[T]he disagreement between polls this week was on the high end, and that makes it harder to know exactly what the baseline is heading into Monday’s debate. The polls-only model suggests that Clinton is now ahead by 2 to 3 percentage points, up slightly from a 1 or 2 point lead last week. But I wouldn’t spend a lot of time arguing with people who claim her lead is slightly larger or smaller than that. It may also be that both Clinton and Trump are gaining ground thanks to undecided and third-party voters, a trend which could accelerate after the debate because Gary Johnson and Jill Stein won’t appear on stage.I don't know why When should I panic? is a relevant question (unless it's a trigger to desperate measures). I guess it's mostly that Silver makes it his business to assure Democrats with numbers. He translates the poll data into a more soothing likelihood of winning.
२४ सप्टेंबर, २०१६
"The Case For And Against Democratic Panic."
From Nate Silver.
याची सदस्यत्व घ्या:
टिप्पणी पोस्ट करा (Atom)
पोस्ट्स
२५ टिप्पण्या:
Shorter Silver: "Who knows?"
I started panicking in the spring. It's the enthusiasm gap that could do the Dems or Hillary in. The Allan Lict or Lictman thing is more meaningful IMO than anything Silver does.
"He translates the poll data into a more soothing likelihood of [Democrats] winning."
Maybe so, but in 2012 I fell for all that "the polls are oversampling Dems!" stuff and Silver nailed it, so I've been checking in on fivethirtyeight.com pretty often.
If I were a Dem I'd be most worried about that "winding path" graphic over there that our host likened to an intestine. Seeing that all he has to do is hold all the states currently in his column or leaning that way, plus flip NH or CO or PA or MI? All of a sudden that "Trump has almost no path to 270" isn't so convincing.
Of course, as Silver and others acknowledge, Monday night could change a lot, one way or the other.
I don't think this election will be decided until the debate, i know that historically the debates don't matter much poll wise but i argue this year is different. Clinton is a poor retail politician, except for screaming how HORRIBLE! Trump is, i don't have the first damn clue as to why she wants to be president.
"I guess it's mostly that Silver makes it his business to assure Democrats with numbers." Win-win-in: he makes money, the Dem fund-raisers exploit the panic, and Dems easily win anyway.
"i don't have the first damn clue as to why she wants to be president." So what? She will be a Dem president doing Dem things, appointing Dem officials, supping Dem legislation in Congress, and turning the Supreme Court left.
I think the support is skittish on both sides, but heading slightly in Trump's direction, and there is so much more of little things that could happen and upset the applecart for Hillary between now and Nov. 8.
And she is of the incumbent party, which will be blamed if something really bad happens.
Blogger james conrad said...
"...i don't have the first damn clue as to why she wants to be president..."
It's all about ego and wanting pay back after all of the years of humiliation from Bill.
I have a rabid pro Hillary, feminist liberal friend.Normal in every other way. Even she is concerned about Clinton's health status.
While I would not hope for Clinton to have a stroke or fall and break a hip, a few coughing and harmless mini seizures spells would break the expected monotany od the debate.
BTW, much is made of Clinton's small audiences lately. Wonder if it is intentional in order to minimize danger of large crowd or noise triggering an episode.
If you believe that Trump is Hitler, and if elected will require that we all wear our underwear outside our clothes, then it's always time to panic.
Of course, if you have the instincts of a Stalinist, then a lot of people look like Hitler. From that point of view, the Politburo has done the thinking for us and has chosen HRC for the succession. The gates of the Reeducation Camps yawn wide for those who resist the Will of the Party.
If you consider the CW, then Trump gets a break by being the Unknown. On the other hand, Clinto gets a sympathy break if she doesn't fall over. So, IMO, the debate CW breaks both ways and it comes down to one thing: who looks stronger? I expect Clinton to be very wonky and Trump to be very general. The debate spin on debate night is that she killed Donald. But the polls will drift in his direction I think. Unless one or the other of them screws up. Then it will be over because debate 2 and 3 don't get near the viewers.
