Home / General / The Evidence That the Comey Letter Threw the Election Is Overwhelming

The Evidence That the Comey Letter Threw the Election Is Overwhelming



Nate Silver’s essay about the Comey effect is very rich, and there are multiple points worth discussing. But let’s start with the bottom line evaluation, which is definitive:

Clinton’s standing in the polls fell sharply. She’d led Trump by 5.9 percentage points in FiveThirtyEight’s popular vote projection at 12:01 a.m. on Oct. 28. A week later — after polls had time to fully reflect the letter — her lead had declined to 2.9 percentage points. That is to say, there was a shift of about 3 percentage points against Clinton. And it was an especially pernicious shift for Clinton because (at least according to the FiveThirtyEight model) Clinton was underperforming in swing states as compared to the country overall. In the average swing state,3 Clinton’s lead declined from 4.5 percentage points at the start of Oct. 28 to just 1.7 percentage points on Nov. 4. If the polls were off even slightly, Trump could be headed to the White House.

Is it possible this was all just a coincidence — that Clinton’s numbers went into decline for reasons other than Comey’s letter? I think there’s a decent case (which we’ll take up in a moment) that some of the decline in Clinton’s numbers reflected reversion to the mean and was bound to happen anyway.

But it’s not credible to claim that the Comey letter had no effect at all. It was the dominant story of the last 10 days of the campaign. According to the news aggregation site Memeorandum, which algorithmically tracks which stories are gaining the most traction in the mainstream media, the Comey letter was the lead story on six out of seven mornings from Oct. 29 to Nov. 4, pausing only for a half-day stretch when Mother Jones and Slate published stories alleging ties between the Trump campaign and Russia.


We also have a lot of other evidence of shifting preferences among voters in the waning days of the campaign. Exit polls showed that undecided and late-deciding voters broke toward Trump, especially in the Midwest. A panel survey conducted by FiveThirtyEight contributor Dan Hopkins and other researchers also found shifts between mid-October and the end of the campaign — an effect that would amount to a swing of about 4 percentage points against Clinton.5 And we know that previous email-related stories had caused trouble for Clinton in the polls. In July, when Comey said he wouldn’t recommend charges against Clinton but rebuked her handling of classified information, she lost about 2 percentage points in the polls. Periods of intense coverage of her email server had also been associated with polling declines during the Democratic primary.


So while one can debate the magnitude of the effect, there’s a reasonably clear consensus of the evidence that the Comey letter mattered6 — probably by enough to swing the election. This ought not be one of the more controversial facts about the 2016 campaign; the data is pretty straightforward. Why the media covered the story as it did and how to weigh the Comey letter against the other causes for Clinton’s defeat are the more complicated parts of the story.

As Silver goes on to say, the data is consistent with both a little (1-2 point) and big (3-4) Comey effect, and I agree with him that it’s prudent to assume the low end. But that was enough, so it’s not important. And given both the amount and steepness of the decline, the idea that regression to the mean explains all of it is massively implausible.

But, to me, the real kill shot is the data about the extraordinary amount of negative media coverage — without recent precedent in an election campaign — about Clinton that the letter demonstrably generated. As with arguments that the bully pulpit moves public opinion, it’s not just that the theory is inconsistent with the data, but that the theory is weak even on its face. To believe that the Comey letter didn’t change the outcome you have to believe that:

  • The director of the FBI implied that one candidate was a crook
  • Generating a massive wave of negative media coverage, one that wouldn’t just reach a few high-information voters
  • At a time when an unusually large number of undecided voters who didn’t like either candidate were making up their minds
  • But this had essentially no effect
  • Even though there was a steep decline in her poll numbers in the immediate aftermath of the letter

And remember, the election was decided by less than 100,000 votes, so to conclude that the letter didn’t affect the outcome you essentially have to be saying that the coverage of the letter didn’t matter at all. This is extraordinarily implausible. You’re never going to see stronger evidence for a campaign counterfactual than this.

As we’ll get to in another post, this doesn’t mean that the Comey letter was necessarily the most important factor. It doesn’t mean that there wasn’t some choices Clinton could have made that would have overcome it. But it is nearly certain that all things being equal the coverage of the letter put Trump in the White House.

  • Facebook
  • Twitter
  • Google+
  • Linkedin
  • Pinterest
  • Yes, but I don’t think that blaming Comey is really to the point. It was the reaction of the corporate media, not the letter per se, that mattered. I’m even willing to imagine that Comey didn’t really grasp what that would be, but even if he did, it’s the media that is the important institutional failure. It wouldn’t have mattered what the FBI director did otherwise.

    Let us not forget that this is really a “for want of a nail” scenario. Comey wouldn’t even have had this opportunity if Anthony Weiner wasn’t a compulsive tweeter of his junk. In other words it is the media that constitutes the long-term, but possibly addressable, problem. We can’t stop Comeys and Weiners from coming along, but they don’t have to matter so much.

    • FlipYrWhig

      The media was awful about the whole email story all along, never explaining what the problem was or why it mattered, but in a partial defense of that late turn: I don’t think any news media anywhere would have been able to resist covering in detail a story about how the head of the nation’s federal police was reopening an investigation into its leading presidential candidate. IMHO Comey is far more at fault for futzing around with a story whose implications were always going to be explosive.

      • Well sure it was a story. But it wasn’t the only story. They had far less interest in Cheeto Benito’s obvious history of fraud, lies and incompetence and his manifest conflicts of interest. And they had for decades clung to this narrative that Clinton was somehow corrupt, if they could just figure out exactly how and prove it. She wouldn’t even have been vulnerable to this if they hadn’t created a caricature of her to begin with.

        • FlipYrWhig

          This is all true but I would just add that I am pretty confident one of the people who believes in the truth of that caricature is James Comey.

        • WhatToDo

          But as the blog post states, it was the story–all other things being equal–that mattered. Without the letter, Clinton would have won. That’s quite clear.

          • one of the blue

            Media malpractice was the baseline factor in what happened, exacerbated some by the Russian/Wikileaks ratf*cking, and decisively as Scott says by Comey’s late-breaking intervention.

            Comey seems to have been driven more by Clinton Derangement Syndrome (he was an early Whitewater investigator), than by Repug partisanship per se. (Sorry I don’t have time to provide a link.)

            • e.a.foster

              it was always peculiar that they labelled Clinton as a crook, but didn’t supply the evidence. In Trump’s case, the evience was out there, but they failed to put many names to it beyond sexual assaulter and still he was elected. it just demonstrates how much the voters either hated Clinton, thought sexual assault was O.K., or just weren’t ready for a female president.

              my guess comey just decided to make trump president because he thought he could be contained. well good luck with that commey.

      • nemdam

        This isn’t an exoneration of the media’s coverage of the letter, but the bigger malfeasance was the hype around for a year and a half prior. Once you hype something up so much, the coverage it got in the last week and a half was the logical outcome when the letter came out.

        And this is why I have no sympathy for Comey. He wasn’t living in a cave, so he knew his letter would generate a tsunami of media coverage. Unless the worst rumors about the NY FBI field office leaks are true, there is simply no defense or sympathy for his actions.

        • Again, no sympathy for Comey, just that he needed the massive institutional failure of the corporate media to have the effect that he did. Of course I agree with you that said massive institutional failure did not suddenly appear on October 28, 2016.

          • CP

            The massive institutional failure of the media is what put the election within stealing range. Comey is what put it over the top for Trump.

            • Rob in CT


        • Scott Lemieux

          This isn’t an exoneration of the media’s coverage of the letter, but the bigger malfeasance was the hype around for a year and a half prior. Once you hype something up so much, the coverage it got in the last week and a half was the logical outcome when the letter came out.

          And this is why I have no sympathy for Comey. He wasn’t living in a cave, so he knew his letter would generate a tsunami of media coverage. Unless the worst rumors about the NY FBI field office leaks are true, there is simply no defense or sympathy for his actions.

          Agreed entirely, except for the FBI field office qualifier. Guilani’s goons leaking it would not have had the same impact, and if he can’t manage his staff it hangs on him anyway.

    • upstate_cyclist

      Comey was part of the Senate Whitewater commission. I have a hard time imagining he wouldn’t understand how any of this material would in the media salivating for anything to take Clinton down a peg.

      • Well right – but it’s the last part of your statement that is exactly what I’m saying — “the media salivating for anything to take Clinton down a peg.” What I said.

        • upstate_cyclist

          We definitely agree on that.
          I just want Comey’s boy scout image to die in a fire.

          • efgoldman

            I just want Comey’s boy scout image to die in a fire.

            In any other context, he’d have been fired on the spot for violating established policy and procedure, disobeying a direct order, and insubordination.

    • CP

      Yes, but I don’t think that blaming Comey is really to the point. It was the reaction of the corporate media, not the letter per se, that mattered.

      As much as I loathe the MSM, I don’t think I agree. The mainstream media is far less trusted than the FBI as an institution. No matter how they’d reported it, the takeaway from the average Joe would’ve been “FBI is reinvestigating Hillary Clinton,” and any media attempt to downplay it or spin it in the other direction would’ve been taken as “liberal media tries to cover for Clinton.”

    • Yes, but I don’t think that blaming Comey is really to the point. It was the reaction of the corporate media, not the letter per se, that mattered.

      Blaming Comey is to the point. This is why the Justice Department has struct rules regarding how information like this is supposed to be handled during an election. Comey had no issue following the rules where Trump was concerned, so why break the rules for Clinton? Comey knew he was lobbing a bombshell. He fucking new it.

      • Pat

        I’m with you, C.V. All Comey’s pretend indigestion is sprinkles on top of the shit sandwich that he knew he was serving up to the American people.

        • Seriously. He’s a fucking Republican who saw a chance to hit one for the home team. He doesn’t deserve benefit of the doubt. He deserves a fucking Hatch investigation.

    • TroubleMaker13

      Yes, but I don’t think that blaming Comey is really to the point. It was the reaction of the corporate media, not the letter per se, that mattered.

      I see your point, but I think you’re missing the point. The media reaction was entirely predictable. That means Comey was either willfully complicit in the outcome, or thoroughly incompetent and oblivious. The latter requires a degree of obtuseness that just beggars belief.

    • Hogan

      I’m even willing to imagine that Comey didn’t really grasp what that would be

      Really? I’m not. I’m pretty sure Comey reads the NYT on a regular basis and knew how big a hunk of catnip that letter would be for their political team.

      • liberalrob

        Of course he knew what the response would be. He testified under oath that he was made “nauseous” by the decision to send the letter. He knew exactly what would happen; he knew what he was doing.

        The only thing he could possibly in theory not have known was that Chaffetz would immediately leak the letter to the press. But his not knowing that was likely to happen would be almost ludicrously naive for a long-time DC operative like Comey.

    • ForkyMcSpoon

      If I were to exonerate Comey slightly (but not really) I would say that it’s the media hype around the scandal that would even have given the impetus to report it.

      Absent the media hype, I don’t see that Comey would’ve thought there’d be any “scandal” in him not reporting the letter. It’s hard for there to be much criticism of him not revealing a nothingburger about trivialities that no one cares about. It’s only because the media cared so much (of course, all encouraged by the GOP) that it would seem important to report in the first place.

