Slashdot Mirror

Bringing up 538 is pretty random. But since you did. He was the one pollster who consistently said that Trump had at least a 30% chance of winning.

trump-is-just-a-normal-polling-error-behind-clinton

nate-silver-fivethirtyeight-trump-forecast

nate-silver-warns-media-against-dangerous-assumption-trump-isnt-really-closing-in-on-hillary

election-update-why-our-model-is-more-bullish-than-others-on-trump

nate silver forecasts showing clinton with 99 chance of winning dont pass commonsense test

nate silver projects trump will win florida

nate silver 511 chance trump-winning-if-election-held-today

nate silvers terrifying-prediction prepare president-trump

election update as the race-tightens-dont-assume-the-electoral-college-will-save-clinton

Bringing up 538 is pretty random. But since you did. He was the one pollster who consistently said that Trump had at least a 30% chance of winning.

trump-is-just-a-normal-polling-error-behind-clinton

nate-silver-fivethirtyeight-trump-forecast

nate-silver-warns-media-against-dangerous-assumption-trump-isnt-really-closing-in-on-hillary

election-update-why-our-model-is-more-bullish-than-others-on-trump

nate silver forecasts showing clinton with 99 chance of winning dont pass commonsense test

nate silver projects trump will win florida

nate silver 511 chance trump-winning-if-election-held-today

nate silvers terrifying-prediction prepare president-trump

election update as the race-tightens-dont-assume-the-electoral-college-will-save-clinton

You're basically right. I picked colon cancer because that is the one major cancer in which screening really can result in a cure.

The way I learned it, the cancers appear in the inner layer of the colon, and they progress to the outer layers and finally the surrounding tissue. Doctors can screen for them with colonoscopy, and the initial cancer can be removed, sometimes directly during the colonoscopy, and sometimes in a separate operation which might remove more of the colon. Once it spreads outside the colon, though, it can't be cured by surgery any more, and there are no drugs that can cure colon cancer.

I think I've seen long-term studies which showed that people who have the cancer surgically removed in the early stages ultimately live a normal life span, whereas people with the same early stage cancer who don't have it removed go on to develop later stage cancer and die of it within 5-10 years. I admit I can't quickly pull up a citation right now.

Christie Aschwanden had a good story about this in FiveThirtyEight, although it's one of those stories that makes you work for the reward at the end.

http://fivethirtyeight.com/fea...
The Case Against Early Cancer Detection
By Christie Aschwanden
FiveThirtyEight
Nov 24, 2014

Papillary tumors are like turtles -- they move very slowly and never pose an escape risk. They don't need screening, because they will never cause trouble. Then there are rabbits, which are eager to hop away to other parts of the body, but can be confined if they're found and fenced. These are the cancers that can be helped by early detection and treatment. Birds, on the other hand, are so flighty and quick that they can't be confined. Screening makes no difference for bird cancers, because they're so aggressive that they can't be detected before they've begun their deadly course.

No cancer screening has ever eliminated the majority of cancer deaths. Instead, the best screening can do is reign in the rabbits. Birds remain unstoppable, and they're the ones responsible for most cancer deaths

You haven't provided any evidence whatsoever to support the idea that there would be any bias. You have no indication that your selection of incidents is not itself biased (because it very much is).

Doing studies on these kind of issues is pretty difficult. We cannot directly measure bias like we can temperature. Any study is necessarily going to have to codify assumptions about what is bias, and selecting the dataset is also an excellent way to bias the results in one direction or another. Do you just look at account closures, or do you try to examine the content as well? How do you determine political leaning? The link is a demonstration of why I would examine such studies very carefully.

But I admit, even a bad study would be better evidence than cherry-picking incidents or your bare repeated assertion of bias. I am sure you are aware that anecdotes are not data. I understand that there has been some recent and widely reported brouhaha about this issue, and it may even be true, but I am an empiricist: show me the data. A bunch of people repeating each other is useless.

It's great what allegations can do though. If they admit they're biased, people will stop using their service. If they don't they're liars. And why are we concerned about Twitter's bias, and why right now? And why is Facebook's admission that their ToS is only for little people not equally concerning? Did you actually read the article, by chance? It's really not something I would spend a lot of time defending.

This is getting dull. This story proves nothing but that conservatives are easy to rile.

