 is predicting two incumbent Republicans will lose in Iowa. Are Iowans really that disgruntled with the Republican Party? Republican Governor Kim Reynolds is still expected to win.
In any case, we can argue whether we are flipping one coin or two, but the U.S. House is a coin toss, and Democrats may be setting themselves up for another tough beat.
@The Vault Dweller: And he ended winning all three. If he really only had a 25% chance it seems far more likely that he would have only won one of those. Two is even more unlikely. And three should have been even more unlikely than than that. The fact that he won with most unlikely scenario (of the outcome of those three states) makes me think the initial estimate of a 25% chance to win overall was off.
If he had 3 50/50 races--and those three were pretty close to that for him, given his margin--his chance of winning all three is actually 12.5%. (1/2)*(1/2)*(1/2). (Exactly the same as his chance of losing all three.)
And if Nate Silver had said exactly that--Trump needs 3 to win, and his chance of doing that is 12.5%, you would be saying now that the odds couldn't have been that bad if he won all three. But that's not true.
And if he'd lost all three, in that situation, you'd have said "well 12.5% chance to win all three of course he lost them"--even those losing all three was exactly as likely as winning all three, nothing has changed in the math.
People do not have a a good intuition about probability. You have to work it out.
I suspect Fivethirtyeight has not factored in Trump doing 6 big Rallies a week during two months before the day. That's easy. Remember President Trump was doing 5 a day in the weeks up to the 2016 election. DRAGON ENERGY!
Inga trying to recruit Crack is funny. Like, 70s sitcom funny.
No path to 270.
Stupid white people.
Yeah. I can see why so many people missed his valuable contributions when he was gone.
Does President Hillary Clinton know about this polling info?
I hope everyone knows that The Crack Emcee is a white teenager.
One thing I wonder, particularly in the House, is how many people split their vote and voted for Clinton for President, and then a Republican congressman to restrain her worst impulses, in the mistaken belief that she was a lock to win the Presidency. My recollection is that in some swing states, she actually ran ahead of the local Democratic candidate for the Senate. In Ohio, for example, she got almost 400,000 votes more than the Democratic candidate for Senate (Ted Strickland). In Wisconsin, she got . . . about 2,000 votes more than the Senate candidate (Russ Feingold). But Trump ran about 70,000 votes behind Ron Johnson. I was going to use Pennsylvania as an example too, but actually both she and Trump ran ahead of their parties' Senate candidates, so it doesn't really work.
Anyhow, I don't know the statistics on ticket-splitting in 2016, but I can easily imagine some strategic voting on the mistaken premise that Clinton was going to be the next President. That would tend to "correct" during this electoral cycle.
I'm not buying the idee fixe that Republicans are not enthusiastic and won't turn out. Many Republican voters, including yours truly, are far more supportive of Trump than they were on 11/8/2016. The constant state of siege mounted by the leftist culture warriors has inspired us to fight back. We have no respect for polls, least of all 538, so we don't take surveys.
Here's my prediction. Republicans lose seats in the House but not the majority. Republicans gain seats in the Senate. Remember, you saw it here first.
Re: Gabriel:
People do not have a a good intuition about probability. You have to work it out.
Off topic, but I feel like this is a real problem for lawyers who are trying to evaluate evidence (and probably for juries trying to weigh evidence). Suppose you have to establish three factors for a man to be guilty: A, B, and C. And they're all independent. And you're 75% confident each factor is met individually. That sounds like you're pretty confident! But that would imply 0.75^3 = only a 42% chance that all three requirements are met such that he's actually guilty. How many lawyers actually think through the weight of evidence in multi-factor tests logically? Very few, I would guess.
We need Mick.
He would know.
so you're telling me there's a chance
@Balfegor: How many lawyers actually think through the weight of evidence in multi-factor tests logically? Very few, I would guess.
Made even harder because in real life probabilities relevant to criminal cases are rarely known numerically and precisely; and often they're not independent.
Suppose you know a murder was committed by a) a man b) aged 42 years old c) height 5' 10" who d) lived in the same apartment complex as the victim--all of which characteristics are shared by the gentleman in the dock. But a), b) and c) are not statistically independent. And if the apartment complex was for middle-aged men only, well...
Nate can factor normal elections. How does he factor
a) insanity. The positions and rhetoric of many Democratic candidates is...offputting to normal people. So Dems will win Stupid Ignorant 90% Dem Districts like they always will...but the moderate areas are not likely to be enthusiastic for voting for a Socialist as they read about the meltdown in Venezuela.
b) Republicans and Conservatives don't talk to Nate Silver. We just don't. We don't like the media, have had our fill of push polling and I personally have hung up on three pollsters this year.
