Here's an excellent analysis from five-thirty-eight: http://fivethirtyeight.com/features/what-a-difference-2-perc...

I'm starting to wonder if the overconfidence in a Clinton victory boils down to assuming that errors are independent when they really aren't. Specifically, did they assume swing states might go one way or the other individually, rather than that they might move as a block? There was a lot of talk about how there were a bunch of states that favored Clinton and she only needed to win one of them. Pulling numbers out of my rear end, if she had a 50% chance of winning each of 8 states, and she only needed one, you might assume her odds of winning the election are 99.6% (255 out of 256). But if they all swing together, then her odds of winning would only be 50%.

Personally, I think polling is worthwhile. I think it's possible they made honest (and sometimes incompetent, but not malicious) mistakes, and acknowledge that attempting to predict the outcome of a future event is inherently difficult. I think it is possible to do honest work while still voting.

Is this a fair representation of what you're saying? I don't feel I've captured it quite right, as I can't figure out how to tie in the entertainment aspect of polls. Also, I'm not happy with my choice of the word "fear" as the driving aspect of motivation.

Acknowledging that I may not have captured your argument correctly, I'm going to move forward, fully willing to abandon it if it ends up relying on parts that I've misunderstood.

For one, I'm not sure where to draw the line of who is involved in a campaign, and therefore who should know about poll results. There are people who are paid to work on the campaign, so it seems clear they are part of the campaign. And there are people who spend a lot of time volunteering, so I think it's fair to consider them as part of the campaign. There are people who volunteer less, for example donating a few evenings working for phone banks, or perhaps working in IT or data management, or working on the polls themselves. Then there are people who aren't directly working for the campaign, but are passionate about their candidate (or perhaps passionately opposed to another), and either work on their own or self-organize. Maybe the line is that only people who are involved in decisions that require polling data should know the results? What justifies "needs to know"?

Another question I have is about GOTV in general. The numbers I've seen show that over 40% of eligible voters didn't vote in the 2016 election. I think we agree that more people should vote. Some GOTV methods I've seen are general, non-partisan "Please vote" campaigns. Partisan ads often include admonishments to vote. There are also programs to help people get to the polls (for example, rides). Some of these are non-partisan, some partisan. Are these legitimate GOTV efforts? If the electorate is to be accountable for turnout, what other methods can they use? Should partisan ride-providing efforts have access to poll data to best allocate their resources? How about non-partisan groups, as polls may show which areas are expected to have lower turnout?

As I said, I've responded to what I interpreted as your position before confirming that my interpretation is correct. I've done this in the interest of expediency, not to build a straw man. So please let me know how my interpretation can be improved if it's not correct.

Thanks!

So...

Internal polls are not generally released. Nor are they shared to PACs. This is one benefit of public polling. It helps coordinate between the campaign and the PAC side.

Re entertainment, I think watching polls is as entertaining as watching any sport for a lot of people. (I think that is pretty obvious actually - just my observation, though I worked with people in the political consulting industry, so perhaps my circle is wonkier than most).

Now, the big crux and where we will disagree... People should vote because they are civicly minded and want to participate in the process. That requires education and sophistication, which is something both side lack en masse. The only thing worse than people not voting, IMO, are dumb people voting for the wrong reasons. The latter is mob rule and is why we're where we're at. Stupid is a bipartisan quality.

As for access to voting, I think we could benefit from electronic or more mail-in. I don't want to suppress the right to vote; I just don't want to see sides manipulate the idiots of the electorate into voting. If you ever see how a GOTV strategy is put together, they look for people who are "persuadable", which is just another way of saying dumb. (Those analysts will admit it too if you put enough drink in their belly)

I completely agree that people should vote because they're civically minded and want to participate, and ideally make their decisions based on a well-educated understanding of the issues (and all they entail). What does this mean practically? What level of education is adequate? What are the wrong reasons? Single-issue voting? Being inspired by a leader? We're human, not perfectly (nearly?) rational, imperfect in general. What level of sophistication is necessary? I'm asking these questions of myself as much as of you.

I have similar questions when it comes to persuasion. One person's education is another's persuasion is yet another's malicious manipulation. How do we convince people to be civically minded without persuading them that it's the right thing to do? What's the distinction between appealing to people who are likely to be amenable to your position and looking for those that are persuadable? How do you educate people who are, as you put it, "dumb"?

Like I said earlier, these are questions I pose to myself as well. I'm asking here because I want to get other's opinions. Thanks again for taking the time to consider them. Discourse can be an important part of reaching understanding.

Those, however, allow for vote buying and coercion, so I utterly oppose them.