      This is all canceled out, of course, by the fact that Comey should fucking know that it’s not actually that important AND that the reaction of the media would be over the top, thus affecting the election. And as has been noted many times, it makes no sense in light of his decision to stay quiet about the Trump-Russia investigation, particularly given that rogue agents were also telling the media that there was no connection there.

      That is, the media heightened the stakes for either decision he would make, but not in a way that makes his decision justified.

      I would also say that without the media playing along with the GOP in 2015 and hyping up the email scandal… there might not have been an investigation in the first place! But again, this doesn’t really justify Comey’s decisions regarding it. He could’ve simply said there’s no reason to think there’s anything there after a cursory investigation, and maybe he’d take heat from the GOP… but who cares? He’s a Republican and served under Bush, that’s enough to insulate him from real heat.

      I guess in conclusion, I can’t decide who was worse, Comey or the media.

  • mojrim

    I still don’t buy it. This is Silver’s (and by extension all the pollsters) apologia for being so dismally wrong about this election. These people have been desperately searching for something, anything, to make themselves “right” again. Given that (1) the last decade has given us GOP dominance at every level and (2) we’ve developed a habit of switching parties every eight years, Trump’s election should not have come as a surprise to anyone.

    We know that people lie to pollsters, especially in front of their spouses, and we know that people lie to those spouses about their voting. There are no more “undecided” voters, I’d helped hunt them to extinction in the 80’s and 90’s. The Comey letter simply made it okay for those people to say it out loud.

    • Silver wasn’t wrong. The popular vote was very much consistent with the latest polls. The difference, as Scott says, was a few thousand votes that happened to be in exactly the right place — and the pollsters were picking up the movement.

      • Tony Pius

        Silver was the only major poll aggregator/pundit to say, “Hey, Trump isn’t as much of a longshot as people think.” Going into Nov. 8, Sam Wang had HRC as a 99-1 favorite. Most places had her winning 85%-90% of the time.

        Silver had her at 71% on Election Day and wrote “If the popular vote turns out to be a few percentage points closer than polls project it, Clinton will be an Electoral College underdog.”

        And sure enough, here we are in the darkest timeline.

        • randy khan

          Seriously. If you’re going to pick out someone as wrong on this election, Silver isn’t the guy.

          • Silver was, actually, the least wrong of all the major polling aggregates. He also appears to have been the only one who correctly picked out Cheeto Benito’s path to victory, and rarely gets acknowledged as such.

            • mojrim

              Okay, I’ll accept that defense of Silver, but “least wrong” really isn’t much of a claim to fame. And the popular vote is irrelevant, something every pollster and pundit knows.

        • Spider-Dan

          I still don’t see how this is distinguishable from CYA.

          Prior to the Comey letter, 538 said the outcome would be somewhere between narrow Trump victory and Hillary landslide. After the Comey letter, they said the same thing.

          The LA Times poll consistently had Trump winning for many months before the election. Do we give them credit for calling the election correctly?

          My main problem with Silver is that he keeps conflating what pays the bills at his website (picking winners correctly, which is why you hear about 538 getting 50 of 50 races right in 2012 etc.) with statistically valid analysis. In other words, we can’t judge statistical validity by simply looking at the winner, but 538 is happy to write article after article insisting that they were the only outlet to give Trump a reasonable chance to win (and implying that this makes them the most accurate).

          This “There’s too much uncertainty to predict who will win (but I had the best prediction of who won)” game is too cute by half.

    • upstate_cyclist

      Sorry to be annoyingly technical about it, but what sort of problems do you see with the simulation and outcome modeling that make you question the analysis? And I mean specifically about the material within that analysis.

      • Dilan Esper

        Don’t bury the obvious, true point, that polls are not elections, beneath a bunch of jargon.

        Trump voters had obvious reasons to tell pollsters they weren’t supporting him. You can use all the statistics jargon you want, but if the underlying phenomenon isn’t measurable, you aren’t going to measure it.

        • randy khan

          There’s really not much evidence for the shy Trump voter. Trump rarely outperformed the polls in the primaries (in fact underperformed them sometimes) and that the national polls were unusually accurate on the whole. In fact, there were pieces written during the campaign about how Trump voters were proud that they were going to vote for him, and not shy at all about it.

          • Dilan Esper

            The Republican primary electorate is true believers. The general election includes reluctant Trump voters.

            Trump gained like 25 points in Utah to win. That wasn’t Comey.

            • efgoldman

              Trump gained like 25 points in Utah to win.

              And the last Democrat to get Utah’s EVs was….?

            • Aaron Morrow

              The general election includes reluctant Trump voters.

              If the general election included a statistically significant amount of reluctant Trump voters, the final national polling would have been less predictive of the final national vote. There is no evidence that your assertion is true.

              Trump didn’t gain “like 25 points” in Utah to win. Given the conservative make-up of the Utah electorate, what unified break of undecideds to Trump there was probably Comey, per Silver.

            • randy khan

              25 points from what? Seriously, I have no idea what you mean.

              Looking at the last pre-election polls, it appears that Trump gained some votes in Utah compared to the polls, but so did Clinton. Also, Utah was the only state where there was a third party candidate of any significance, and of course third party candidates drop a lot of votes towards the end as people realize that it’s pointless to vote for them. Trump’s share of the two party vote – that is, excluding McMullin – was lower than in two polls and higher than in three polls, and within 2-1/2 points of the average of those polls. That’s pretty good when you consider that most of the McMullin bleed off probably went to Trump.

        • John F

          The polling “error” in the 2016 election was no worse than the average error 1980-2012. Where folks were wildly off was in figuring out the probabilities.

          But people have narratives/memes they want to push:

          1. The Polls said HRC, Trump won, therefore Polling Fail:

          Sure, knock your self out. If you want to say that a poll that had HRC winning the popular vote by 4 points “failed” while the LATimes poll that had Trump winning the popular vote by 4% “won” knock yourself out. If you want to say the LATimes poll was more “accurate”- you are just wrong.

          2. Trump won because a huge wave of folks liked what he was selling, and swept him to victory, all unseen by pollsters who weren’t counting his support:

          This is the narrative that Trump is pushing, why anyone on the left is pushing it baffles me- he won with 46% of the vote, he lost the popular vote, he won the electoral college because he won a bunch of states by about 1%. There was no huge wave, no hidden Trumpian base- there was no major polling fail (there was a major analysis/interpretation fail- HRC had a 4 point general election polling lead- using that to calculate a 90-99% chance of victory (especially when she had a 0-4% lead in many individual states) was just unskewed level wrongness

          3. Pollsters are just making excuses, why should we listen to them?

          Ok, some are, but a lot of these guys are nerdy math geek types, they always do post postmortems to figure out what went wrong and trying to figure out how not to do it again. As opposed to pundits who for the most part are using the outcome to assert they were right all along.

          All those RW trolls and noncompoops (Dilbert cough) whop said Trump was gonna win by a landslide? And who are all now claiming they were right? They were all wrong, they were far wronger than the pollsters they are now mocking- did ANYONE say, hey Trump’s gonna do better than expected, 1-2 points better, he’s still gonna lose the general by 1-2 points but eke out an EC win?

          Silver didn’t say Trump was gonna do that, but he did say it was possible that Trump might do that. Do any of you have anyone better?

    • nemdam

      *Nate wasn’t wrong unless you think putting a candidate at a greater than 50% odds of winning constitutes a rock solid prediction. He gave Trump about a 30% chance which is far from trivial and much higher than most other pollsters.
      *Regarding (1), I hear there was some guy named Obama in the White House, and he passed some sort of important health care law. I may be mistaken, but I don’t think this is possible if the GOP was dominant at every level for the past decade.

      • postmodulator

        Clearly, you didn’t spend your morning arguing with a Bernie-or-Buster who claimed that the ACA wasn’t a real health care law, and that if Obama had been a Troo Progressive he would have passed single payer. Which the House wouldn’t have repealed. Because something mumble mumble.

        By contrast, guess how I spent my morning.

      • sonamib

        It’s like those people who think that the weather forecast is all wrong because when they say it’s gonna rain with 80% probability, 1 time out of 5 it actually doesn’t rain!

        • John F

          Actually when forecasters say it’s gonna rain with 80% probability, something like 2-3 times out of 5 it actually doesn’t rain.

          Forecasters know:

          1. Forecasting is not a precise science.
          2. Viewers have unrealistic expectations.
          3. People get angrier when it rains when it “wasn’t supposed to” than vice versa.

          So if a forecaster thinks there is a 10% chance of rain, he’ll say 25%. If she thinks it’s 50%, then she’ll say 75-80%.

          • randy khan

            Actually, no. Forecasts multiple the chance of rain by the likely area of rain within the forecast area. So, if there’s a 70% percent chance of rain in 70% of the forecast area, it’s a 49% chance of rain. Granted, they usually do round to the nearest ten percent, but that’s about it.

            That’s what the NWS says

            • John F

              Actually yes, (Although I suppose I should have said TV Weathermen” instead of forecasters)

              There’s some pretty startling trends here. For one, The Weather Service is pretty accurate for the most part, and that’s because they consistently try to provide the most accurate forecasts possible. They pride themselves on the fact that if you go to Weather.gov and it says there’s a 60% chance of rain, there really is a 60% chance of rain that day.

              Weather.com often forecasts that there’s a higher probability of raining than there really is.

              This phenomenon is commonly known as a wet bias, where weather forecasters will err toward predicting more rain than there really is. After all, we all take notice when forecasters say there won’t be rain and it ends up raining (= ruined grill out!); but when they predict rain and it ends up not raining, we’ll shrug it off and count ourselves lucky.

              The worst part of this graph is the performance of local TV meteorologists. These guys consistently over-predict rain so much that it’s difficult to place much confidence in their forecasts at all. As Silver notes:

              • randy khan

                In my defense, you said “forecasters,” not “TV personalities telling you what the weather might be.”

              • liberalrob

                I’ve heard some local TV forecasters relate it as “a 20% coverage over our viewing area” rather than a 20% chance of rain. That at least sounds more accurate.

      • djw

        He gave Trump about a 30% chance which is far from trivial and much higher than most other pollsters.

        And not only that he correctly identified Trump’s path to victory. I don’t understand why people keep saying he was “wrong.” It’s just transparently false for anyone who has the faintest understanding of probability.

        • Denverite

          It’s just transparently false for anyone who has the faintest understanding of probability.

          That could cover upwards of maybe even a dozen of the commenters on here.

          • Dilan Esper

            Silver was right, not wrong, about the probabilities ex ante. Probably. :)

            • j_doc

              Can we rewind the universe a few times to see? Then leave it running on one of the higher probability outcomes.

        • sam

          and, AGAIN, every time he raised his head above water to to warn people that Trump had a very real possibility of winning, he got shit on by everyone for: being crazy/trying to drum up pageviews/trying to overcorrect for the michigan primary.

          • Rob in CT

            Yup. Hell, I was giving him some side-eye.

            And he was absolutely right.