This is data that Nate Silver misses. And those people slamming the phone down? They are PASSIONATE people. They are 'crawl over broken glass' Republicans.
It is an entire data pool Nate has very little entry into.
But I could be wrong. Maybe he knows how much spin to cypher out all the hang up calls or 'hide the decline' like Anglia-gate.
Which makes his estimates less accurate suppositions.
Personally, I HOPE he is wrong, but I am open to a split Congress.
Because Trump still has the Pimp Hand.
But, but, but: what about the Rasmussen poll yesterday that had black approval of Trump up to 36% from 19% three months ago. If - big if -the Dems have the House I am willing to bet that the R's end up with 57 seats in the Senate. Anything coming from the House will die in the Senate. Court appointments will continue to be made, trade treaties have nothing to do with the House and those are the next big things on the agenda.
Do you really think that all thoseTrump voters whose lives are being made better by Trump's policies are going to be stupid enough to stay home and not vote? Really?
Trump will nationalize the elections in order to up turnout. Do you think the rallies he has been holding are just for his own entertainment? What if Kavenaugh has not been confirmed; what's the impact going to be? How about when RBG is hospitalized just before court opens?
I think polling is just as abysmal now as it was in 2016.
"Off topic, but I feel like this is a real problem for lawyers who are trying to evaluate evidence (and probably for juries trying to weigh evidence). Suppose you have to establish three factors for a man to be guilty: A, B, and C. And they're all independent. And you're 75% confident each factor is met individually. That sounds like you're pretty confident! But that would imply 0.75^3 = only a 42% chance that all three requirements are met such that he's actually guilty."
-- ... That sounds like trying to determine how many wounds a gun line of dwarves will actually get through before a death block of infantry closes on them. "First, To Hit, then To Wound, then Saves."
I, for one , am much more positive about Trump and his administration and what's been accomplished than I was in 2016. In 2016 I voted against Hillary - absolutely - in 2018 I am voting for the way things are going in the country right now. I agree with Tyrone Slothrop that there are many like us who have gone from marginally positive to fully on board.
Meanwhile the population is not reminded of
The Attempted Democrat Assassin.
Kathy Griffin
Antifa
The Various Anti Free Speech Statements by Liberal Academics.
Crumbs Pelosi
Andrew "America ain't that Hot' Cuomo
Anti-ICE Democrats
A few racial atrocities by illegal immigrants
And some Trump tweets which engage the voters.
And what about this headline in the WSJ today U.S., China Plot Road Map to Resolve Trade Dispute by November? Won't that have an impact?
There's a whole lot of old man yells at cloud in this thread.
Anyone forecasting political events to tenths of a percent is fooling themselves even more than they're fooling you.
To paraphrase Nassim Taleb, the chance of the Dems taking the House is 50/50. Either they do it, or they don't.
That's all you can say right now if you're not selling snake oil.
Unfortunately, fake news runs on speculation and predictions for which no one will ever be held accountable.
It's pretty much snake oil 24/7.
The Comfort and Security of Black People ... come for the poverty, stay for the murderous violence!
Didn't realize they were still around.
"Just a few weeks ago our financial advisor told me and my wife that if the GOP loses the House he will go "straight to safety" with our investments, "
You should find a different financial advisor.
Damn this sounds familiar.
@Stan - perhaps Americans pine for higher taxes, more illegal immigration, more mandatory homosexuality and more pandering to foreigners. But I am guessing no.
"Just a few weeks ago our financial advisor told me and my wife that if the GOP loses the House he will go "straight to safety" with our investments, "
You should find a different financial advisor.
I don't know. The Democrats would like to crash the economy before 2020. If they get the House they might get a chance to do so.
If you think they give a shit about the American people, at least the bottom 99%, I have a bridge for you.
Numerate people know that, aside from all the other issues, giving numbers to parts in a thousand is lying. Silver is lying to you.
@Gabriel
If he had 3 50/50 races--and those three were pretty close to that for him, given his margin--his chance of winning all three is actually 12.5%. (1/2)*(1/2)*(1/2). (Exactly the same as his chance of losing all three.)