What does this suggest about all the other stories the mainstream media (CBS exec: "How could Nixon have won? I don't know a single person who voted for him!") tells every day, in exactly the same way, about all sorts of issues that can't be so easily measured?

And what does it suggest about the academic institutions that enforce conformity to exactly the same set of opinions and narratives and not only filters information but fires and otherwise silences dissenters? And anyone who suggests that a system so utterly lacking in diversity of thought that there is no one left on campus who will tell them if they're wrong can't be trusted is called "anti-intellectual" and "anti-science".

It doesn't tell me that they are wrong about everything, just that they are likely to be just as wrong about some other issues they are adamant about as they were about this one, but that promoting their dogma, not working to find the truth, is what they really care about, which doesn't make them wrong about any particular thing, just not reliable.

The vote was almost dead even. If you though Trump was going to win big then you were just as wrong as the places that predicted a big Clinton win.

About as big as I expected, although not quite as big as I hoped, then again, there are always surprises in Presidential elections. I didn't expect Trump to get Pennsylvania (or Toomey to win), I expected Russ Feingold to win (yesterday saw a report on the general issue of Team Hillary ignoring/taking for granted the Rust Belt, including ignoring a plea for help from Feingold), and one of the biggest surprises is that Minnesota truly was in play (margin looks to have gone from ~200K in 2012 to ~40K).

>..a 1% shift from Trump to Clinton would have resulted in her victory.

Here's an excellent analysis from five-thirty-eight: http://fivethirtyeight.com/features/what-a-difference-2-perc...

Or maybe AP calls it if a candidate concedes? Guess we'll just have to wait, and for California, they've got over 4 million absentee ballots to count....

The only example I've heard of where "Ah, fuggit" kicks in is when enough ballots have been counted that the absentee ones can't change any race, and the extra labor is not considered to be worth it.

Except, for, you know, who administratively runs the country for the next 4 years, has the bully pulpit and therefore potentially great influence over the Congress, etc. etc. etc.?

It sure seems to mean a whole lot based on all the sad faces and rioting in the streets I've seen over the last few days.

And, circling back to the real point, it's the rules of the game, claiming some sort of greater meaning only when your side loses the popular vote is beyond pointless.

To remind you, my point is that the election was very close. It is not that the result is bogus or nobody cares or it's meaningless.

A tiny shift in sentiment would have changed the outcome of the election. The fact that the margin of electors is about 14% doesn't change that fact. The electors are not a meaningful measure of how close the race was.

If two countries play a best-of-5 soccer games, and one beat the other 3-0 with three consecutive 1-0 games in the last minute. It would have been hard to claim that one team isn't obviously better and the margin of winning is small.

Your analogy is flawed. The individual "games" weren't that skewed. Instead, imagine two basketball teams play best-of-5. The scores are 100-98, 101-99, 100-101, 99-100, and 98-100. It would be bizarre to say that one team was greatly superior based on that, even though they did have a big margin in percentage of games won.

On further reflection, I do agree that the election was close. I'm guessing my previous response was conditioned on Trump being the underdog, arguing on a different point you were making.

Your "most logical analogy" still isn't quite right. The individual scores for each game were very close, and the total scores actually slightly favor Clinton. But Trump won more games, after weighting the games in a certain way. But I don't know how to do it better, I don't think there's any sport that's really played anything like this!

I'd say the best measure for closeness would be either how much of a shift in the overall vote was needed to flip the result, or how much of a targeted shift could generate a change in the result. In this case, an overall shift of 1-2% would have changed the result, and a targeted shift of some tens of thousands of votes could have. It's not the closest election ever (2000 probably takes that prize by these criteria) but it's fairly close compared to, say, 2012 and 2008.

It would take an exceptional, once-in-a-generation, beloved Republican to win states like California, New York or Massachusetts, and I don't think Trump even tried there. Just the margins of victory in those three states are around 1/12th of Clinton's entire vote total.

I had some CNBC shows saved on my Tivo and I recently went back to look at what they were saying. Here's an extreme example of the deception. Just the day before the election CNBC was showing states such as Montana, North Dakota and South Dakota as "leaning" GOP, not "likely" GOP. But here are the results for those states:

   Montana      Trump 57% Clinton 36%
   North Dakota Trump 64% Clinton 28%
   South Dakota Trump 61% Clinton 32%

Yes there were true battlegrounds or toss-ups. E.g. Florida only went for Trump by about 1.3%. But there were more than 2x as many Trump voters than Clinton voters in North Dakota. It was a bald-faced brazen lie to call that "leaning".