            • Pat

              Honestly, if you were going to hack the Diebold machines and change the election results, wouldn’t you test out your attack on a primary, like Michigan’s?

              If you were feeling paranoid, maybe.

          • StillWithHer

            Who was it again that was shitting on his pro-Trump analysis?

        • To be fair, most people don’t have the faintest understanding of probability, which is a serious problem. I think there is actually research suggesting that people evaluate probabilities as something along the lines of 100%, 99%, 50%, 1%, 0% regardless of what they are. I haven’t looked into it in depth, but I don’t think most people possess an intrinsic understanding of what probabilities like 70% actually signify.

          • erick

            That’s why Casinos (unless Trumo is running them) make lots of money (seriously just how incompetent do you have to be to lose money running a Casino?)

          • randy khan

            I’ve mentioned this before, but in my job I sometimes have to put probabilities on possible outcomes, and I try very hard not to use percentages for the very reason you mention.

        • erick

          I’d say winning 3 states by a few thousand votes to win the EC while losing the overall popular vote by 3 million is like catching an inside straight on the river.

          So if anything even a longer shot than the 30% chance Silver gave him.

          People who say Silver was “wrong” are showing their ignorance of statistics, probabilities and polling.

        • John F

          for anyone who has the faintest understanding of probability.

          That is an astonishingly small percentage of the population

      • mojrim

        ACA was, indeed, a stunning achievement, especially considering how slim a majority the dems held and for how short a period it lasted. It’s been a repub congress since 2010, which is why nothing else has gotten done.

        So, a slim, short-lived control at the national level while the states are essentially red. Yeah, I’ll call that running the tables.

    • sk7326

      This shows a severe lack of understanding about probability.

      Silver said Trump had a 30% chance of winning. That is – if you turn the crank on the math machine 10,000 times with the inputs … Trump would have been the winner 3000 times.

      One of the unlikely outcomes came up. Silver made no prediction, just indicated the probabilities.

      You are right about #2 – but that is not a “developed habit” (a recent one). That has been true in all but once case since the 22nd amendment.

      • mojrim

        I’m actually not here to criticize his predictions, only his postoperative justification for the outcome. He, and other quants, are trying to pinpoint some particular, statistically readable, event which caused this disaster. My point is that they are all looking in the wrong direction.

    • Scott Lemieux

      This is Silver’s (and by extension all the pollsters) apologia for being so dismally wrong about this election

      Silver’s ex ante analysis of the election was, in fact, eerily prescient, so pull another one.

      We know that people lie to pollsters

      This is just anti-intellectual horseshit. Polling of national elections is accurate, and it was in 2016.

      • Moondog von Superman

        Polls indicate that people are lying when they say they lie to pollsters.

      • mojrim

        There are no national elections in this country, and every social researcher I’ve worked with has a discount rate for lying respondents.

  • nemdam

    I just want to say there may be some that think you are writing about this story too much. But I want to opine that this cannot be written about enough, and I’m glad you are giving it the attention it deserves. Whatever your thoughts on Clinton, we all need to understand that if we don’t understand or do anything about the egregious outside factors that were decisive in her defeat, then the same story will repeat itself in the future. This work is just as important as any other in succeeding in the future, and I am glad you will not let this story die.

    • upstate_cyclist

      Given that certain corners of the lefty-net screen bloody murder whenever Hillary so much as comments about the outcome of the election and the media has yet to own up to its own failures, I definitely happy to see more data driven commentary on this topic.

    • Alex.S


      It’s clear that —

      1. A Presidential candidate can easily end up being investigated by the FBI. Both Sanders and Trump had investigations surrounding them. I would not be surprised for it to be revealed that Jeb Bush and other candidates were also investigated.

      2. The FBI now realizes that they can interfere with an election, and as long as their target doesn’t win they will be rewarded.

      3. The political media does not have the desire to not follow the FBI’s lead and question what they are doing.


      But hey, at least the political world and media and FBI and DOJ have realized that this might be an issue and are working to immunize themselves form an out of control partisan FBI. Or maybe they haven’t, and in a few years we’ll find out that every single Democratic Senate candidate is under FBI investigation in a really remorseful press conference from James Comey.

    • SatanicPanic

      People are trying to obscure this because their claim that the party needs to retool requires the party to have been wildly wrong. Which it wasn’t. I wish the people who want the Democrats to focus on their pet projects would start arguing for those things on the merits rather than make unproven claims about their electoral possibility, but I’d also like to drive a Tesla and neither seems likely.

    • Joe_JP

      Like “we all know that” comments of some in the media during the Bush Administration/Iraq War lead-up, the fact the people around here as a whole agree with Scott doesn’t settle it either.

      Some things require repetition.

      • Right. This seems particularly necessary because the commentariat may not even be particularly representative of the blog’s readership as a whole. I read the blog for several years before I started commenting, and I’m not sure I would have begun commenting as frequently as I have if I hadn’t wound up leaving multiple other Internet communities (and if one other hadn’t basically destroyed its own interactivity). I don’t have any way of knowing for sure how many people read this blog regularly, but I suspect it’s many more than comment regularly.

    • Agreed. The Democrats need to clean house, build stronger state parties, and move on from the Obama strategy of a centralized, carefully controlled messaging strategy. They need to get younger people and activists more represented in party politics. And they need to have a strong progressive economic platform as well as a strong social justice platform. This all needs to be hashed out over the coming years, and there will be a lot of arguments. Arguments are good. We don’t believe in following the leader.

      But we also can’t ignore that there are forces working to promote reactionary politics in the US, and ignoring those forces leaves us vulnerable to them.

      • Agreed on basically all counts. On the plus side, it does look like a lot of these things are already well under way; there is now more on-the-ground activism at a local level than I’ve ever seen before in my life. But we definitely need a strategy to counteract the reactionary forces in American politics, and while these are commonly discussed here, I don’t see them being as widely discussed as a whole. LGM probably isn’t the largest or most influential blog out there, but it’s definitely a good thing that it’s kept putting this out there. (And it’s also definitely good that Silver is continuing to discuss the Comey letter’s impact.)

      • Origami Isopod

        The Democrats need to clean house, build stronger state parties, and move on from the Obama strategy of a centralized, carefully controlled messaging strategy.

        For anyone interested, there’s a good discussion about that in this guy’s recent diaries on “Red America.” It’s less about economics vs. “identity politics” and more about the value of having political networks in place everywhere — and running Democrats regularly enough in “impossible” districts.

        • eclare

          I read a decent piece over at GOS a few days ago arguing essentially that the reason people think Democrats are out of touch and don’t share their values or whatever has nothing to do with policy (obviously) and everything to do with the fact that there’s no one within their communities running for local office and saying “I’m a Democrat and I want to help you.” Pretty much every local politician is a Republican. This is why Dems need to run in every race, no matter how futile. Just let people know that you’re there and that you care about their issues, no many how small. Helping someone fix a pothole on their street can win the whole party a lot of good will.

          • If you still have a link to that piece, I’d like to read it. I do suspect we may be underestimating how important that can be. Of course, recruiting candidates for local office is often extremely difficult, though I suspect recent events will have made it somewhat easier than it was in the past, because people are now pissed off nationwide.

            • Brien Jackson

              It….depends. I’ve got a surprising number of people interested in running for the House, and enough potentials that we’re going to have a full slate for state legislature. I will probably be able to drum out enough candidates for school board, a non-partisan race.

              I can’t get anyone anywhere near exploring a run for county commissioner, because a) it’s probably still impossible and b) you have to deal with A LOT of bullshit from Reublicans in our county up to and including death threats.

              • That sucks. :( (To be fair, I said “somewhat” rather than “much” for a reason; I do expect lower-level races to be substantially harder to recruit for than congressional/state races.)

          • Origami Isopod

            Agreed. (Actually, I think that might be one of the diaries I’ve linked.)

            Also, this chart found in those discussions could be useful. Democrats as a party really need to counteract that entire “I’m from the government, and I’m here to help you” distrust that was always extant in the body politic but which Saint Ronnie amped up to 11.

          • Domino

            Is there enough people who are willing to spend time on an essentially doomed campaign? It’s not exactly easy to put a contention of running for office – the traveling, the speeches, the fundraisers, the phone calls. It’s a lot to ask of someone.

      • Pat

        I think that electoral security, keeping the process free from foreign interference, is a crucial element of democracy.

        That includes the press, which was manipulated, the FBI, which broke its own rules for unknown reasons, and security in voting.

    • erick

      Exactlly, we have a crisis in our Democracy and people just want to move on and pretend there was nothing to see here. Historians will study this like the election of 1876, Watergate and Iran Contra to pick just 3.

      There is an underrated late Sidney Lumet film, Q & A with Nick Nolte, Timothy Hutton, Armand Asante and just about every “that guy” actor in Hollywood (seriously check out the cast on IMDB). At the end after Hutton’s idealistic young guy figures out all the police corruption only to see the powers that be sweep it all under the rug he asks his mentor something like How can you let this happen? And he replies “you can ignore even bigger than this”. Implying obviously that he has done so many times.

      We just continually ignore stuff in the name of moving on.

    • The Lorax

      Amen. We should never forget what Comey and the media did. This is one reason why (places like the) NYT’s being self-righteous in the face of postmortems like Silver’s makes me so fucking angry.

      • Domino

        Doesn’t help matter when Liz Spayed is clearly out of her depth and not good at her job.

  • N__B

    A) It’s nice to see the data laid out like that.

    B) Putting this up now that PM is banned and can’t respond is the height of cruelty hilarity.

    • sonamib

      Wait, PunidtusMaximus was banned?

      • N__B

        The previous post. He was masturbating figuratively and philosophically (and quite possibly literally) and Scott reached his limit.

        • sonamib

          I’ll grab the popcorn and check it out!

          • jim, some guy in iowa

            I’ll take Mr Bear’s word for it

            thank you, Scott. Much appreciated

    • sibusisodan

      Give it a week. When PunditusMaximus4, Herewego^74, UnderTheShip and whichever random Troll du Jour are all competing to be most disruptively pure, you’ll wish you had never counted your chickens…

      • N__B

        None of them are anywhere near as pure as Dagchester, because none of them know how to self-flagellate the way he does.

        • sibusisodan

          I used to have a theory that Jennie was actually SEK doing fieldwork for a piece on semiotics in distributed communities.

          I miss SEK.

          • Domino

            He said he one day would do numerous shot breakdowns of “BladeRunner” one day. Now that the sequel becomes closer to debuting and they pick up an ad campaign, I keep getting reminded of that.

            Damn it all, he didn’t deserve what happened to him. I miss him too.

    • Dr. Ronnie James, DO

      Does anyone know: was NoMoreAltCenter banned? Or did he just storm off like JoefromLowell, last seen hanging out at Eschaton?

      • Rob in CT

        “StillWithHer” is NMAC.

        • Q.E.Dumbass

          Erik previously banned NMAC in one of the Gillibrand threads, but [s?*]he resurfaced as SWH about two weeks ago (and before that DJW confirmed them as being “I Guess Bust”).

          BTW, who was PM?