It is not over until the count is in - and that may be in the Electoral College rather than the election, if it is close.
Nobody ever tells us WHEN to panic, only when NOT to.
"It's all about ego and wanting pay back after all of the years of humiliation from Bill."
Yup. She is a hater and an egomaniac.
Fits right in DC.
With fivethityeight their all lefties so best to ignore the commentary and focus on the data. For the most part they show their work.
Most of the polls I've see haven't drifted as much in her direction as one might think.
When things drifted in Trump's direction, the assumptions changed the R and D numbers were pretty close with Dems holding a slight lead.
The New polls have Dems with 8-10 points over Reps. Thus, Hillary leads.
The numbers are all coming out the same though. Just the weighting that is changing. Which I don't have an explanation for.
Another thing I noticed. I like to compare one poll against the same pollster.
NBC/Wall Street Journal has her up 7. The last one from the same pollster had her up +9.
McClatchy/Marist also has her at +7. The last one had her at +15.
So while the +7 in both are much better than she was getting, against those two, they are worse.
In one case, much worse.
… because Gary Johnson and Jill Stein won’t appear on stage.
Suggested Gary Johnson campaign slogan: Stranger Together
The Democrats should have started panicking in 2014.
JPS,
Most polls over-represent Dems or Reps- it is just the nature of random sampling that there is variance in the data sets. However, in 2012 the polls were adjusting to the demographics of 2008, so it was natural to weight towards that election and it would look like an oversampling to a Romney supporter, and Obama did largely repeat his coalition from the 2008 landslide, losing only a couple of percent of the white vote.
The polls this election are more uncertain, in my opinion- do you weight like it was 2008 and 2012, or does it look more like 2000 and 2004? I don't think there is a chance in Hell of Clinton pulling the gross numbers of African-Americans, even if she somehow manages to hold Obama's percentage in that demographic (something I think is very unlikely, too). I do think the polls weighting the turnouts of 2008 and 2012 will undercount Trump's support by at least 1 to 4%.
New ABC/Washington Post poll tonight has Clinton at +2.
Two weeks ago the same poll had Clinton at +8.
I'm not very good at math, but, this doesn't seem like a good thing for Clinton in the new ABC/Washington Post poll.
Clinton gets 88% of Dems, 38% of Indies and 5% of Republicans.
Trumps gets 90% of Republicans, 43% of Indies and 8% of Dems.
So, he gets more party support, more independent support, and a larger share of the other party than Clinton, and he's behind by 2 points?
The rule in the stock market is "Panic early, before everyone else does." Pretty good advice most of the time, but probably not applicable here. The voters can see and smell the panic and react against the panicking candidate. (See Dukakis + Tank = Ridicule.)
Rob McLean, 6:02:
"Nobody ever tells us WHEN to panic, only when NOT to."
Ever see A Shock to the System? "Whoa, whoa, whoa! Let's not all panic! You, you and you panic; the rest of you, stay calm."
Yancey Ward, 11:25:
Good points. I tend to agree.
"eric said...
So, he gets more party support, more independent support, and a larger share of the other party than Clinton, and he's behind by 2 points?"
There really are significantly more people who identify as Democrat than there are who identify as Republican right now. Heck, whether I identify as a Republican or as Independent depends on the mood you catch me in.
Republicans are more likely to be "a pox on both their houses" than Democrats, right now.
The big question in translating polls to electoral results is always turnout -- The election probably depends on whether the Democrats get more of their "hold my nose and vote against Trump" voters to the polls than the GOP do their "hold my nose and vote against Hillary" voters in a handful of "swing" states.
I was aware of more "crawl-over-broken-glass" GOP voters four years ago, but the same seems to be true on the other side. And how do you accurately measure that?
टिप्पणी पोस्ट करा