This study proves that working for a publicity-hungry quack clinic damages your ability to distinguish between association and causation.

The author http://www.amenclinics.com/sta... works for a clinic http://www.amenclinics.com/ that sells dubious treatments based on dubious SPECT diagnoses.

Quackwatch has this to say:

https://www.quackwatch.org/06R... A Skeptical View of SPECT Scans and Dr. Daniel Amen by Harriet Hall, M.D.

I believe it is improper to charge thousands of dollars for a test that has not been validated and may not be safe. I don't think any of Amen's research has provided clear evidence that patients who have had SPECT scans have superior clinical outcomes to adequately treated patients who have not been scanned. That's really the bottom lineâ"especially with an expensive test that involves significant radiation. At the very least, he should be describing the test as experimental.

Some of Dr. Amen's treatment suggestions also worry me. For example, he recommends: (a) uses for dietary supplements that are not supported by good evidence, (b) EMDR (a highly questionable approach), and (c) hyperbaric oxygen therapy for conditions not generally considered to warrant such therapy.

I don't doubt that many patients who visit the Amen Clinics are helped. The key question, however, is whether or not SPECT scanning is justifiable for most of them. I, personally, would not undergo the test at Dr. Amen's clinic even if it were free. In my opinion, based on current knowledge, the possibility of harm outweighs any potential benefit. Pictures showing that "this is your brain on drugs" may impress some people, but I am far more impressed by quantifiable data (such as tests of mental performance) and clinical consequences (such as improved behavior) than by nonspecific pictures of "holes" in the brain.

So this is an operation that is selling diagnoses and treatments not supported by legitimate scientific research. They wound up with thousands of SPECT scans and decided to do some data-dredging on them, a process that we know is guaranteed to produce false positives http://fivethirtyeight.com/fea... https://xkcd.com/882/ , along with any real causative association. They found an association with marijuana, and rushed to publish.

Once it was published in a journal, they made claims in the press release that weren't supported by the data:

According to Daniel Amen, M.D., Founder of Amen Clinics, "Our research demonstrates that marijuana can have significant negative effects on brain function. The media has given the general impression that marijuana is a safe recreational drug, this research directly challenges that notion. In another new study just released, researchers showed that marijuana use tripled the risk of psychosis. Caution is clearly in order."

Clearly false. Association is not causation.

Well played, sir. I looked at the quackwatch site, which had this quote from the Amen Clinics site "Brain-Soul connection." and found that sufficient. Not that that's not an interesting question in general; just that this current research direction seems no more likely to elucidate it than any of the others pursued over the previous 10,000 years.

This study proves that working for a publicity-hungry quack clinic damages your ability to distinguish between association and causation.

The author http://www.amenclinics.com/sta... works for a clinic http://www.amenclinics.com/ that sells dubious treatments based on dubious SPECT diagnoses.

Quackwatch has this to say:

https://www.quackwatch.org/06R...
A Skeptical View of SPECT Scans and Dr. Daniel Amen
by Harriet Hall, M.D.

I believe it is improper to charge thousands of dollars for a test that has not been validated and may not be safe. I don't think any of Amen's research has provided clear evidence that patients who have had SPECT scans have superior clinical outcomes to adequately treated patients who have not been scanned. That's really the bottom lineâ"especially with an expensive test that involves significant radiation. At the very least, he should be describing the test as experimental.

Some of Dr. Amen's treatment suggestions also worry me. For example, he recommends: (a) uses for dietary supplements that are not supported by good evidence, (b) EMDR (a highly questionable approach), and (c) hyperbaric oxygen therapy for conditions not generally considered to warrant such therapy.

I don't doubt that many patients who visit the Amen Clinics are helped. The key question, however, is whether or not SPECT scanning is justifiable for most of them. I, personally, would not undergo the test at Dr. Amen's clinic even if it were free. In my opinion, based on current knowledge, the possibility of harm outweighs any potential benefit. Pictures showing that "this is your brain on drugs" may impress some people, but I am far more impressed by quantifiable data (such as tests of mental performance) and clinical consequences (such as improved behavior) than by nonspecific pictures of "holes" in the brain.

So this is an operation that is selling diagnoses and treatments not supported by legitimate scientific research. They wound up with thousands of SPECT scans and decided to do some data-dredging on them, a process that we know is guaranteed to produce false positives http://fivethirtyeight.com/fea... https://xkcd.com/882/ , along with any real causative association. They found an association with marijuana, and rushed to publish.