          *DJW describes IGB/NMAC/SWH using female pronouns, but if there was a confirmation as to gender I missed it. Given that it used the same beats as Twitter account @FreddyLovesKarl in reference to the Kriss article, I’ve assumed they were one and the same person & thus tend to refer to the account as a “he.”

          • N__B

            My memory is hazy, but I believe PM was ProgressiveLiberal.

            • Rob in CT

              Herewegoagain is Progressive Liberal.

              PM, I’m 99% sure, was someone else.

              • djw


              • tsam

                Didn’t recognize him without the MEAT IS MURDER YOU HYPOCRITES theater.

            • I don’t think so. PL keeps re-registering as herewegoagain and variations, and was posting in one of the same threads as PM a few days ago. PL also demonstrates far, far more CDS than PM ever demonstrated; basically every single post PL has made over the past several months complained about Clinton’s approval ratings, which I don’t think PM ever once brought up. PM may actually have been a completely different person from any of our previous trolls, but I’m not sure. He didn’t actually irritate me as much as many of them at first, but quickly got tiresome so I can’t say the ban really bothers me.

              edit: beaten by Rob. Oh well. I provided some explanatory reasoning so I won’t delete it.

              • eclare

                He didn’t actually irritate me as much as many of them at first, but quickly got tiresome so I can’t say the ban really bothers me.

                The content of his arguments didn’t bug me that much, but dear god, the repetition.

                • That was most of what did it for me as well, I think. He wasn’t nearly as repetitive starting out.

                • Q.E.Dumbass

                  Repetition is also what did it for me wrt King Goat. Speaking of, has he been banned?

                • I haven’t seen him recently, but I also haven’t seen any reference to KG being banned. It could go either way. I seem to remember him disappearing for awhile and then showing up to ruin several threads and then disappearing for awhile again and so on.

                  And yes, his repetitiveness is the main reason I wanted him banned as well. And PL. And… basically everyone I’ve ever wanted banned here, come to think of it. It is, as I said in an earlier thread on KG, that Churchill thing: “A fanatic is a person who won’t change their mind and won’t change the subject” (paraphrased slightly, I think).

                • sibusisodan

                  I have no problem with a certain kind of repetitiveness: we all have our idees fixes. And arguments often require several tries to communicate successfully, still more to change minds

                  But this assumes you are having a conversation, and that the other person is worthy of good, patient and polite argument. It’s never simple repetition because you’re always responding and flexing with that response.

                  By contrast, our trolls show loudly that there is no conversation to be had. We can offer no useful critique or pertinent question. We can only assent.

                  It’s a dominance display.

                • Agreed. And I’ve certainly repeated myself on a number of points as well, particularly on topics like people’s general credulity regarding those they perceive as authority figures. But I’m not doing it in a manner that brooks no opposition from others. I’m actually trying to have conversations with people. The trolls don’t seem to want to have conversations.

                • Q.E.Dumbass

                  Thing is, JG could only really be considered trollish on one subject…but was enough of an argumentative asshat that he’d inevitably destroy any thread with that as a topic. Which seems to be a recurring problem with crankish commenters: JfL rage-quit, the TJ persona suffered a four-month mental breakdown near the end, and ProgressiveLiberal became unhinged enough to cross over from crank into unambiguous troll.

                  What I’m saying here, is that Murc should take special precautions.

                  *”Crank” here denoting argumentativeness, not lack of credibility.

                • Q.E.Dumbass

                  ETA: should read KG

            • wjts

              I don’t think so. ProgressiveLiberal comes back these days as “Herewegoagain[suffix]”. I’m not sure what the rationale is for thinking PM is any other commenter.

              • Q.E.Dumbass

                Brad said it could either be slothrop or TVTray; sharculese thought it was slothrop, and stepped pyramids half-facetiously hypothesized him to be the same guy as BJ’s lib.* Although like you, I haven’t seen any conclusive evidence for PM being a sock puppet for anyone.

                *Though this isn’t plausible– for one thing PM has far less severe a case of CDS.

                • I’ve never seen Slothrop post more than a couple lines or hijack a thread as thoroughly as PM did. He also posted in one of the same threads as PM (which isn’t disproof in and of itself, to be fair). I’ve also never seen TVTray post more than a couple lines.

                • Murc

                  You know what really enrages me about our merry band of trolls?

                  They’ve put me in the position of needing to defend Hillary Clinton, vigorously, loudly, and repeatedly.

                  I may never forgive them.

                • Origami Isopod

                  PM didn’t read to me like any of the other trolls. That’s not conclusive, obviously, not only because I’m not an expert in trollology but because people have been known to adopt very different “voices” for different online personae. For all I know that might be a topic of study at Troll Farm University.

                • jim, some guy in iowa

                  (wrong place wrong time)

                • Q.E.Dumbass

                  @jim: Maybe? Their ideologies are distinctly different in a way TJ & NR’s weren’t, so I’d think more “derp minds think alike.” I also noticed that “anonymous” hasn’t posted here ever since we outed him as realville/Jenbob; for me the connection was clear when realville started ranting about the Southernization of the Midwest.

              • PM did seem to have a very different style than any previous troll. The suggestion that he was TVTray, which I saw someone else make (can’t remember whom or where) is even more laughable; TVTray never posted more than two or three sentences, while PM’s posts often got very long.

                • Origami Isopod

                  PM was sui generis in style. Not that self-contradictory word salad for the purpose of disrupting a thread is unusual. But I haven’t seen any other troll, here or elsewhere, make observations about the importance of intersectionality or the role of patriarchy or white supremacy, then turn around and mock idpol. And then insist their comments don’t contradict each other, you’re just too dumb to get it, and they’ll prove it by typing “LOL” or “fapfapfap” at you.

                • Yeah, I’d never seen that from any of our other Leftist Betters. It’s a rather bizarre form of cognitive dissonance, to say the least.

                • brad

                  Some trolls are capable of a kind of learning, and PM’s style in some ways mirrored TvT’s, is why I consider it possible. TvT, in so far as I noticed and remember, also tried to be friendly and personable and agree when possible, when not doing childishly simplistic trolling that was basically memes with punctuation. I consider it plausible PM was an attempt to flesh out soundbites, as it were.
                  Assuming it was someone having another go and not just a new specimen, which I also agree is by no means a mandatory assumption.

                • Q.E.Dumbass

                  I distinctly recall TVTray calling drones a necessary, acceptable evil, however.

                • jim, some guy in iowa

                  the tray and the maxass had on occasion a smarmy chumminess that reminded me of throttlejockey. but I don’t think much about it

                • Q.E.Dumbass

                  PM is obviously nothing t ThrottleJockey, as IIRC TJ never made any pretenses of being especially left-wing.

            • N__B

              Obviously my memory was hazier than I thought. In my defense, I skim troll comments at best, so all of these assholes merge in my head into one enormous asshole.

              …that sounded better before I typed it out…

              • Dilan Esper

                Is there a program I can purchase that identifies all the trolls?

              • Origami Isopod

                so all of these assholes merge in my head into one enormous asshole.

                Resistance is futile. They will be ASSimilated.

          • djw

            describes IGB/NMAC/SWH using female pronouns, but if there was a confirmation as to gender I missed it.

            I don’t remember why exactly I came to that conclusion, but I assume I had a reason, maybe even a good one.

            • liberalrob

              Is there such a thing as a female Internet troll?

              • There aren’t many, it seems.

                • Pseudonym

                  I guess you must not spend much time on Twitter.

                • I don’t. I also avoid tumblr. I would avoid accepting any troll’s self-identification on either of those platforms at face value, however; I am quite aware of a number of trolls who claim identities that aren’t actually their own as a derailing tactic.

                • Pseudonym

                  There are plenty of female trolls on Twitter with blue check-marks next to their names too.

            • Pseudonym

              If a previously banned troll has returned and is disrupting conversations once more, why are they not banned again?

              • Beats me. Maybe the staff simply aren’t that organised. If it were up to me, every single banned member that returned would be re-banned immediately, and all their posts with second/third/etc. accounts would also be wiped. Then again, I tend to take a pretty hard-line stance on banning overall.

                • Thom

                  Not “that organised” seems a good theory to me. These are people with full-time jobs whose main concern with the blog is writing good posts.

                • Q.E.Dumbass

                  DJW confirmed that this is the case; not helping matters is that IIRC Scott, Rob and Erik are the only people with banhammer privileges.

      • Domino

        I don’t understand Eschaton’s comments – 300 2 sentence posts about personal matters that in no way relate to the topic at hand. It’s one of the most baffling things I’ve come across.

        • jim, some guy in iowa

          it’s an ongoing conversation among people who have a more or less shared political viewpoint but isn’t about politics all that much. It’s about the last place I would expect joe from Lowell to comment on any kind of regular basis

  • FlipYrWhig

    I think one angle a lot of the postmortem analysis may be missing is the number of people who might have turned against Hillary Clinton not on the basis of a _strong_ version of the story (“Now I know she’s a crook, I’m voting for Trump!”) but on the basis of a _weak_ version of the story (“Not this shit again, I can’t stand the idea of hearing about it for 4 more years, I don’t care anymore, I’ve had it”).

    • Justin Runia

      Yeah, I remember thinking immediately after the news hit, over lunch: “JFC, I’m not sure if I can stand four years of this”, followed shortly by “We are so screwed.”

      • rea

        “The American people are sick and tired about hearing about your damn emails,” as somebody once said.

      • Rob in CT

        Right, even we felt this way.

    • sibusisodan

      Ooooh. Well put.

      (do Chaffetz’s (?) confident predictions of Day 1 investigations also fit into that narrative)

    • This is an excellent point. I was an enthusiastic Clinton supporter by the time the emails came out and even I was dreading the constant bullshit that the story heralded at that point.

      • Cheerfull

        I know that visceral emotional reactions are not data, or at least not very much data, but when I saw the little pop-up on my screen from the Wash. Post about the letter my heart immediately sank, and the day after I had to simply stop reading the news because I couldn’t stand the steady drum beat about the letter, and I had this feeling of it all slipping away.

        It was almost literally a gut reaction, and it’s probably why I find it plausible that the Comey letter had a decisive effect.

        But then there was also the amazing phenomenon over the months; however big a lead Clinton would build when Trump showed himself to be creep he is, the polling would always creep back. I think the letter just hastened the reversion to a mean.

        • Right. People often underestimate how much of an effect enthusiasm can have on elections. (At the same time, there were a ton of enthusiastic Clinton supporters out there, and the media rarely, if ever, reported on us. Which was just another of far too many examples of CDS to list here.)

          I suspect some of this was natural regression to the mean, but wouldn’t have been as severe without the letter, and since there were only a limited number of days before the election when the letter came out, it’s difficult to imagine regression to the mean being decisive in the outcome without it.

          I also blame media coverage for some of the regression trends in the first place: every time Clinton got unfiltered access to the media, she rose in polling. They just rarely gave her unfiltered access. The polling trends reflect how dishonest their coverage was. The regression reflects dishonest coverage; the polling after she got unfiltered access doesn’t. So, in other words, the regression was a reflection of the media’s CDS. This is a serious problem and I don’t know what we do to fix that.

