Loki Liesmith
Loki Liesmith's JournalHow the Trumpists are trying to game debate expecations
Lowering Expectations:There is a conventional wisdom hardening which holds that Hillary Clinton’s most urgent task at the first debate is to bait Donald Trump into acting like the ignorant, dangerous, hateful, bigoted, temperamentally unhinged character he has periodically displayed to the nation for the last year. The corollary to this is that if Trump acts relatively controlled and projects minimal seriousness, he might “defy expectations” and emerge the winner.
I don’t buy it. While this is certainly a factor worth considering, I’m going to suggest that Clinton can win this debate — in the minds of voters, if not in the minds of pundits — even if Trump does pull off that magic transformation for ninety minutes on Monday night.
No question, Team Clinton probably would like to see Trump come unglued before an audience that could be as large as 100 million people. The New York Times reports that Team Clinton is trying to determine “how to knock Mr. Trump off balance,” in the belief that “she needs the huge television audience to see him as temperamentally unfit for the presidency, and that she has the power to unhinge him.”
And no question, Team Trump believes that avoiding that outcome is key to his hopes of prevailing. The Associated Press reports that Trump’s advisers are counseling him to avoid letting Clinton rile him up, noting, remarkably, that “some Trump aides are more concerned about Trump’s disposition on the debate stage than his command of the issues.”
A tweet/tag to retweet if you want to help remind the media to do it's job:
https://twitter.com/lambertglowbug2/status/779335252204457985
#LoweringExpectations
Election Model Update 9/22/2016
#############################################
# INSTANTANEOUS MODEL
#############################################
Instantaneous Probability of a Clinton Win: 0.684513006655
Mean Electoral Votes Clinton Currently Winning: 285.9956
Median Electoral Votes Clinton Currently Winning: 288.0
Maximum Likelihood Electoral Vote Scenario for Clinto: 287
#############################################
# PROJECTING RESULTS TO NOVEMBER
#############################################
Probability of a Clinton Win if Current Trends Continue: 0.627906976744
Mean Electoral Votes Clinton Projected to Win: 283.764
Median Electoral Votes Clinton Projected to Win: 284.0
Maximum Likelihood Projected Electoral Vote Scenario: 270
Predicted National Point Spread (Clinton = +, Trump = -) 3.03117822284
Today's polls show that the Kerry Wall has held
That is a massive advantage for Democrats in this election and going forward. If the last two weeks didn't breach that wall, I don't think that much can.
Election Model Update
Last Update######################################
# INSTANTANEOUS MODEL
######################################
Instantaneous Probability of a Clinton Win: 0.676879661359
Mean Electoral Votes Clinton Currently Winning: 285.7502
Median Electoral Votes Clinton Currently Winning: 288.0
Maximum Likelihood Electoral Vote Scenario for Clinton: 304
######################################
#
# PROJECTING RESULTS TO NOVEMBER
######################################
Probability of a Clinton Win if Current Trends Continue: 0.632426988922
Mean Electoral Votes Clinton Projected to Win: 283.84
Median Electoral Votes Clinton Projected to Win: 285.5
Maximum Likelihood Projected Electoral Vote Scenario: 288
Estimated National Spread: +3.04, Clinton
Instantaneous Win probability has shown some decline from last update, but given a bad few weeks of polls, I think the hints of recovery today have helped a lot.
No more analysis, somehow someone will accuse me of concern trolling for posting what I think are some positive numbers.
Maybe I'll post some graphics later.
Hypothesis on the nature of this election: 1980 redux
Hackneyed pols often trot out the example of 1980 and Ronald Reagan as key to understanding the effect of debates.
In that election, RR was tied or trailing Jimmy Carter for months until the debates "showed" that RR wasn't a complete corpse on stage. It
assured the public (or so we are told) that Reagan was ready...and people fell in line to vote for him.
A lot of people have been presuming this to be the template for the current election...believing that HRC is carter and Trump is Reagan.
But what if this is wrong? Clinton's numbers are the most elastic: she passes 50% frequently in the polls, when not under press attack.
Trump NEVER does. People want to vote for her over Trump, but are concerned about the bad press she has gotten.
So maybe this means that people are waiting for a reason to get on board with Hillary. Maybe SHE is Reagan in this scenario.
Just a thought.
Trump has made essentially no progress in Wisconsin for months
Clinton has only slumped a bit. She wins this on turnout
New Hampshire Monmouth Poll
HRC 47
DJT 38
GJ 10
JS 1
Epic result.
We own the map now. Probably.
If Marquette's Wisconsin poll is good for Clinton I'll eat my hat
They have had Trump closing for weeks. Brace yourselves...incoming.
NC Poll by PPP: not a great result
Donald Trump 45
Hillary Clinton 43
Gary Johnson 6
Clinton/Trump head to head is tied at 47
This is not a "sick Hillary effect". Poll was Sunday-Tuesday.
NC will be a hard state to win.
Here is why 538's odds on Hillary are decreasing while The Upshot is increasing.
I knew this, but had put it out of mind. 538 aims to be a predictive model and in doing so it projects trends in horse race numbers forward in time. Today's polls by Monmouth and St. Leo in Florida both had larger leads for HRC way back in August. So todays lead for her is indicatrive of downward trend in her numbers. So her numbers, coupled with weirdo Reuters semi-polls, actually got WORSE.
Frankly, I think this methodology makes no sense. If you do believe in statistical momentum (and I'm not sure I do), incorporate it into your model using the agggregate horse race numbers, not individual polls. If you fit the trendlines to individual polls, you effectively penalize that poll for having drawn an outlier in the past. I can't see a rationale for that at all.
Fitting data in this manner will cause your model to jump around haphazardly, based on the coupling coefficients to each polls trendline, because the polls come at stochastic time. Using an aggregate trend over all polls makes more sense if you value consistency in a model.
Nate Cohn, at NYTimes Upshow seems to have noticed this as well, and I don't think he agrees with Silver based on his twitter timeline.
Profile Information
Member since: Thu May 26, 2016, 09:07 AMNumber of posts: 4,602
