������Ƶ

It’s hard to understand politics.��Some people don’t try to because they don’t care. They are totally disengaged, and they get picked on all the time by people who care a lot about politics: no vote, no grumble.

Political understanding is also elusive because people who are most passionate about politics — they can’t imagine not voting, for instance — typically think about politics in ways that limit understanding.

Many of these politically engaged people, let’s say the sort of folks who read Civil Beat, let their passion get in the way of their understanding.��They may be engaged, but in very limited, even counter-productive ways.��

These engaged people are confident that they know what’s right.��They substitute opinions for facts. Their concern is with justifying their own preference rather than understanding others. They lack humility and introspection.

Think of two parallel universes: Passion and Introspection.��Politically engaged people typically occupy only one.��They would be better citizens — and have a better understanding of themselves — if they found their way into the Introspection universe from time to time.

There are feasible, proven ways to blunt these single-universe predispositions and improve political understanding.

Well-Educated Tunnel Vision

Here is a typical conversation about Republicans and conservatives, versions of which I often have with friends who, like me, are highly educated, politically aware and liberal.�� (To be clear and fair, studies show that conservatives behave this same way regarding liberals.)

Friend: How can anybody who is smart or rational be a Republican?

Me: Well, I’m not sure if you are asking a rhetorical question or not. If you really want an answer, I can tell you what people who study political behavior say about that.�� (Then comes my analysis, which seems quick yet really incisive to me but, I can tell from the look, is long, tedious and irrelevant to my friend.)

Friend:��(Responding to this attempt at a teaching moment.) What? Are you turning into a conservative or something?

Me: No. I am just trying to answer your question, to get you to understand why conservatives believe what they do.

Friend: So what?��Conservatives are wrong.

Me: Don’t you at least want to know your enemy?

Friend: They are wrong.��Paul Krugman says so.

These friends have no interest in getting beyond a right versus wrong view that leads to tunnel vision. �� Their understanding of the opposition is based entirely on the opinions of those they agree with.��They don’t reality-test their beliefs.��They know they are right.

Of course they read about politics, but almost all of their information comes from pundits they agree with.��Most pundits are as bad about this as anyone, maybe even worse because they have such a bully pulpit and because they are so convincing in passing off their opinions as facts.��

Pundits are typically, in the novelist Jonathon Franzen’s words, “cherry-picking anecdotalists” disguised as empiricists.

There is nothing particularly surprising about this.��Generally people seek information that reinforces their beliefs.��What’s more, for all kinds of demographic and political reasons, conservatives and liberals today are much more likely to occupy separate universes.

Changing the way we think about politics is a challenge because, to oversimplify but not by much, our brains are hardwired to choose a position first and then search for information that justifies that position.��In the polarized political culture we inhabit today, that works out pretty well.

But these tendencies limit our politics and ourselves.

With Understanding Comes ‘Superforecasting’

Phillip Tetlock and Dan Gardner’s is the most interesting book I have read in a long time.�� (I’ll refer to the author only as Tetlock because the book is written in the first person, and he clearly is that person.)

Tetlock’s book is about forecasting the outcome of political events. The skills and approach necessary to be a successful forecaster are precisely the ones that increase political understanding.

Tetlock’s famous earlier research showed that on average, experts, including pundits and academics, did no better predicting the outcome of events than if they would have chosen the outcome randomly — the so-called chimpanzees throwing darts experiment.

However, that study also showed that some of these hundreds of forecasters actually did very well, beating the average by a long ways.

These superforecaster characteristics are precisely the ones that many people engaged by politics lack.

So Tetlock and others designed the Good Judgment Project, which involved hundreds of screened volunteers from many walks of life who made thousands of complex political forecasts over a number of years.��A panel of scientists assessed the accuracy of these predictions.

“Superforecasters” reports on what happens when a variation of the Good Judgment Project was one of the competitors in a forecasting tournament sponsored by the Intelligence Advanced Research Projects Activity,��a low-profile but extremely important U.S. government intelligence agency.

A significant group of Tetlock’s best forecasters—“superforecasters” he calls this group — did much better than any other group, including professional intelligence analysts who had access to classified information.