          • Origami Isopod

            At the same time, there were a ton of enthusiastic Clinton supporters out there, and the media rarely, if ever, reported on us. Which was just another of far too many examples of CDS to list here.

            Enthusiastic Clinton supporters leaned female, dark-skinned, and maybe very poor too. Obviously the enthusiasm of such individuals was irrational and misguided and in many cases vagina-driven, unlike the enthusiasm of the stalwart White Working Class Male for Trump.

            • Pat

              I think that there was one story about Clinton supporters, compared to hundreds about Trump supporters.

              • Anecdata, of course, but the only piece I ever saw during the cycle that directly provided portraits of Clinton supporters was on Vox, which I wouldn’t count as part of the MSM in the first place. I do think 538 may also have had a piece on why the “enthusiasm gap” was bullshit, but I don’t remember it delving into specifics.

          • randy khan

            Your point about unfiltered access really can’t be overemphasized. Clinton rose meaningfully in the polls after the convention and after each debate. But while Trump’s speeches and fake press events got wall-to-wall coverage – including lots of shots of empty podia while the press was waiting for him to show up – Clinton got almost no unfiltered coverage of her events during the campaign.

            • Yeah. This is a factor that has very seldom been acknowledged in analyses of media coverage, and it may actually be the most meaningful one. The Villagers’ claims that they don’t have any effect on popular perception of the candidates are completely indefensible in light of polling data.

    • Nick never Nick

      This is a great point, I felt this way myself sometimes. It’s not a fair reaction on anyone’s part, but it’s totally human — not wanting to deal with the squelching Swamp Thing that follows the Clintons around.

  • upstate_cyclist

    2000: Al Gore is a serial fabricator
    2016: Hillary Clinton is crook
    2028: ???

    Shall we just assume that DC corporate media gets bored with *successful* Democratic administrations and has this overwhelming desire to kneecap their successor?

    • Justin Runia

      Let’s not discount a not-insignificant chunk of voters who only voted for that one Democrat, and have no memories of having to be a functioning adult while Republicans loot the public trust. Basically, the electoral equivalent of vaccine-deniers who have no memory of how nasty it can get.

    • FlipYrWhig

      The media hates overachievers who look like they practice too hard, like Gore and Hillary Clinton, and maybe, to a degree, H.W. Bush. They like “authentic” people, like W. Bush and McCain and Trump, and maybe, to a degree, Sanders.

      (My map breaks down a bit with Reagan and Obama, though.)

      In retrospect one of the bigger surprises of my politically-aware life is that the media couldn’t drag John McCain across the finish line in 2008, after he worked for years to burnish his reputation as Mr. Straight Talk.

      • N__B

        The media hates overachievers who look like they practice too hard, like Gore and Hillary Clinton, and maybe, to a degree, H.W. Bush. They like “authentic” people, like W. Bush and McCain and Trump, and maybe, to a degree, Sanders.

        So…life as a jocks-versus-nerds straight-to-video movie?

        • Pat

          Obama was quite the jock, but still pretty nerdy.

      • CP

        Did they really dislike H. W? I thought they liked him.

        • wjts

          Outside of the coverage he got during the Gulf War, I don’t remember them being super keen on him (“the wimp factor”, “read my lips”, “a thousand points of light”).

          • Aaron Morrow

            Not in 1998; they loved “a kinder, gentler nation.” To the media Dukakis was, at best, a dork.

            • wjts

              Dukakis definitely got more stick from the media than Bush did, but I don’t think the media was particularly gaga for Bush. (With the caveat that this is based on memory and I was a more politically aware than average 11 year old at the time. Maybe it was just Bloom County and Saturday Night Live.)

              • John F

                Bush 1, by far the least loathsome GOP candidate 1980-2016, got more flack and abuse by the MSM than any other GOP candidate- which is really perverse the more you think about it.

                • FlipYrWhig

                  Unless you interpret him as the Republican version of Hillary Clinton or Al Gore, which I think he is. He got slagged for “read my lips,” got slagged for the supermarket scanner, got slagged by MoDo herself for saying “splash of coffee” IIRC… DIDN’T get slagged for the Iran/contra affair, interestingly enough. He’s probably the Republican with the most grounds for complaint about _unfair_ harsh treatment by the press, somewhere in the neighborhood of Carter if not quite to the level of Gore or Hillary.

            • FlipYrWhig

              That’s not how I remember it. 1988 was a stalemate, I think; Bush was more jock than Dukakis but Bush is pretty dorky himself. And in 1976 Ford was AN ACTUAL ATHLETE but still came across as a dork, making that race dork vs. dork too. And the Dana Carvey impression of HW Bush was that he was whiny and incoherent.

      • sigaba

        Palin ruined him. A thousand MTP appearances couldn’t undo that.

      • upstate_cyclist

        Now that I think about it, ’04, ’08, and ’12 show some of the same patterns. I guess just that the 3rd Term Democratic candidate elections are seared into my memory for all eternity.

      • CP

        In retrospect one of the bigger surprises of my politically-aware life is that the media couldn’t drag John McCain across the finish line in 2008, after he worked for years to burnish his reputation as Mr. Straight Talk.

        2008 wasn’t unwinnable, exactly, but Republicans had so thoroughly shit the bed that even most of their traditional sources of support were finding it impossible to cover for them anymore. (Not that they didn’t try). And, as Sigaba says, he doubled down on it all when he nominated Palin.

        (Naturally, the Republican narrative about this is that the media crucified poor little Sarah Palin and John McCain and that’s why they lost).

      • Origami Isopod

        The media hates overachievers who look like they practice too hard, like Gore and Hillary Clinton, and maybe, to a degree, H.W. Bush. They like “authentic” people, like W. Bush and McCain and Trump, and maybe, to a degree, Sanders.

        They hate smart people who don’t take pains to hide it and don’t feel ashamed of having an intellectual bent. N__B’s “jocks vs. nerds” comment isn’t that far from the truth.

        I don’t actually think Sanders is stupid, but his message is extremely simplistic, which works for The Village.

        • farin

          And they fucking love conventional masculinity, which both Bill Clinton and Obama have by the truckload (albeit with extremely different styles).

        • Pat

          OI, I think that the highly provincial NY Times also hates Southerners.

          (and maybe the women that love them.)

          • Origami Isopod

            I don’t really think it’s regional. I think it’s a combination of classism and, since at least the late ’80s, whom the right-wing noise machine tells them to attack. I haven’t seen them attack McConnell, a Kentuckian, in the same way they attacked Mike Dukakis (a Bostonian), John Kerry (a Northeast blueblood), and Hillary Clinton (originally from Illinois and with no Southern accent that I can discern). Gore is the exception among them.

            As for ordinary citizens, the NYT is fine with working-class or poor people from all regions as objects of pity, as in their recent articles about the “WWC.” They’re also all right with such people when the latter prop up social hierarchies. However, if such people start getting ideas above their station about making our society any fairer, they get ignored, and when they’re too big to ignore, they get smacked down.

            • Pat

              I suppose the best example for your argument would be G.W. Bush, who claimed Texas as his home state, drawled a whole lot, but had a pedigree as long as your arm. The NY Times really bent over backward to support his administration, even as it destroyed the economy.

              So do you think that their bias is more towards those born in the right society, or just to those born to money?

              • Origami Isopod

                I don’t think they distinguish that much between the two, to be honest. It’s a matter of which powerful ass they need to kiss that day.

      • eclare

        Obama’s more like Chris Klein’s character in Election. Smart, but cool and affable. He doesn’t have to try hard – he makes things look easy, and people love that. Basically, he’s the Homecoming King.

        • FlipYrWhig

          I can see that, but the unusual thing is that both Reagan and Obama have reps for (1) affability (2) communication/rhetorical skill. (2) is all about… stagecraft. I guess it’s a performance that doesn’t look like a performance, rather than a performance that looks too much like one? It’s a dire sin in media eyes to be affected, and it’s a special virtue to be artless.

    • Q.E.Dumbass

      Digby and T.R. Ramachandran have documented media figures straight-up admitting in 2015 that they intended to sandbag Clinton & bolster her Republican opponent (and in at least the Digby post, one of the cited figures was Dylan Byers).

      • sigaba

        I’d take that link if you had it, digby is not in my manic reload rotation.

        • upstate_cyclist

          *We* all saw the Equivalency Train roaring down the tracks from many miles away. Being powerless to derail it was infuriating, though maybe the fact that once again NO ONE has been held to account is worse.

          • mongolia

            more infuriatingly, they fail upwards to the post/times/cnn, where clicks and subscriptions have increased, rewarding them for their disastrous coverage.

        • I should really be reading Digby regularly. I don’t know why I forget to check her on at least a weekly basis. I’d also like the link if you can find it.

        • Q.E.Dumbass

          Here. Naturally Byers is now working over at CNN.

      • Aaron Morrow
    • Joe_JP

      2020 at least.

  • epidemiologist

    For those very interested in this topic, the American Association for Public Opinion Research released their evaluation of the 2016 election polling as well. I haven’t had time to read the whole thing but they have a helpful summary of findings that agrees with Silver on several points. (He wasn’t involved in the report as far as I can tell.)

    In particular they concluded that national polls were pretty accurate and while state level polls accurately showed more local competition, they also had methodological issues in key states. And real voter preference did change in the final week before the election, particularly in the Midwest where a lot of voters made up their minds later. They also call out election forecasting as contributing to the perception that the outcome of the election was certain. This would include 538 although famously, it was the least confident of the major forecasters. They suggest trying to improve key state level polls, which critically will require funding first. Something to think about as we all decide where to put our “fuck no” money in the next few years.

    I really can’t agree with mojrim that these post-mortems are just apologia. Many people involved in polling and forecasting have looked at this outcome and other than being in the same broad field, their self-interests do not all align. (The AAPOR remark that “Forecasting models do something different [from polls] – they attempt to predict a future event. As the 2016 election proved, that can be a fraught exercise, and the net benefit to the country is unclear.” is an illustrative example.) Understanding what happened this election has broad implications for social science research as well, and really any discipline that relies on polling and surveys. Actually I found out about the AAOPR report from my institution’s survey research lab.

    Besides, what is the alternative? Throw out all of modern polling and statistics and try to find a new way of gauging public opinion based on something other than asking people? Anyone qualified to figure out what worked and what didn’t will have a stake in the answer. The solution is for many people, with a variety of approaches, to actually analyze what worked and what didn’t, find ways to reduce bias, and keep getting better. It’s science! Any alternative is in “Fucking magnets, how do they work” anti-intellectual territory.

    • sonamib

      They also call out election forecasting as contributing to the perception that the outcome of the election was certain. This would include 538 although famously, it was the least confident of the major forecasters.

      Wait, do they call out the existence of election forecasts? Because that cat’s out of the bag. There will always be some sort of electoral forecast because, understandably, people are anxious about the election, and want to know what’s gonna happen.

