Illusion of Accuracy

The spheres are in commotion
The elements in harmony
She blinded me with science!
And hit me with technology
--Thomas Dolby

After the 1992 presidential election results came in, I remember my company's former CQO telling me that the accuracy with which the polls forecasted the election results spoke to the power of statistical sampling in estimating outcomes for a population. Seemed reasonable to a rookie like me back then.

Were I to respond to his assertion today, I would say that there is more to predicting outcomes from human subjects than a reasonable sampling plan--which is no small task in and of itself. How the questions are asked also makes a difference. For example, ex ante questions such as "Who do you plan to vote for?" has a different response profile than "Who will you vote for?" as does "Who did you vote for?" ex post.

Another issue involves the difference between what people say they will do and what they actually do. Survey respondents might say they will vote for one candidate, but actually vote for another. Or perhaps they don't vote at all. People might indicate that they do not plan to vote but subsequently head to polls and vote--perhaps for the candidate deemed least likely by other survey responses.

The +/- X% 'margin of error' numbers provided by pollsters provide an illusion of accuracy when underlying methodologies fail to compensate for the true sources of error in survey research.

This is precisely what we saw in the run up to election day. People blindly hanging their hats on forecasts provided by pollsters who offered the illusion of accuracy when, in reality and with notable exceptions, the polling methods generated estimates that were spectacularly wrong.

Why the error among pollsters was so skewed in one direction (which, fittingly, is known as 'bias' in survey research) is a subject for another day.

New New Deal Rally

I was dreamin' when I wrote this
Forgive me if I go astray
But when I woke up this morning
Coulda sworn it was Judgment Day
--Prince

Would be remiss if we didn't recount the tremendous amount of market movement since election night. When the tables began to turn in favor of Trump on Tuesday nite, futures markets immediately fell out of bed. The Dow was down nearly 1000 pts and many markets were 'limit down' at about -5%.

However, when Trump spoke after the election had been called about getting the economy going again via government 'investments,' markets began trimming their losses. By the time the opening bell rang yesterday morning, equities were in the green and didn't look back. The Dow lifted 300 pts and finished with in spitting distance of all time highs.

What happened? Beyond the maniacal behavior of traders seeing themselves caught on the wrong side of 'common wisdom' about the election, a primary factor in the move higher is hope that Trump will drive the investment in infrastructure projects to stimulate the economy. Banks, retail, healthcare, heavy equipment, et al. Many stocks in these sectors are closing in on 10% gains since Tuesday's close. A New New Deal rally

As we know, the prospect of more stimulus peps up addicted markets.

Unfortunately bonds didn't play along. As stocks partied higher yesterday, Treasuries were crushed. Ten year yields are now north of 2%--a place they haven't been in nearly a year.

Why the selling? Government stimulus projects need to be paid for. Bond markets are beginning to discount how much Treasury debt will be required to fund these projects. Interest rates are being remarked to reflect higher borrowing costs.

If govvies continue to get splattered, then it is only a matter of time before bond yields stop the New New Deal stock rally in its tracks.

positions in gold, SPX

Trump's Turn

To everything (turn, turn, turn)
There is a season (turn, turn, turn)
--The Byrds

The swing felt similar to twelve years ago. As initial results began to trickle in on election night in 2004, polls and pundits were firmly in John Kerry's camp. A few early states were called in his favor. Prediction markets, as I recall, assigned a 90%+ probability of a Kerry victory.

I decided to go for a late evening jog. As my feet crunched leaves on the sidewalk beneath the streetlights, I remember thinking that Bush will need to pack his bags soon.

By the time I had returned, however, the entire race had flipped. W had unexpectedly won a couple of blue and battleground states, and the incoming vote tallies suggested that Bush's election night momentum was building.

I went to my computer to check out the prediction markets. The Bush and Kerry contracts had essentially traded places. It was now the Bush contract that was valued at 90%.

Just like that.

Last night Donald Trump was elected president of the United States with much the same election night dynamic. Polls, prediction markets, and pundits overwhelmingly favored a Hillary Clinton victory. Early election results did not offer much encouragement for Trump supporters. For example, in my home county considered central to 'battleground' Ohio, initial counts of early voter ballots spotted Hillary a 20% lead.