Once it was published in a journal, they made claims in the press release that weren't supported by the data:

According to Daniel Amen, M.D., Founder of Amen Clinics, "Our research demonstrates that marijuana can have significant negative effects on brain function. The media has given the general impression that marijuana is a safe recreational drug, this research directly challenges that notion. In another new study just released, researchers showed that marijuana use tripled the risk of psychosis. Caution is clearly in order."

Clearly false. Association is not causation.

Oh, we believe she prefers Hillary over Trump, no question about that. But that preference is the real motive for doing it, or she'd want an audit in states like NH, as well, which was won by just 2,732 votes. Of course, Hillary won that state, so we don't care if that result was fair or not, right? And MI that was recounted? They use nothing but paper ballots.

We know why she lost and it wasn't "hacking" as some of the #fakenews has been pushing lately -
http://fivethirtyeight.com/features/demographics-not-hacking-explain-the-election-results/

> When you try to turn this into a "smart" vs "not smart" you are asking for trouble
Actually Nate Silver did some analysis showing Clinton surged in the 50 most educated counties, and collapsed in the 50 least educated counties. http://fivethirtyeight.com/fea...

Nate Silver has pretty much debunked this one already: http://fivethirtyeight.com/features/demographics-not-hacking-explain-the-election-results/

Everything was rigged to make Hillary look better than reality,

That makes no sense, sorry. Why would they do that? If anything, they'd want to do the opposite, to get their vote out.

But bhe best pre-election analysis showed that Hillary's estimated lead was roughly equal to the statistical error in the polling, and the post-election analysis pretty much confirms this. No need for a wacky conspiracy theory.
http://fivethirtyeight.com/fea...
http://fivethirtyeight.com/fea...

Everything was rigged to make Hillary look better than reality,

That makes no sense, sorry. Why would they do that? If anything, they'd want to do the opposite, to get their vote out.

But bhe best pre-election analysis showed that Hillary's estimated lead was roughly equal to the statistical error in the polling, and the post-election analysis pretty much confirms this. No need for a wacky conspiracy theory.
http://fivethirtyeight.com/fea...
http://fivethirtyeight.com/fea...

It's worth noting that if you actually read the article, he doesn't say that the ballots actually were hacked: in fact, what he says is "Were this year’s deviations from pre-election polls the results of a cyberattack? Probably not. I believe the most likely explanation is that the polls were systematically wrong, rather than that the election was hacked."

If you were talking about one or a few polls I would agree with your premise that we should not investigate these allegations. However, it is valid to notice that every single poll agency in the United States, including the agency hired by Trump, they all failed to even come closed to their statistical predictions.

Nope. The analysis by 538 published before the election was that Clinton was ahead by a number that was equal to the polling margin of error. "it shouldn’t be hard to see how Clinton could lose. She’s up by about 3 percentage points nationally, and 3-point polling errors happen fairly often, including in the last two federal elections." (November 6: http://fivethirtyeight.com/fea... )

Another one worth reading: http://fivethirtyeight.com/fea...

It's worth noting that if you actually read the article, he doesn't say that the ballots actually were hacked: in fact, what he says is "Were this year’s deviations from pre-election polls the results of a cyberattack? Probably not. I believe the most likely explanation is that the polls were systematically wrong, rather than that the election was hacked."

If you were talking about one or a few polls I would agree with your premise that we should not investigate these allegations. However, it is valid to notice that every single poll agency in the United States, including the agency hired by Trump, they all failed to even come closed to their statistical predictions.

Nope. The analysis by 538 published before the election was that Clinton was ahead by a number that was equal to the polling margin of error. "it shouldn’t be hard to see how Clinton could lose. She’s up by about 3 percentage points nationally, and 3-point polling errors happen fairly often, including in the last two federal elections." (November 6: http://fivethirtyeight.com/fea... )

Another one worth reading: http://fivethirtyeight.com/fea...

Polls were rigged by oversampling democrats vs. republicans/independents so many were flat wrong. Aggregate sites like 538 were wrong.

Um. 538 predicted a 3% lead for Clinton in the popular vote. In the end she won the popular vote by 2%. If just 1% of voters preferred Trump but were embarrassed to admit it, then the polls would have sampled perfectly.

