2016 elections - because it's never too early

[Hamilcar]

I love their logo.  Looks like something you'd find on a chocolate candy package.

That's actually more or less the idea. Brand your right wing party with a smiley, friendly sun. Genius.

[celedhring]

snip

Never voted SVP, never will. Ugh.

Try again 

I love their logo.  Looks like something you'd find on a chocolate candy package.

It even says "Swiss quality" on it.

[Razgovory]

snip

Never voted SVP, never will. Ugh.

Try again 

[Hamilcar]

In any case, the left in the western world needs to take a hard look at itself, shape up and step up, or the fascists really will take over. Le Pen for example scares me far more than Trump.

[celedhring]

I just find it troubling to "forget about race", when race is such a major component of inequality in the west.

[Hamilcar]

I just find it troubling to "forget about race", when race is such a major component of inequality in the west.

I should clarify: we need to drop this absolute obsession on race, this elevation of race over so many other relevant variables. It's actually deeply racist. "All black are like X", "Whites all are like Y".

[Barrister]

Oh man, all the SJW types in my social networks are in hysterics. Don't get me wrong, I think Trump will not be a good president, but a small part of me can't help but smile at the meltdowns.

After such a clear defeat, maybe my fellow liberals can cut the virtue signaling identity politics crap, and focus on liberty, equality and human progress again.

I don't know about "smile", but I'm already getting tired of all the hand-wringing going on in my office.

It's okay to be interested in the election result (I know I was) but fundamentally it's not our country.  The US can elect whomever the hell they want to.

[Malthus]

Oh man, all the SJW types in my social networks are in hysterics. Don't get me wrong, I think Trump will not be a good president, but a small part of me can't help but smile at the meltdowns.

After such a clear defeat, maybe my fellow liberals can cut the virtue signaling identity politics crap, and focus on liberty, equality and human progress again.

I don't know about "smile", but I'm already getting tired of all the hand-wringing going on in my office.

It's okay to be interested in the election result (I know I was) but fundamentally it's not our country.  The US can elect whomever the hell they want to.

Two different issues.

1 - acknowledgement that the US can elect whoever they want to.

2 - horror at who they wanted to elect, the likely damage that choice will do to both us and them.

[The Minsky Moment]

I really dislike the "got their call wrong" talk.  If I told you you have 83% chance of surviving the game of Russian roulette, and you blow your brains out, did I get the call wrong?  Validating predictions is a very difficult challenge that requires a lot fo data.

I'm having trouble following your logic.  Are you saying if we re-ran the election many times, Hillary would win 5 out of 6 (or whatever)?  That the actual outcome was just the tailish result of certain people staying home, that kind of thing?

Not exactly it's a different kind of uncertainty.
It's more like forecasting economic variables, like GDP, or unemployment.  Or making assessments of likely actions by an adversary relying on various forms of intelligence.  There are indicators that are used for these kinds of tasks that have real correlation with outcomes but there is uncertainty associated with them.  But using the indicators however uncertain is better than just randomly guessing.  So you say something that - that data/indicators/intel indicates X will happen, but because there is noise or the data is imperfect we assume a certain % chance that it's not capturing the true state of affairs

HRC would lose 100 out of 100 but there is no way to know that a priori.  You can only estimate a priori using the data you have, knowing that there is uncertainty involved.  So you have a probabalistic model  that recognizes that uncertainty i.e. the probability that your indicators don't precisely and accurately reflect the true state of the world.

That's why Silver seems to have "got it right": he basically said the data shows strong evidence of a Clinton victory but there is a lot of uncertainty so there is a significant probability that the data has it wrong.

[Admiral Yi]

Not exactly it's a different kind of uncertainty.
It's more like forecasting economic variables, like GDP, or unemployment.  Or making assessments of likely actions by an adversary relying on various forms of intelligence.  There are indicators that are used for these kinds of tasks that have real correlation with outcomes but there is uncertainty associated with them.  But using the indicators however uncertain is better than just randomly guessing.  So you say something that - that data/indicators/intel indicates X will happen, but because there is noise or the data is imperfect we assume a certain % chance that it's not capturing the true state of affairs

HRC would lose 100 out of 100 but there is no way to know that a priori.  You can only estimate a priori using the data you have, knowing that there is uncertainty involved.  So you have a probabalistic model  that recognizes that uncertainty i.e. the probability that your indicators don't precisely and accurately reflect the true state of the world.

That's why Silver seems to have "got it right": he basically said the data shows strong evidence of a Clinton victory but there is a lot of uncertainty so there is a significant probability that the data has it wrong.

Would another way to put it be that Guller is not defending the pollsters (who did in fact fuck up) but rather the people who used those polls to predict the outcome?

[derspiess]

I read an article this morning that quote someone from The Atlantic as saying, "The press takes him literally, but not seriously; his supporters take him seriously, but not literally."

I think that is 100% spot on. 

[jimmy olsen]

So weird. GOP loses ground in the House and Senate, but takes back the presidency with the most unpopular major candidate in modern history.

It was all about turn out. Trump didn't get any more votes than Romney. Clinton simply woefully underperformed Obama.

[Fate]

So weird. GOP loses ground in the House and Senate, but takes back the presidency with the most unpopular major candidate in modern history.

It was all about turn out. Trump didn't get any more votes than Romney. Clinton simply woefully underperformed Obama.

All of the votes aren't counted. He likely did better than Romney and McCain by 2-3 million votes. Agree on the woeful under-performance of Clinton.

[The Minsky Moment]

Would another way to put it be that Guller is not defending the pollsters (who did in fact fuck up) but rather the people who used those polls to predict the outcome?

yes
but it's also not clear how much the pollsters are to blame.  You can't assess the size/direciton/possible cause of error until after the fact.

[jimmy olsen]

So where did the Democratic Party go wrong, and more importantly, where do they go from here?
It's beyond Hillary: the Dems simply can't win at any level.

They gained seats in the Senate; and will almost certainly take control of the Senate and probably the house as well in 2018.

2018 is going to be a bloodbath for the Dems in the Senate. There are a whole lot of blue seats in precarious states in this next cycle. Maybe they could get Nevada, but can you really make an argument for them picking up anything else?

2018 Senate map...


Right now what I see are a lot of Trump favorable rust belt states and red states that democrats really have no right winning anymore. If the Reps ran the table they could get a supermajority in the Senate for the first time since the early 1900s and Reconstruction. Absolutely no way Dems could get a majority unless there's a shit ton of unexpected retirements in the next two years.

The president's party almost always loses seats. SInce there's likely to be a bad recession that will probably happen again.