      In that light, it seems particularly unfair to hit on 538. The only way to look at 538’s forecast and think the outcome was certain is to :

      1) be innumerate, since a 30% chance is a very real possibility, we’re talking playing Russian roulette with two bullets in the barrel here

      2) not have read any of their analyses, where they explicitly pushed against the narrative that Clinton was a lock-in

      • epidemiologist

        But they didn’t target 538 specifically. Maybe I didn’t make it clear, but both Silver and AAOPR agree that exaggerated certainty about the election may have influenced the outcome– although the effect is difficult if not impossible to measure. Silver focuses on the media’s certainty, while AAOPR draws out the role of forecasting in creating that certainty. They’re perfectly correct to point out that prediction is qualitatively different from measurement.

        My point was more that both analyses are arguably shaped by their authors’ interests and biases, but those interests are certainly not the same. Nate Silver is obviously not going to suggest that election forecasting in itself might be harmful and provide little benefit, only that other analysts did it wrong or the results were interpreted wrong. Pollsters presumably don’t want to be blamed for any failures of analysis and interpretation made by anyone using their data for forecasts. But as experts in this field they do have valuable points of agreement and I think the areas where they differ are probably useful to think about as well.

        For a lay reader who wants to be aware of bias, it’s not helpful to mash these people into one big field that is equally unreliable because they are or feel equally responsible. Especially since I agree with you that the cat is out of the bag on forecasting. AAOPR also know this though– it is one reason they recommended finding the money to improve state level polling.

        • epidemiologist

          Also I probably should have linked to it if I want to pick apart their argument, huh? Here it is.

    • Dilan Esper

      To answer your last question, it’s OK to poll, but people need to understand that there are real limitations to what polls measure and there’s no way to actually measure who people will actually vote for, only how they feel at the moment.

      As a PS, certain people here need to look at Obamacare polling. Did pre-Trump polls accurately measure Obamacare’s popularity? Or what people wanted to say to pollsters?

      • Aaron Morrow

        Are you comparing polling on voting preferences versus voting on policy preferences?

        • Dilan Esper

          Is there a reason to think that voters treat election related polling as a solemn obligation while saying whatever they currently feel to policy questions?

          If anything we actually KNOW that voters treat election polls as non-binding, because the swings far exceed the actual number of swing voters.

          • Aaron Morrow

            we actually KNOW that voters treat election polls as non-binding

            What if we don’t ignore the accuracy of presidential polling?

            the swings far exceed the actual number of swing voters.

            Do you mean “undecided voters,” since there were more swing voters this year than swings?

      • Hogan

        certain people here


        And do you mean polling on “Obamacare”? Or polling on exchanges, subsidies, community rating, ban on rescission, etc.?

        • Dilan Esper

          I wasn’t trying to be childish. And yes, I mean polling on Obamacare.

      • there are real limitations to what polls measure and there’s no way to actually measure who people will actually vote for, only how they feel at the moment.

        This is superficially true. A poll taken on October 28 only measures how people feel about the election at that time, not how they will vote on November 8. So?

        certain people here need to look at Obamacare polling.

        which probably did pretty accurately measure what people felt about “Obamacare”, whatever that meant to them. So?

        Did pre-Trump polls accurately measure Obamacare’s popularity?

        See above.

        Or what people wanted to say to pollsters?

        Why would people want to say they disliked “Obamacare” if they did not, on some level, actually dislike it? The fact that their dislike may have been ill-informed and unreasonable says nothing about the accuracy of polls, or anything relevant to the effect of the Comey letter?

    • I agree with basically all of this. I would also add that any projection that doesn’t bake some uncertainty into the model is flawed, because of the observer effect (the act of observing a phenomenon changes the phenomenon). 2016 is about as strong a demonstration for the observer effect as is possible in politics; the widespread belief that Clinton was inevitable was, ironically, a major contributing factor to her loss, and it seems that that belief sprung from how many electoral projection sites presented her as, in fact, inevitable.

  • Bruce Vail

    The real story is that the Comey letter had a fairly large and measurable impact, probably enough to cost Clinton the election. It wasn’t the only thing that mattered, and it might not have been the most important.

    This is what Nate actually wrote. Not exactly ‘Overwhelming.’

    • Cheerfull

      The post does not say the effect was overwhelming, but that the evidence there was a measurable and significant effect is overwhelming.

      I am curious about the polling that seemed to show in the last couple of days before the election a slight uptick back in Hillary’s favor. Would that again have been a reversion to the mean? If the election had been a few days later, is there a better chance she would have won?

      • Pat

        I think the two key aspects are that (1) the Comey letter was written in complete disregard of the FBI’s rules of permissible behavior and (2) it had a measurable effect on swinging the election to Trump.

        Never forget (1).

      • liberalrob

        Why wait that long? If the big media orgs had DONE THEIR FCKING JOB and come out 2 days later with page 1 stories about “FBI: New Clinton email probe targets Weiner, staffers, not Clinton” or even “New ‘Clinton emails’ not actually new and not actually Clinton’s, FBI clarifies” it might have helped. But that would have required actually following up on the Comey story with FBI sources, which would constitute real work. It’s much easier and less controversial to simply reprint press releases and leaks from anonymous sources. Besides, there had to have been something worth investigating for the FBI Director to have sent such a letter to Congress, right?

    • djw

      He gives high-end and low-ends for the probable effects of the letter. Take away the low end swing, and Clinton wins. Is it your assertion that the choice of the word “probably” in that sentence vitiates the substantive findings of his analysis?

      • Bruce Vail

        No. I read the piece by Nate and he just never said that the evidence was overwhelming.

        You are free to conclude what you wish.

        • Aaron Morrow

          In the article, Silver provided a lot of evidence to build his case that the Comey letter had a fairly large and measurable impact, probably enough to cost Clinton the election. He even called it “a lot” of evidence.

          Since overwhelming means “very great in amount,” Silver presented an overwhelming amount of evidence.

        • djw

          It wasn’t presented as a quotation, it was presented as a characterization. If you’ve got an actual argument the characterization is wrong, OK, but “he didn’t use the word” is silly nitpicking.

          • Bruce Vail

            It’s not nitpicking. Scott’s headline says ‘Overwhelming’ but Nate said its ‘probably enough.’ There is a meaningful difference.

            • If you have a meaningful argument that characterising the evidence Silver presents as “overwhelming” is a misreading, present it. You are not doing that. You are nitpicking his choice of words. Scott didn’t use quote marks. He was characterising Silver’s arguments, not quoting him.

            • Aaron Morrow

              Silver said there was “a lot of other evidence” besides that which he went over in detail. Did you not think to read that sentence, did you not understand that an “overwhelming” amount means “a lot,” or are you confused because you don’t understand the difference between “a fairly large and measurable impact” and the use of “probably” to measure a level of uncertainty less than 100%?

    • Scott Lemieux

      The real story is that the Comey letter had a fairly large and measurable impact,

      Yes. He reached this conclusion after presenting overwhelming evidence. What the hell is your point?

  • AdamPShort

    I think you are being too kind (or, more likely, too cautious) with this last bit:

    “As we’ll get to in another post, this doesn’t mean that the Comey letter was necessarily the most important factor. It doesn’t mean that there wasn’t some choice Clinton could have made that would have overcome it.”

    This sets such a high standard for judging something “the most important factor” that we could really never judge something the most important factor.

    If a fighter is winning a fight easily (and a 6 percentage point lead in the polls with two weeks to go is a BIG lead) and then he gets cut in round 3 and the fight is stopped in round 6 and his opponent is declared the winner, the cut was the reason he lost, and the most important factor.

    The fact that the cut fighter could have knocked his opponent out between getting cut and being ruled unfit to continue is true, but it isn’t very useful information analytically. It certainly doesn’t contradict the judgment that the cut was the most significant factor in the outcome.

    Hillary Clinton was not my idea of an ideal candidate, but in hindsight I think I was wrong about that. She was very close to an ideal candidate. A convergence of factors mostly beyond her control led to her losing to a candidate with what appeared to be very bad candidate dynamics. The two most important factors were Comey’s interference in the election and the electoral college system vastly overweighting the votes of rural white people. Take away either of those two things, Hillary Clinton is President. No other condition was clearly decisive in that way. Take away both of those things and Hillary Clinton is President and enjoying the honeymoon associated with a massive electoral blowout.

    No other single factor is clearly decisive in that way, and since the EC is a fixed element of the system (though its effects in this particular election were unusually large) the only decisive factor specific to this election was Comey’s letter and the associated media coverage.

    • Tracy Lightcap

      Yeah, this is correct.

      The best way to look at the 2016 election is as an (ahem) “interrupted time series”, i.e. a series of observations divided by a particular event. If things are different after the event then before and all other conditions are the same, then the event is the obvious suspect for the change. As Drum pointed out in a recent post, the entire electorate was exposed for the campaign to all the things that were supposed to have a greater influence then Comey’s letter. We all had months to consider Clintons e-mails, her bank speeches, her occasional slips on the podium, the foundation, and all the rest. Result = a 6% lead on October 27. Then the letter was released. And she ended up losing the EC. With a 5% lead (some regression to the mean) on November 7, she wins. Why didn’t she have it? Comey.

      It really is as simple as you say.

    • nemdam

      Yes, yes, and yes.

      There are two reasons I hate all the “Well, Clinton wasn’t perfect and made mistakes” qualifier to any Comey/Russia/Media/other outside factors discussion.

      First, it doesn’t matter. Even if Clinton was genuinely a horrible candidate, that still doesn’t justify the outside factors aligned against her. The merits of the candidate don’t justify outside interference or media malfeasance, and when you say this, it’s implies that somehow Clinton’s treatment was justified. It’s like saying a pitcher lost a game because the umpire was paid off but qualifying any discussion with “Well, he didn’t pitch a perfect game.” It’s completely immaterial to the discussion of whether outside factors influenced the game and at best obfuscates the issues.

      Second, if our standard for candidates is that they have to be perfect, then we won’t win another election because there is no such thing as a perfect candidate. Whatever your thoughts on the merits of Clinton, if you think she should be vilified for being imperfect or the lesson learned form 2016 is to nominate a mythical perfect candidates, then we have learned nothing. Even if Clinton is far from your ideal candidate, the fact is that we must come to terms with how she was treated because if we don’t, we will repeat the same mistakes and continue to lose again in the future. Protecting Clinton’s legacy is vital to the future success of the party.

  • Sebastian_h

    Nate gave Trump a big chance to win, he just thought on average Clinton’s numbers looked a touch better.

    I don’t buy the late breaking voters going to Trump thing as proof of the Comey letter’s effect though. The exact same thing happened in the Brexit polling under what I would characterize as similar sociological circumstances (those who have been getting hammered by globalism making a vote against continuing the course). The effect of the Comey letter there seems more difficult to prove…

    • StellaB

      No, the Brexit polling was much closer.

      • Sebastian_h

        The result was closer? Yes. But it had the same characteristic where the undecideds very largely broke for Brexit rather than breaking evenly or nearly evenly for Remain and Brexit.

        Because of that, I question whether or not there is something going on with that on a wider scale than merely Comey’s letter (which is the explanation used for why they largely broke for Trump at the end).

        Comey’s letter could not have influenced the Brexit vote (because the Brexit vote predated the letter) but it exhibited the exact same surprising characteristics seen in the Trump vote.