538 predicted Trump had a 29% chance of winning. In the end, he won. If something has a predicted 29% chance of happening and it happens, that's not a failure in the prediction.

http://fivethirtyeight.com/features/demographics-not-hacking-explain-the-election-results/

Here you go.

Nate has since turned it into a full article:

http://fivethirtyeight.com/features/demographics-not-hacking-explain-the-election-results/

Education big time http://fivethirtyeight.com/fea...

Wrong. White women without college degrees chose Trump . In fact, educated folks generally voted for Clinton, regardless of race or gender.

There's fairly good evidence that October Surprises don't have any real effect. By the end of the election most people have already made up their mind. Wikileaks and Comey had no effect on the election, no matter how much the DNC wants to blame someone other than themselves.

Are these the same polling experts that predicted that Clinton would comfortably win the election?

I feel like the story of the pollsters being wrong is somewhat distorted.
Just before the election Five Thirty Eight was predicting about a 70% chance Clinton would win, but noted

The track record of polling in American presidential elections is pretty good but a long way from perfect, and errors in the range of 3 percentage points have been somewhat common in the historical record. . . .
In three of the last five presidential elections, in other words, there was a polling error the size of which would approximately wipe out Clinton’s popular vote lead — or alternatively, if the error were in her favor, turn a solid victory into a near-landslide margin of 6 to 8 percentage points.

Yes, though she also didn't do quite well enough in Florida. Florida was achievable, with just a little more turnout for the Democrats. But Florida alone wouldn't have saved her; it would just have been insurance against losing one of the three "defectors" (as ultimately she did).

But the three defecting states - Pennsylvania, Wisconsin, and Michigan (still not officially called, but it no longer matters) - would have been enough to put her over the top (228 + 46 for 274 total). And those three hadn't voted for a Republican Presidential candidate in 28 years.

Now, various people have been saying for a long time that the Democrats' "firewall" was much weaker than popularly believed. But given the very, very close totals in those states, it should have been possible to retain them with better campaigning. For the most part (and particularly with Wisconsin, a state the Clinton campaign basically ignored - though voter-impeding by the state Republican government hurt too), the Democrats didn't do a good job of maintaining their constituencies there, or reaching out to undecided voters. And the last-ditch scramble in Michigan looked desperate and seems to have fired up rural Republican voters much more than urban Democrat ones.

She lost by a couple percent. There are a lot of things she could have done differently that would have swayed things in her direction (Nate Silver says it well).

That's evidence-free. Seems to be a pattern with you.
Actual evidence: Trump voters in the primary had average incomes roughly $16K higher than the median income and over $10K higher than democratic primary voters.

Trump's ascendancy is primarily about racism. For decades now its been a truism that being a republican doesn't make you racist, but if you are a racist then you are probably a republican. Trump was just the first to finally take everything the GOP has been saying in subtext and make it text.

Most everything you've said is wrong:

Polls say Hillary can win even without PA.
Hillary doesn't need Florida to win, either.
Hillary doesn't even need North Carolina to win, though one of those 3 is necessary. PA is heavily in her favor, and she's ahead in the other two as well.

Trump needs to win all THREE above listed swing states (which are polling against him) to even stay alive, and that's assuming he wins every other close state, too.

See:
http://projects.fivethirtyeigh...

As of this moment Florida is 54% blue and I expect that to go up. It was around 77% blue before the Weiner emails and it's headed back there now that she's cleared.

In your dreams. 538 has him at 35% probability of winning as I write this;

I'm not a great statistician, but a 1 in 3 chance looks like a fairly high likelihood to me. Silver has hedged his guess six ways to Sunday because of the inherent unpredictability and volatility of the situation. I think there's grounds to be confident of a popular vote and electoral college win, but that's not to say that we should assume they'll actually happen.

One item that's of greater concern to me is the slightly greater than 1 in 10 chance that a recount will be required in one state's results. If we get another Florida, I don't think we can trust the people involved to step back from the brink. Much as people love to revile Gore, his decision to accept a very flawed result saved American democracy—or at least allowed it to survive a little longer. Likewise with the Kennedy election, when Nixon's strategists decided that it made more sense to fight another day.