Now I might be wrong about this it has been a while since I've studied stats. But if Trump was in a position that he needed only 1 of those states to win the way to calculate the probably of that would be 1 - (probability of losing all 3 states) and if assuming each state is roughly 50% chance either way and they are independent of one another, the chance to win at least 1 state would be (1 - 12.5%) or 87.5% Now that percentage also includes the chance to win all 3 states, but as you calculated above that is just 12.5% of that total, which leaves 75% it was another combination other than winning all three. Of that 75% there are 3 combinations of him winning 1 of each state he needed to win individually and 3 combinations of him winning 2 of those states he needed to win as a combination of the three. So my point is that because he happened land in that one 12.5% segment of winning all three instead of the 75% segment of the other two combination possibilities of win suggests to me that the initial assment was off. Now I know it isn't an unheard of possibility, it is after all 1/8th. I just tick that result off as a factor in giving me my hunch that the initial polling was off.
@Ken B, I know where you're coming from. In mathematics and statistics I've heard it called the "Chinese Emperor's Nose Fallacy." You ask hundreds, maybe thousands, of people to estimate the length of the Chinese Emperor's nose. You collect the data, average the results, and calculate an estimated length for his nose to as many digits as your computer supports.
But, you see, no one is actually allowed to look at the face of the Chinese Emperor, so you are averaging hundreds or thousands or wild guesses, and the result is a wild guess, but to three or four or more significant digits.
@Grabriel
Oh sorry slight follow up, in my above post we were talking the chance of him winning all 3 states assuming a roughly 50% chance of him winning, and that percentage was 12.5% Now if the reality of the probabilities were closer to the polls where it was 90% Hillary 10% Trump then the probability of him winning all three of those states was .1%. And that is my main point in that because of the margin of his victory in the electoral college I assume the probability of him winning is much closer to a 50% 50% split than 90/10 and 75/25 predictions we had seen from other pollsters.
@Vault Dweller, the problem you are having is that the three events are not independent nor are they uncorrelated. Let’s look at the conditional probabilities. For the sake of argument, lef’s accept that the a priori probabilities of winning Pennsylvania, Wisconsin, and Michigan are each 10%. Now let’s look at conditional probabilities. Would not you say that the probability of winning Pennsylvania given that Trump won Michigan is much, much higher? Maybe closer to 80-90%? And the probability of winning Michigan goes up if Trump has won Pennsylvania, though certainly not as dramatically.
But my biggest problem with Nate Silver comes when he looks at historical data as part of his analysis, because I think we are living in times that are fairly unprecedented. For instance, one of the reasons that the party holding the presidency loses seats in Congress two years later is that the new president’s coattails pull in some weak candidates that lose their first reelection campaign. But the GOP actually lost a few seats in the House in 2016, so it’s unclear — at best — what Trump’s coattails were like. Put another way, a Congress Critter who won in 2014 and was re-elected in 2016 is not as vulnerable in 2018 as someone who won for the first time in 2016 and is facing his or her first reelection in 2018.
See also my comment at 12:18.
CJinPA said...
"Tribal comfort seems to be what will dictate politics and culture in New America."
Blacks left the South looking for the law. Whites made sure we never found it. You can't expect any trust after that.
"Trump (and the GOP) didn't make those rules, but I guess we're all obligated to play by them."
See how you lied, to yourself, there? If whites made sure there was no law for blacks leaving the South - and Trump first came to public notice in a trial for discriminating against blacks in his buildings - how didn't whites make "those rules" we are "all obligated to play by" now? I'll tell you how:
You let yourselves off the hook for making the rules others live by.
It's hard to deal with people who are so dishonest, it carries over even to themselves.
Doug said...
"The Comfort and Security of Black People ... come for the poverty, stay for the murderous violence!"
And the inability to hear white men talking shit.
It's a blessing.
@ Crack - I think that’s wrong. I think that Blacks left the South looking for economic opportunity.
@THe Valut Dweller:nd that is my main point in that because of the margin of his victory in the electoral college I assume the probability of him winning is much closer to a 50% 50% split than 90/10 and 75/25 predictions we had seen from other pollsters.
But we did the math. Probability does not work that way. When we started with 3 50/50 races we got a 12.5% chance of winning all three. Winning all three was PERFECTLY consistent with the odds being 50/50 on each one. Things that have a 12.5% chance of happening, happen all time. 12.5% of the time.
You have a 1/6 (17%) chance of rolling a 6 in one try. If you do it, and you roll a six, you are saying that you now have evidence that the odds must really have been higher than 1/6 or you couldn't possible have done it in one try. That's not how probability works, and casinos and state lotteries and people who run sports betting could not possibly make money, if probability worked the way you argue here that it does.
Post a Comment