        If only one or the other had exhibited the surprising characteristic of having the undecideds largely break for the anti-globalism side (or if you want to say anti-immigrant side?), than there wouldn’t be a question. But they both exhibited that characteristic, which at least tentatively suggests that something which allegedly influenced only one of them might not be as strong an explanation as is supposed.

  • xq

    There’s a lot to like about Silver’s article. I think it’s superior to pretty much all previous takes on the subject. And I’m in agreement on the media criticism angle. I don’t think skepticism that Comey was decisive in any way excuses the behavior of the media.

    But I admit I’m still not convinced that the degree of confidence that Comey was decisive is justified. Silver presents a 1 point Comey effect as within the realm of plausibility. In that scenario, of the 4 points separating Clinton’s lead on Oct. 27 from the final result, 1 is due to Comey and 3 is due to other factors. Comey is decisive. But what if the Comey effect was more like 0.5 points and 3.5 points was from other factors? Then Trump still wins PA and the election.

    What makes the 3/1 division so much more probable than a 3.5/0.5 division? Silver treats 1 as the lower bound to the Comey effect, but doesn’t appear to do any calculation to justify it. It’s just his intuition that heavy coverage of a story harmful to Clinton during the last week must have had an effect at least that large. But is his intuition really so fine-tuned to be able to strongly distinguish between 3/1 and 3.5/0.5? Mine isn’t.

    So, I dunno. There was a large drop late in the race. We seem to agree that, more likely than not, a majority of that drop was not Comey-related. So how to partition it? We lack the data to make an informed estimate, so it depends a lot on your priors on how much you expect an event like the letter to change the race. If you buy the argument that news events tend to have tiny effects, exaggerated in the polls by nonresponse bias, I see nothing in Silver’s argument that should lead you away from that view.

    • Aaron Morrow

      I thought “differential turnout in voting” was a bigger thing this year; I guess it depends upon what baseline to which you compare 2016.

    • Dilan Esper


      Silver’s assumption that regression to the meam can never cause a big polling drop is pure speculation. It could all be regression.

      • Jackov

        From Dan Hopkins’ writing on the ISCP panel surveys referenced by Silver:

        Still, we shouldn’t discount the possibility that voters might have gravitated to Trump anyhow. Research has long suggested that over the course of a campaign, partisans come home to their party’s candidate. Between mid-October and our post-election wave, Trump picked up almost 4 percentage points from people who had backed Romney four years before, suggesting that Republican identifiers were doing just that.

        Even the strong version of the Comey swing, might have just been Republicans returning to Trump. (Though one could argue media saturation of the letter prevented Clinton’s closing arguement from being heard or diminished the number of third-party backers who would have broke for Clinton or depressed lean-Dem turnout and on and on)

        • xq

          One of the strange aspects of this debate is that so many people on the “Comey was definitely decisive” side took and continue to take that article as strong evidence for their position even though Hopkins himself says that his data is consistent with a normal “coming home” process.

          That the late swing appears to be composed largely of Romney voters is very strong evidence for “coming home”–and while it’s possible that Comey caused many more Republicans to do so than would have otherwise, it’s hardly obvious from the data. Partisan Republicans ultimately vote for Republican presidential candidate despite misgivings is not something that requires special explanation; that’s just how elections work in the hyperpartisan US 21st century.

    • Srsly Dad Y

      I’m agnostic, but the American Association for Public Opinion Research post mortem, mentioned by epidemiologist above, also makes the point that the Comey letter came in what now looks like the middle — not the beginning — of a fairly steady erosion in Clinton support that lasted until Election Day.

      To rehash someone’s boxing analogy above, if a fighter starts taking hit after hit to the head in one round, and in the next round takes another and gets cut, and three rounds later takes more hits to the head and the fight is stopped because the cut is too bad, it’s hard to say the cut really made the difference. It was all the hits.

      • xq

        Looks like an interesting report. I’ll need to read in detail later. But the conclusion accords with my view:

        Based on all of the data examined here, we would conclude there is at best mixed evidence to suggest that the FBI announcement tipped the scales of the race. Pairing this analysis with the preceding one on NEP data for late deciders, it remains unclear exactly why late-deciding voters broke for Trump in the Upper Midwest. Anecdotal reporting offered a number of other suggestions (e.g., Republicans skeptical of Trump finally “coming home,” Clinton’s campaign – believing the Upper Midwest was locked up – allocating time and money elsewhere, Democrats lukewarm on Clinton deciding to stay home), but ultimately the data available do not offer a definitive answer to this question.

        • Murc

          Clinton’s campaign – believing the Upper Midwest was locked up – allocating time and money elsewhere

          This isn’t. Fucking. True.

          I don’t even mean as a piece of analysis, I mean as a matter of fact. Clinton dumped a shit-ton of time and money into PA, and she hurried to dump resources into WI and MI once it became clear those resources were needed!

          It’s true she didn’t have resources in WI and MI through the summer and into the fall. And that’s because there was no evidence resources were needed there.

          Why the fuck should I take seriously something that gets such basic facts wrong?

          • xq

            Your third paragraph seems to be saying the same thing as the blockquote? PA is not Upper Midwest

            • xq

              That said, I agree that it makes little sense as an explanation. They’re just referencing “anecdotal reporting”, though, not endorsing it.

            • Even if you don’t include PA as part of the Upper Midwest, she still spent a significant amount of time in OH and IA, and generally at least parts of those states are counted as part of the Upper Midwest.

              Every single analysis that faults Clinton for not spending time in MI and WI fails to account for the fact that she did spend time in other demographically similar states (also including PA) and still lost them, and that MI and WI alone wouldn’t have been enough to win her the election. If she diverts resources from OH and IA to MI and WI, maybe she wins those two, but she still doesn’t win the election.

              And, again, without the Comey letter, she probably doesn’t need to spend time in MI and WI. Even if the letter only shifted about 30,000 votes in those states, that’s still decisive.

              • nemdam

                All the “Clinton didn’t campaign in Wisconsin!” hot takes in addition to ignoring PA, also ignore OH. Clinton contested OH as hard as any state and the polls were off in OH as much as WI and PA.

          • Sebastian_h

            Meh. The problem in the Midwest wasn’t that she didn’t throw money at it in the last week. The problem was that Democrats in general, and Clinton especially, hadn’t been particularly concerned about the deterioration of the area for decades. Clinton was easily the wrong person to deal with that if she had ever thought to–and by all appearances she never thought to.

            Making it about the last minute allocation of campaign resources is missing the forest for the trees. I’m sure that campaign managers told reporters crap like that, but it doesn’t make it true. The Democratic problems in that area of the country were sown long ago–Obama kept things afloat (and being from Illinois probably didn’t hurt) but without change, this was coming. Which is why focusing on the Comey thing is so annoying. Just like Clinton won’t be the next nominee, Comey won’t be able to release reports on the emails of the next nominee.

            But if we don’t address the problems in the Rust Belt, Democrats won’t get the area back.

            • Well, again, Clinton did spend time in the Midwest. Lots of time. She just didn’t spend time in MI and WI. She spent plenty of time in OH, IA, and PA, though, and lost all of them.

              The deterioration of our coalition in that region is a serious problem and needs to be addressed, but it’s not as though Clinton bears unique responsibility for this here. The fact that several of these states have had Republican governors for awhile seems rather noteworthy. A lot of people look at this and blame the national party, but in truth, this is not the national party’s responsibility, nor has it ever been. I don’t have time to delve into the specifics right now but it’s always been the state and local parties’ responsibility to a rather large extent. I don’t fully know what the solution is, but it may be correcting itself already with the increase in activism due to the shitgibbon, the healthcare bill atrocity and other factors.

              • FlipYrWhig

                Clinton also won most of those states in the Democratic primaries in 2008. Which I think needs to be taken into account before concluding that Bernie Sanders’s apparent strength in those states is reliable proof that he’s a genius at communicating with them going forward. (Not that you were doing this, CL, but many people in these threads appear to be inclined to do it.)

                • That’s a good point. If people are going to criticise Clinton’s appeal to those regions, they at least need to explain why her message in 2008 resonated but not her message in 2016. Most of these people haven’t addressed any of this.

                • FlipYrWhig

                  My feeling is that the white working class in the post-industrial Midwest probably prefers Hillary Clinton to Obama and Sanders to Hillary Clinton BUT wouldn’t necessarily prefer Sanders to, say, Bill Clinton or Brian Schweitzer or Joe Manchin. Hence I think looking for mini-Bernies is barking up a very wrong tree. It’s mistaking the symptom for the cause, or something of the sort…

        • randy khan

          I don’t have all the data handy, but there are a couple of obvious, let’s say, caveats to the conclusion that there’s no more than “mixed evidence” on the Comey intervention. One of them is that a good part of it, by their own admission, depends on anecdotal claims about other potential influences, which is to say you won’t find any real data to support those claims. Another is that in any election polling results will vary a bit day to day, so you have to be at least a little careful about proclaiming a two or three day event to be a trend.

          • Dilan Esper

            We don’t have any data on Comey as a cause either. We only have polls, which don’t measure actual voting intention.

            It’s turtles all the way down.

            • No, that’s wrong. There are also early voting returns. Comey’s letter appears to have had a measurable impact on those as well. I can’t remember if Silver specifically addressed those in this specific piece, but I’m certain they have been analysed, and I’m fairly certain Scott mentioned those analyses in one of his previous pieces.

              • xq

                No, notably, Silver did not address than in his piece. Probably because he knows it’s really weak evidence. To see why, you just need to go a few articles back in this series. The intro to that article:

                Data on early voting, for instance, usually doesn’t provide much predictive insight. Historically, the relationship between early voting in a state and the final voting totals there has been weak, and attempts to make inferences from early voting data have made fools of otherwise smart people.

                • Well, there’s a difference between comparing early voting to the final count and comparing early voting from one day to early voting from a day, or a few days, later. My understanding is that the analyses compared early voting on a day-to-day basis (as opposed to the final vote total) and found that Comey’s letter caused a measurable shift in the returns. Of course, not all states have early voting, and not all of the ones with early voting had it during a period that would’ve been shifted by the letter, but my understanding is that the analysis compared early voting within the states that had it over the relevant period and found a notable, measurable shift at the time of the letter. That said, I don’t have time to look into it further right now.

            • randy khan

              Good Lord – of course *we* *have* *data.* The polls are data. And you can keep saying that the polls don’t measure voting intention, but that in fact is what they do. (I see your comment above about how the swings in polls don’t match the number of swing voters, but (a) the group of “swing voters” actually changes during an election campaign; and (b) in this case in particular, we’re talking about shifts in sentiment, and aggregated polls over time do measure that pretty well.)

            • sibusisodan

              We only have polls, which don’t measure actual voting intention.

              For what may be the millionth time:

              Polls are a measure of voting intention. That’s what the words ‘Who are you voting for?’ mean.

              They measure this imperfectly. For a variety of reasons, chiefly sampling difficulties. Quite a lot of that uncertainty is quantifiable using broadly agreed assumptions.

              This is very different from saying they do not measure it at all.

              • Dilan Esper

                No, they measure voters’ current mood, which is not a vote.

                Any time you assume it measures voting intention you are making an unjustified jump.