Clinton's 65% rating seems to show how low its possible for her to fall. So yes, the odds are in her favour. She seems to have weathered the absolute worst the opposition could throw at her. But stranger things have happened than a Trump upset. The Cubs somehow came back from a 3-1 deficit.

If Trump wins, then God help you all. It looks like that may not happen. But don't count your chickens just yet.

In your dreams. 538 has him at 35% probability of winning as I write this;

And that's only naively counting the electoral college. You think they are going to let Trump-the-Pussygrabber onto the highest seat in the land? Not likely. How many Republican party luminaries have already stated "country over party" and refused to back Trump? This many! This election is the perfect excuse for the electors to vote their consciences. If they need tom which is doubtful.

And then there's a disparity in the women's vote: currently standing near 50% in favor of Clinton.

There's no way Trump can win. None. You're just fooling yourself. Enjoy it while you can, because come Wednesday morning, we'll have a woman in the white house. And Washington will go on just as it always has, Republicans blocking every bit of progress they can, democrats fumbling around in a non-unified manner, and idiots like the tea partiers throwing monkey wrenches in everywhere they can, and all the while the lobbyists will be proffering handfuls of money to everyone in sigh, the president not fulfilling a plugged nickle's worth of the promises made on the campaign trail.

You come back here Wednesday, you'll see I called it exactly. If you can find this post after the load of butt-hurt moderation falls on it, lol

have left

You've been misinformed. Hillary is in the lead by a mile. http://projects.fivethirtyeigh... You should be talking about all that Trump has left, which is apparently, his Russian connections, and an army of alt-rightists. Both will lose interest in Trump when loses the election.

No, it is most definitely not a mere average of polls. It's a weighted average, and almost every poll was given 0.00 weight. In most cases, that is to reflect the fact that "that was then, this is now" but in many others it is to reflect the fact that the authors of the poll were flaming idiots.

http://projects.fivethirtyeigh...

I am not a Trump supporter, but I do not want the FBI and the AG to drop this investigation. It's clear that Hillary is guilty of breaking multiple laws, but because her party has power in the executive branch, she's not being held accountable to the degree that anyone else in the country would be.

My solution for the whole thing is to not put up Trump signs, but to put up "Hillary for Prison 2016" signs. She'll make a terrible president. Trump will make a terrible president. What I'm secretly hoping for is that McMullin figures out how to sneak in, people take Kotlikoff seriously as a write-in candidate, or that something terrible happens to Trump/Clinton when they win and the VP has to take over.

Can your read the statue, or your just making a snap judgement.

I am not a Trump supporter, but I do not want the FBI and the AG to drop this investigation. It's clear that Hillary is guilty of breaking multiple laws, but because her party has power in the executive branch, she's not being held accountable to the degree that anyone else in the country would be.

My solution for the whole thing is to not put up Trump signs, but to put up "Hillary for Prison 2016" signs. She'll make a terrible president. Trump will make a terrible president. What I'm secretly hoping for is that McMullin figures out how to sneak in, people take Kotlikoff seriously as a write-in candidate, or that something terrible happens to Trump/Clinton when they win and the VP has to take over.

By definition, Trump can either have extreme opinions or he can represent the majority view, but he can't do both. Getting elected is the art of getting lots of people from the middle of the political spectrum to agree with you.

Getting nominated of course is a different matter. To get nominated you just have to get a plurality of a subset of voters, and ones predisposed to agree with you at that. News organizations were also predisposed to like Trump. He sells a lot of newspapers, and drives a lot of pageviews. However, he seems to have mistaken "getting attention" for "getting votes", and even if everything else were going his way, what drives the media out of his corner is his decision to attack them for anything resembling negative coverage. Threatening lawsuits is probably not a good move there. Saying that as President you would push for more expansive libel laws is flat-out stupid.

Trump is socially pretty extreme. That's why people know who he is. It's possible to be socially outré as a politician (Churchill comes to mind), but pretty difficult. His politics are also pretty extreme, and that puts him at a mathematical disadvantage with the electorate. However, if there is a media conspiracy against him, [1] they don't have much work to do, given the above, and [2] he should probably have gone for a strategy of appeasement rather than aggression. There is an appropriate phrase here: "Never argue with a man who buys ink by the barrel." Given that Trump continues to ramp-up his anti-media rhetoric, are you really surprised that the media is less inclined to support him?