Overall, these superforecasters had IQs that were higher than average, but very few were geniuses.�� Tetlock shows that what made them so successful was not innate talent but rather��a particular way of thinking, of gathering information, of updating beliefs.��These habits of thought can be learned and cultivated by any intelligent, thoughtful, determined person.

In fact, the project developed a simple superforecasting primer based on its findings.��Those who took the hour needed to study these ����significantly improved their forecasting success.

Some of these commandments refer to certain skills, but the most important refer to the humility, skepticism and diligence needed to gather and assess information.

Superforecasters are extraordinarily diligent in seeking out information.��They are careful not to let their pre-existing opinions affect the information they seek.��They constantly look for new information pertaining to their prediction, whether that information supports or challenges their opinions.��They emphasize the degree of probability rather than choose between certainty and uncertainty.��They seek help from groups with diverse opinions, including ones that challenge their own.

These superforecaster characteristics are precisely the ones that many people engaged by politics lack.

It’s a leap from this successful experiment that required people to make probability predictions about things like the arc of the Ebola epidemic, the Nikkei Average, and Putin’s designs on the Ukraine to, say, predictions about elections, sustainability, or rail in Hawaii.

Still, it is possible for some people to develop the superforecasting skills and mind-set.��You may not get as good as Tetlock’s group, but some improvement is better than none. You don’t give up on the ukulele just because you can’t play as well as Jake Shimabururo

How can Better Understanding Actually Help?

To get a better sense of how this attempt at political understanding would help, think about answering these three questions about Hawaii:

• What is the probability that there will be an election in Hawaii in 2016?

• What is the probability that Brian Schatz will be re-elected U.S. senator in the 2016 election?

• What is the probability that the rail system on Oahu be completed by 2021?

The first question is a no-brainer.��The election is a constitutional requirement. People would not even bother to look for information that supports this sure thing.��And that would be all the understanding they need.

The Senate election is more complicated but manageable.��You can consult the history of previous elections in Hawaii, the relative strengths of the political parties, and some decent data about the national scene to come up with a probability estimate.

The Honolulu rail project, to put it mildly, is a work in progress, so as a superforecaster, you would constantly reassess your understandings in light of new information.

That’s not what most people do of course.��They would rely on what cherry-picking anecdotalists are saying without assessing how accurate or objective these experts are.��And of course we would rely on our own preferences more than we care to admit.��Thinking in probability terms?��Not likely.

The rail question?��There is a hard way and an easy way to answer this question.��The easy, constantly used, but unreliable way is to assume you know the answer and to look for information that supports it, most of which comes from people whose view is much stronger than the evidence to support it, sometimes because they have a stake in the outcome.

The hard but much more accurate way is to look at all kinds of other, more disinterested sources of information, like the history of other rail projects, and to keep reassessing your position in light of any new information you come across. The rail project, to put it mildly, is a work in progress, so as a superforecaster, you would constantly reassess your understandings in light of new information.

Doing this for rail is less feasible because it requires more digging and probably more introspection, but still manageable and certainly useful.

In the spirit of superforecasting, it would be wrong for me to make confident, sweeping predictions about how superforecasting would affect politics in Hawaii.��So take my predictions as hypotheses to be tested and reassessed over time, falling somewhere in the murk between fact-based claims and hopes and dreams.

• There is a low probability, somewhere in the vicinity of 10 percent, that more than a very small fraction of people in Hawaii try to use superforecasting to increase their political understanding.�� ��

• There is a high probability (85 percent) that those who do will significantly broaden their understanding and increase their humility and introspection about their beliefs.

• There is an uncertain probability (for now hedge at 50 percent) that over time this approach to understanding adopted by a few will pass on to others.

• There is an extremely low probability, 1 percent, that the understanding that superforecasting creates will alone create affordable housing or get rail built.

• There is a higher probability, 75 percent, that the understanding that comes from forecasting will make it easier to understand why and make it more likely for people to draw on more perspectives to try to meet these challenges.

You’re skeptical about my assertions?��You want to figure out what types of good information are available to examine them and make your own predictions? �� �� ��

If so, that’s an excellent beginning — a training exercise for the real thing.