However, as ballots from Eastern Time Zone polls were being tallied, it became evident that support for Clinton in crucial states was weaker than expected. In fact, Trump was amassing sizable leads in many of these states.

Prediction markets and other odds-making sites such as the NYT forecasting site began to flip dramatically, with Trump moving from 20 to 50 to 70 percent odds of winning within a couple of hours-leaving many commentators flabbergasted. By the time battleground states such as North Carolina, Ohio, and Florida began to be called for Trump, his chances of winning had increased to 75 - 90%.

Trump subsequently crossed the 270 electoral vote mark early Wednesday morning.

Just like that.

How did all of the prognosticators get it so wrong? That question will occupy analysts for some time. My sense is that, similar to 2004, the "had enough" factor was significant. Many people have had enough with status quo politics, and they view Donald Trump as an outsider, albeit a 'wildcard' one, with capacity to make change.

Many onlookers also perceived a lack of procedural fairness in how Trump was being treated by the media et al during the course of the campaign. At some point those onlookers had enough of the biased treatment and acted at the ballot box to express their displeasure.

The "had enough" factor brought people out of the woodwork to vote for Trump who otherwise would not have. Pollsters did not adequately account for this factor in their models.

Trump's entire campaign, including his dramatic turnaround on election night, will go down as one of the biggest stories (and case studies) in modern American politics. Where his term as president goes may be as unpredictable as was his path to the White House.

Get Out and Vote?

"It is a kingdom of conscience, or nothing."
--Balian of Ibelin (Kingdom of Heaven)

"Get out and vote." The mantra reaches a crescendo on election day. Since childhood, we are told that because we live in a democracy, voting is our civic duty. Ballots cast reflect 'the will of the people.' And that voting processes grounded in the democratic principle of 'majority rule' are ultimate expressions of freedom.

Perhaps it is the frequency with which these claims are drummed into our heads that causes us to generally accept them as is. However, people willing to don their critical thinking caps in search of truth of likely to conclude that some if not all stock claims underlying the "get out and vote" argument are contestable.

Let's consider one counterargument here.

The proper scope of government is to assist people in protecting their property (broadly construed) from aggression by others. When government oversteps its rightful boundaries by forcibly taking property from some for the benefit of others, then it becomes the aggressor.

When people vote for candidates or laws with demonstrated capacity for committing such aggression, then those voters become partners in crime. When they cast their ballots, those individuals are pulling levers of aggression. They become principals of violence who, by casting their ballots, are contracting with strong armed agents to do their bidding.

Under such circumstances, is there a case for casting a vote? If the situation involves voting for a candidate who sponsors less aggression than another candidate (i.e., the lesser of two evils), then the answer is no. Voting for a candidate who promises aggression against a minority of people is still employing offensive force. Greater good accounting does not suffice.

In some cases, however, it may be possible to vote for pushing government back toward its proper scope in a peaceful manner. Although they are rare, candidates occasionally surface with capacity to shrink government without side helpings of aggression. One can also vote against initiatives such tax levies in order to decrease the degree of forcible appropriation of property in the system.

Beyond those few cases of voting with peaceful profiles, "get out and vote" involves securing the services of political agents to act aggressively against others.

Evil Voting

"There will be a day when you will wish that you had done a little evil to do a greater good."
--Sybilla (Kingdom of Heaven)

The primary heuristic people seem to prefer when voting for president this time around is: vote for the lesser of two evils.

No big deal, some might say, that's how it nearly always is. Few candidates are perfect. Compromise is required.

If that has been the dominant mindset all along, then it should not be a surprise that we are where we are.

When you compromise with evil, you increase evil.

Comey Redux

I can't stand this indecision
Married with a lack of vision
Everybody wants to rule the world
--Tears for Fears

FBI Director James Comey sends another letter to congressional leaders saying investigation of Clinton emails show nothing new. Media of course takes this as clearing Clinton once again.

Certainly a plus for Clinton two days before Election Day.

However, Comey still faces problem of internal mutiny. There is also the investigation into the Clinton Foundation.

Remember, Remember

"Truth be told, if you're looking for the guilty, you need only look into the mirror."
--V (V for Vendetta)

Lots of angst and finger pointing as election day approaches and likely to be more after Tues. As V reminds us on this Fifth of November, our government is only a reflection of its people.