> actually seems to garner extra support from their followers by being outlandish.

I wouldn't say extra support. Trump absolutely knows his fanbase, the reality-tv loving, racially insecure (but not financially insecure, trump primary voters average $11K more in yearly income than both clinton and sanders primary voters) authoritarian-leaning types. For them it is not about policy, its about the feels and he gives them the best feels. But they only make up about 40% of the republican party, and everybody else is pretty much grossed out instead of turned on by that shit.

I think this is the key.

Ever since Trump went after the Khan family it's been obvious that the general electorate does not like Trump going after private citizens.

Yet that's exactly what he did after the first debate with Alicia Machado and again after the second debate with the women who accused him of assault.

The first debate proved that the general electorate does not like him acting unpresidential.

Yet he keeps doing it and ended the first debate by calling Hillary "a nasty woman" and creating a completely unnecessary controversy by talking about contesting the election results.

All of his advisors would have told him these were bad ideas ahead of time and would cost him votes. And it's not like they asked him to memorize the constitution word-for-word, avoiding these controversies would have been trivial.

I honestly see two main possibilities for his behaviour:

1) Trump really does have a major deficiency, either comically low levels of self-control or he's living in a massive bubble and somehow thinks these are brilliant ideas.

2) Trump has no intention of winning the election and is just trying to preserve his fortune. He lost his TV show and realizes he's destroyed the mainstream appeal of his brand (possibly his biggest asset), so he's now trying to transform his new base into a market for TrumpTV.

> actually seems to garner extra support from their followers by being outlandish.

I wouldn't say extra support. Trump absolutely knows his fanbase, the reality-tv loving, racially insecure (but not financially insecure, trump primary voters average $11K more in yearly income than both clinton and sanders primary voters) authoritarian-leaning types. For them it is not about policy, its about the feels and he gives them the best feels. But they only make up about 40% of the republican party, and everybody else is pretty much grossed out instead of turned on by that shit.

On the topic of Russians, I'm going to assume that this is Putin trying to help his employee Trump win. If Trump can't tweet, he can't keep reminding voters of all the reasons they want to vote against him. And the only way to keep Trump from tweeting is to take out Twitter.

It's mainly affecting the east coast, sure, but also Ohio which Trump needs to win.

Seems like a much more straightforward than using trolling to help him win.

See also: http://fivethirtyeight.com/fea...

Which, as Nate Silver pointed out, is a very poor example considering the lack of polling at the time, and the fact that even with the limited data, there was some indication of Truman pulling ahead.

http://fivethirtyeight.com/fea...

FTFV: A key Clinton operative is on camera saying, “It doesn’t matter what the friggin’ legal and ethics people say, we need to win this motherfucker."

Lock her up and throw away the key. James O'Keefe has video proof. This guy:

http://wonkette.com/424143/jam... ...has video proof. Proof. On video. That people who work for Hillary's campaign are fired up. "Damn the torpedoes" they say (even though there are no torpedoes).

Meanwhile, back in the States:

http://projects.fivethirtyeigh...

Elections don't win themselves!

This one does.

http://projects.fivethirtyeigh...

I think that if Bernie had gotten going 2 months (or better, 6 months) earlier in the run-up to the primary, the actions of the party and likely the results of the primary would have been totally different.

Nah, still would have been the same result, only more Bernie Bros would have squandered more in small donations. Also, don't misinterpret his support either. For instance, my wife & I both voted for Bernie in the primary. We were more interested in seeing Hillary lose than we were in seeing anyone in the Republican Party win. And as it turned out, he won our state and that helped keep his campaign chugging along for a few more months. Neither of us would ever vote for Bernie in the general election. I'm sure a lot of his other "supporters" are in the same boat.

The jobs aren't going away because people here are being replaced by better technology, the jobs are going away here because people are being replaced by workers in other countries who can work for less. These actions are of course being rewarded by the boards of the companies who are doing this.

Simply not true. Improvements in automation have made it more economical to automate than to send jobs overseas.

Also, the world doesn't have an endless supply of people willing to work for pennies. Sooner or later, people in developing countries will demand a higher standard of living. When that happens, labor costs increase dramatically.