                • GeorgeBurnsWasRight

                  I’d argue that it’s up to you to show that voter’s mood doesn’t strongly correspond to the vote they’re going to cast.

                  Are you claiming that voters decide who to vote for on primarily intellectual conclusions? The history of advertising says otherwise.

                • sibusisodan

                  No, they measure voters’ current mood, which is not a vote

                  They literally measure voting intention. That’s just what words mean.

                  It would be unjustified to say this gives us a perfect picture of a future election. No one is saying this.

      • AdamPShort

        It’s kind of funny that you make that point because that is a fairly accurate description of what happened in the Lewis/Klitschko fight which is te fight I was thinking of in making the analogy.

        This requires you to disagree with the Silver analysis, and I think if you disagree with that analysis and find that the Comey letter did not affect the polling by at least a point then you are correct. The analogy is murky in this case because of course a fight stopped on cuts is officially acknowledged as having been stopped for that reason.

        The point I’m making is that if you accept that the Comey letter swung the polls by at least a point, then you can declare without much fear of being wrong that the Comey letter was the decisive factor.

        • Srsly Dad Y

          Right, which is like saying the losing boxer would have held on to win despite all those hits to the head if he just hadn’t gotten cut. You know … maybe?

  • Morse Code for J

    What angers me most is that it was so completely fucking unnecessary of Comey to have sent the letter.

    Precedent and custom argued against sending the letter. Common sense argued against it, insofar as nobody had even had a chance to study the “evidence” found on Weiner’s laptop, and thus making any claims about it to someone bound to leak those claims was irresponsible. I bet you could look back over Comey’s career and find no instances where he did this before. Or instances where he had an opportunity to talk prematurely about unexamined evidence from an investigation to interested parties in government, but did not because doing so would have been unethical.

    And of course this is what always comes of appointing Republicans in a Democratic administration, who are believed to be above politics. Never again.

    • Never again indeed. I really want this to be a question at the primary debates in 2020. Candidates should pledge to only appoint Democrats.

      • upstate_cyclist

        Good luck getting Cuomo to do that. ;-)

        • Cuomo would pick a GOP VP and instruct him to break ties in favor of the Republicans.

          • upstate_cyclist

            As NY Senate goes, so goes the Nation.

      • Brien Jackson

        This is kind of interesting to flesh out. I basically agreed with Scott’s critique that appointing Hagel SecDef was bad politics, but I don’t remember any complaints that he wasn’t in line with the administration and didn’t do a basically good job. LaHood seems to have been good at Treasury, but he’s the exception that proves the rule: blue state Republican with an idiosyncratic interest in a “minor” cabinet department. Comey, by contrast, was always a hack with massive red flags going up.

        So basically you can appoint either heterodox Republicans who are already mostly out of line with the party like Hagel, or give a “less important” spot to someone like LaHood…but these Reublicans don’t really exist anymore anyway. But Obama just wasn’t clear eyed enough about what a hack Comey was, and that was downright suicidal level delusion.

        • Morse Code for J

          Or you just don’t appoint any whatsoever, since they’re not about to reciprocate and any gestures towards restoring comity between parties are wasted.

    • Tracy Lightcap

      I agree with all of this, provided that Comey could have sure that his decision wouldn’t have been leaked from inside the FBI. He might have been able to prevent that by calling for a determination that the new e-mails hadn’t already been examined. Problem = the very people who found the e-mails – the NYC office – had been after Clinton from the first and he couldn’t depend on them to keep their mouths shut. So he sent the letter.

      This is no excuse for abandoning DoJ policy like he did. I’d fire him for it, if were president, but I can see why he did it.

      And, yes, no more of this bipartisan stuff in future appointments. The president’s party or a true neutral (they exist) from now on.

      • Taylor

        He put his own reputation ahead of the country.

        What a guy.

    • Joe_JP

      And of course this is what always comes of appointing Republicans in a Democratic administration, who are believed to be above politics. Never again.

      A consistent practice would work both ways.

      So, no Democratic appointee for a Republican either. Or is that different?

      Just having Democrats appointing a Republican is a rather limited “above politics” principle. Comey in particular was chosen because he particularly allegedly had cred. Some have shown this is b.s. but “Republican” isn’t the only problem or the only reason he was appointed.

      There are actually Republicans out there who might not have done what he did. Anyway, I think there is some reason — somehow — to have a norm where independence is an important qualification for FBI directors. Shouldn’t just be tossed to a crony above and beyond some Cabinet position. The length of the term alone shows this.

      Just appointing a person from another party would be a suspect way of doing that, sure.

      • Morse Code for J

        If there are Republicans out there who wouldn’t have thrown this election out of self-interest and partisan leanings, I don’t know of any way to distinguish them ex ante from the Republican hacks who happen to be unusually smooth about being hacks.

        And going forward, I feel confident that the universe of Democratic state or federal prosecutors can and will furnish a suitable candidate every time there is a vacancy at FBI. This person will be prepared to step on any cabals of FBI agents who decide that they are going to smear Democratic candidates for no good reason. I see no reason to settle for less in a director.

        • Joe_JP

          I think there are ways to determine someone who just happen might be a registered Republican is not merely a hack.

          It in fact in the minds of some people gives Comey too much credit to say the problem he’s a Republican. Comey had issues with Clinton (the assumed candidate in ’16 when he was appointed) specifically.

          But, sympathetic with the general distrust. Still, my overall concern is general. I don’t want someone who will treat Republican candidates badly too. There are examples of that happening over the years; that is, political related prosecutions that appear biased.

          It’s a thing & trying to find a means for neutrality here is something I’m sympathetic with as well. It might not be possible but different from a “Republicans for big boy jobs” rule.

      • What’s the last major Democratic appointment in a Republican administration? Wikipedia has a list of cross-party nominations, but said list includes Jared Kushner and Michael Flynn (!!). The highest-level Dem appointees in the W admin were Norm Mineta (Secretary of Transportation) and Surgeon General (Richard Carmona).

        In comparison, Obama appointed two Republican Secretaries of Defense, a Republican Fed Chair, a Republican CIA director, and a Republican FBI director.

        I’m comfortable making that trade.

        • Morse Code for J

          Obama also appointed Ray LaHood, an Illinois Republican from the House, to serve as DOT his first term.

        • Joe_JP

          The list includes Secretary of the Army during the Bush43 Administration. That’s a fairly serious job.

          There is a general concern that Democrats nominate Republicans for certain “big boy” jobs like Defense. I’m concerned about that too but it’s understandable those without foreign policy or military cred like Clinton and Obama would do that especially if the alternative is a lot of trouble in confirmations etc.

          Yes, they did it in part because of a double standard in place that practice shows to be wrong, particularly on policy grounds.

  • David Hunt

    At this point, I think the best (though still shitty) argument that Comey’s letter didn’t throw the election is to say that the media would have just seized on some other anti-Clinton story and talked that up until the election if the email thing hadn’t come out.

    As I said above, this is exteremly weak. Anyone who’s seen a word cloud of Clinton campaign coverage will immediately have their eye drawn to the huge “email” in the middle of it. I can’t think of any other story that would have triggered that big a feeding frenzy short of Dead Girl or Live Boy.

    Addendum: while writing this, it occurred to me that the Gulliani’s New York FBI sources might have leaked the story if Comey had kept his mouth shut. I still don’t think it would have generated as much Buzz. So, Comey is first up against the metaphorical wall.

    • upstate_cyclist

      But we all have to wait it as the whole Russian FARA/RICO/espionage investigation grinds on. Maybe someone should send him an albatross.

      • FlipYrWhig

        A hypothetical truly fair-minded Comey could have said something like “If I hear one thing leaked about the FBI and Hillary Clinton, I immediately look the other way on leaks about the FBI and Donald Trump, which will cancel it out. Mutual Assured Destruction, dumbasses. Let our agency do its motherfucking job in accordance with procedures, and it’s my ass on the line anyway, and I can take it.”

        The fact that _he didn’t even CONSIDER_ doing this is what leads me to believe he was, gee whiz, probably not fair-minded in the first place.

  • Taylor

    I just want to say there may be some that think you are writing about this story too much. But I want to opine that this cannot be written about enough, and I’m glad you are giving it the attention it deserves.


    There’s nothing our fucking professional media would like more at this point than to airbrush their role in the 2016 election out of history. Downplaying Comey and blaming Clinton are all part of that self-serving weasel strategy.

    Never. Forget.

    • Pat

      It’s a lot like their role in promoting the war in Iraq, a country that had not attacked us.

  • veleda_k

    Rebuttal: I hate Hillary Clinton. She is a meanie and gives me sad feelings.

    • FMguru

      Did you know she once gave a speech at a bank?

  • sibusisodan

    Apparently Macron’s campaign has been hacked and the data posted just before purdah begins.

    I can only imagine Le Pen is really nervous about it happening to her…

    • petesh

      I trust you intended to use the code tag for the second sentence.

    • Pat

      Well, we know who Putin favors…

  • joejoejoe

    To quote Stork in Animal House, ‘Well, what the hell we s’posed to do…?’

    What should follow if you assume that the Comey letter tipped the election 3% towards Trump? Fire Comey? An election do-over? Electoral college reform? Find alternatives to corporate media? I don’t understand what is supposed to come from all of this analysis. At 3%, Comey had the same weight on race as Gary Johnson. At 1%, Jill Stein. 3% is roughly one-fifteenth of either Trump or Clinton’s vote total. A heavy straw that breaks the camel’s back is still only a fraction of the load. How many times can you cite the data behind the Comey effect and then go ‘Ah-ha!’?

    • jamesepowell

      Though both were ridiculous and stupid at times, Johnson & Stein were candidates on the ballot. Whatever impact they had on the race was legitimate in our political system. Comey, a supposedly neutral law enforcement officer, has no legitimate role in determining elections. Particularly when there was no evidence of any crime.

      Can you not see that? Or are you trolling?

      • joejoejoe

        So is this just a lot of discussion about what is and is not legitimate? I guess it follows that the entire Trump administration is now fruit of a poison tree because it isn’t legitimate? Great! I’m sure they’ll just step down.

        Is James Comey better or worse than J.Edgar Hoover? The amount of U.S. government power that is illegitimate under the terms of this blog post and debate is a heckuva lot more than any margin Nate Silver can identify with polls and statistics.

        It’s 6 months on from the election. Are we going to be talking about Nate Silver and the Comey effect in another Friedman unit or can we move on to the situation on the ground now?

  • jamesepowell

    I agree in every respect with this post and with Silver’s analysis. But no amount of analysis, explanation, or argument is going to change the now firmly fixed narrative that explains the election. It is now established beyond peradventure that Hillary lost because:

    a) she, like all Democrats, is a coastal elite who is out of touch with Real Americans,

    b) she ran a bad campaign, and

    c) she is a corrupt, lying pawn of Wall Street.

    These are the only permitted explanations in the press/media. So, Scott Lemieux, are you going to move on or are you going to be the new Bob Somerby, correct but ignored because acknowledging that he is correct is just too uncomfortable.

    Also, I’m starting to feel sorry for Matthew Broderick. It was just a character he played in a movie!

It is main inner container footer text