Our dystopian problem does not rest with the other guy. The problem lies within.

Milgram's Obedience Experiments

Male student: I'm getting a little tired of this.
Dr Peter Venkman: You volunteered, didn't you? We're paying you, aren't we?
Male student: Yeah, but I didn't know that you were gonna be giving me electric shocks! What are you trying to prove here, anyway?
Dr Peter Venkman: I'm studying the effect of negative reinforcement on ESP ability.
Male student: Effect? I'll tell you what the effect is. It's pissing me off!
--Ghostbusters

In the early 1960s, social psychologist Stanley Milgram of Yale University ran a series of famous experiments aimed at studying 'destructive obedience.' The gist of the experiments (Milgram, 1963) was that most subjects would deliver excruciating electric shocks, perhaps even sometimes debilitating or lethal, to innocent people when the experimenter told them to do so.

When subsequently explaining his findings in the popular press, Milgram (1973) suggested that one reason why people go along with malevolent authority is that they feel arrogant, uncoorperative, and rude if they do not obey. Commitment to politeness overwhelms commitment to basic moral decency.

Milgram's research helps explain the political correctness movement, unwillingness to testify, party loyalty under conditions of political corruption, and other collectivist phenomena.

Reference

Milgram, S. (1963). "Behavioral study of obedience." Journal of Abnormal and Social Psychology, 67(4): 371-378.

Milgram, S. "The perils of obedience." Harper's, December, 1973, pp. 68-77.

Clinton Foundation Investigation

Always slipping from my hands
Sand's a time of its own
Take your seaside arms and write the next line
Oh, I want the truth to be known
--Spandau Ballet

Multiple sources are reporting that the real elephant in the room for Hillary Clinton is the FBI investigation into possible improprieties of the Clinton Foundation. This investigation is separate from but related to the Bureau's investigation into Clinton's email improprieties while secretary of state.

Momentum of the Clinton Foundation investigation is building due to the Wikileaks publication of thousands of hacked emails including those from Clinton campaign manager John Podesta, and possibility of new evidence found on on Anthony Weiner's laptop.

The primary focus of the investigation is the possibility of 'pay-for-play'--i.e., donors such as foreign dignitaries paying the Foundation in exchange for inner access to US foreign policy information and favors via Clinton's State Dept. Stated differently, the investigation is trying to determine whether Hillary Clinton put US foreign affairs on sale.

Sources suggest that evidence that implicates the Clinton Foundation is mounting at a rapid pace and that, barring obstruction from, for example, Obama's Dept of Justice, that an indictment is likely pending.

Early Voting

"Appeal to their emotions. Make them laugh. Make them cry. Make them mad, even if they get mad at you. But for heaven's sake, don't try to improve their minds."
--Jack Burden (All the King's Men)

For the past few weeks my Twitter feed has been sprinkled with paid messages from Hillary Clinton to vote early. It is one thing for a person to decide to vote early and another thing for a candidate to advise it.

A person might vote early for various reasons. A schedule conflict on election day. Not wanting to deal with the possibility of standing in line to cast a ballot. Dislike of the polls that troll outside voting places.

The primary risk of voting early, however, is that individuals cast their ballots before all the information is in. New information could turn up that might alter voting decisions. Unfortunately, this new information can't be put to use by early voters because their ballots have already been submitted.

But why would a candidate advise people to vote early? Wouldn't a candidate want voters to weigh all information that comes in right up until election day in order to make the most informed choice?

I can think of at least two reasons why candidates would want people to cast their votes early. The first reason is that the candidate has an early lead in the polls and wants to preserve that lead by locking in supporting votes as quickly as possible. Not only does the leading candidate collect those votes, but polls that report early voting trends that favor the candidate might lead to subsequent bandwagon effects that a) push undecided votes toward the leader, or b) discourage those who might vote for opposing candidates in doing so.

The other reason is that a candidate fears that information might emerge between now and election day that paints the candidate in an unfavorable light. Better to lock in voters now before they might possibly change their minds in lieu of potentially damaging information. Grab those voters while the getting is good.

Interestingly enough, Google Trends indicates that the search phrase 'change early vote' has been trending since news broke late last week of the reopening of the FBI's case against candidate Clinton.

Candidates who seek support from the most informed voters possible are unlikely to advise people to vote early.