I am going to laugh my a$$ off if this ends up becoming another Dewey vs Truman statement

According to 538 (who unlike you and me and most of the press are experts at analyzing polling), you have roughly a 15% chance (as of this writing) of getting that laugh. Not where I'd place a bet, but likely enough that you should probably lay in for some spare a$$es just in case.

http://fivethirtyeight.com/fea...

From that fivethirtyeight article some months ago:

So, how do I wind up with that 2 percent estimate of Trump’s nomination chances? It’s what you get3 if you assume he has a 50 percent chance of surviving each subsequent stage of the gantlet.4 Tonight’s debate could prove to be the beginning of the end for Trump, or he could remain a factor for months to come. But he’s almost certainly doomed, sooner or later.

People have to remember that Nate Silver is using statistics based on assumptions, and those assumptions may or may not be valid.

The particular assumption in the link you quoted, is that his chances will not change in the next two weeks.

Let's see what tomorrow may bring, shall we?

http://projects.fivethirtyeigh...

and Hillary's credibility which is in free fall?

You might as well start calling her "President Clinton", so you can get used to it. Call it aversion therapy.

http://projects.fivethirtyeigh...

And just so you can get the bad news from conservative sites, too:

http://www.realclearpolitics.c...

https://electionbettingodds.co...

If you read the e-mails as proof it was rigged, then you read what you wanted to read and would have believed anyway. Fivethirtyeight demonstrated that the convention was not rigged.

You already knew that the DNC liked team-player, loyal soldier Hillary rather than independent, more-interested-in-protesting-than-governing Sanders. The e-mails conclusively proved only that. They did not show that sanders was cheated.

Bernie won the primary by 51% total in all of the states that have a paper trail

???

Sanders actually did best in the Caucus states. In other words, precisely those states where there was no secret ballot, and its possible to strongarm and intimidate voters, and where the fewest % of the population actually votes, so fraud could be most effective.

In primary states with a huge number of voters (eg: Florida, Texas, Ohio, Georgia, California) he got waxed (25, 22, 10, 8, 7 points respectively)

If I were to suspect fraud based on the numbers, I'd suspect it of the Sanders camp. More neutrally, its pretty clear, no matter how popular the guy was online, real live Democratic voters preferred Hillary.

Are there a lot of people in jail for possessing or using pot? Unquestionably, yes there are. The question though is how did they get arrested? They did something that attracted the attention of law enforcement.

The Federal Bureau of Prisons says that number was just under eighty-four thousand last month, 46.4% of all prisoners. I don't know exactly how they build the graph, but I assume someone arrested for murder while high would be counted as homocide rather than a drug offense. The numbers add up to 100%, so it doesn't appear to be overlapping.

That doesn't count state prisons which—while housing far more prisoners—have a smaller percentage locked up for drug offenses. "According to the Bureau of Justice Statistics, there are 1,358,875 people in state prisons. Of them, 16 percent have a drug crime as their most serious offense."

Plurality voting with single member districts leads to two party systems. It would require seriously amending the Constitution to change that.

Actually it wouldn't take amending the Constitution [which says nothing about requiring plurality or First-Past-the-Post voting], only changing Federal election laws, in order to completely break the plurality system.

First, there are two states (Maine and Nebraska) where the Electoral College vote can be split; increasing which states with this system would then magnify the value of 3rd-party efforts [as each such state greatly increases the odds of a minor candidate earning the one or two electoral vote(s) which might deadlock the EC, forcing the election to be determined by the House instead]. As seen by the fact this system already exists, this change could be implemented without requiring changes to the Constitution or federal election laws, only state laws.

Secondly, change could be instituted within the House of Representatives by revising the laws on how members are elected: Federal law requires the current separate district methodology but we could move towards a state-level proportional representation system. This would grant easier third-party access to Congress and, while not directly contributing to Presidential aspirations, would elevate the visibility of those platforms and policies. Again, this change would not require a Constitutional amendment, but only altering existing Federal election laws.

Because FPTP/plurality voting sustains the current two-party system even in the face of such hatred the electorate shows for Clinton and Trump, saying these changes do not require amending the Constitution does seem to discount the resistance these changes would face... but I believe the unprecedented hatred for those two candidates and the extreme partisanship on display by their supporters together indicate the importance of making them.

Convention bounces are bounces. They don't form a month long trendlines. Also worth noting the only real bounce upwards for trump happen pre-RNC.

Read graphs, not bounces.