What kind of a claim is “Biden has an 87% chance of winning”? (on the metaphysics of probability)

Leave a reply

If you’re spending all your time nervously checking the election forecast on FiveThirtyEight.com, your mental health may suffer. You can stop checking and do something productive to improve the world. Or you can become intrigued about what a forecast means, read Alan Hájek’s “Interpretations of Probability” in The Stanford Encyclopedia of Philosophy (Fall 2019 Edition), and write some rambling reflections. That is the Path I have chosen.

Four years ago, Nate Silver’s FiveThirtyEight gave Hilary Clinton about a 75% chance of winning the 2016 election. The fact that she lost did not invalidate this prediction–outcomes with probabilities of 25% happen often. Looking retrospectively, it seems right that Trump’s chances were small. He had to win narrowly in just the right combination of states.

Of course, we now know that he did win. In October 2016, an omniscient deity would have known that already. The deity could have known it in either of two ways: by looking into the future, or by understanding the complete present situation with one week still left to go in the election. Presumably, if you could know exactly what every American was thinking about politics, the precise distance to their nearest polling place, whose contact lists everyone was on, what Putin was up to, how heat and humidity were distributed on the face of the globe, and everything else about the situation with one week to go in the election, you would know what would happen with the vote. (I leave aside the possibility that that the universe incorporates physical randomness at the quantum level that affects things like the outcome of an election a week away.)

Applying that theory to our present circumstances, we would say that either Biden has a 100% chance of winning the 2020 election or Trump has a 100% chance. These are falsifiable claims, and a maximum of one of them will turn out to be true. Every other probability estimate will turn out to be false, because either Biden or Trump will actually win.

Yet is seems rational to say that Biden has almost a 90% chance of winning right now, and wrong to say that he has a 100% chance–and even more wrong to say that Trump has a 90% chance. A lot of data and experience go into a plausible prediction. Even if Trump will win in 2020, he doesn’t have a 90% chance right now. Another Trump victory would be a second improbable event. But again, the actual vote won’t invalidate either a 10% or a 90% estimate of Trump’s chances, because either one is compatible with him winning or losing.

A different way to make sense of this is to say: If the election were held 100 times, Biden would win almost 90% of them. But that is weird in several ways. The election cannot be held 100 times in a row, and if we repeated it at all, the repetition would affect the outcomes. If we imagine 100 identical universes that all unfold separately from now until next week, perhaps Biden would win in 90 of them. Or perhaps the future is determined by the current situation, which must the same in all of the 100 identical universes. Then they must all turn out the same way. We just don’t know which way.

Forecasters like Nate Silver use simulations with random (or pseudo-random) numbers built in. Those are meant to model the actual world. But they are not replicas of the real world, which–leaving aside quantum physics–seems to have just one future that is determined by the state of things now.

Another interpretation is that giving Biden a 90% chance today is simply an assessment of our knowledge level. It’s as much about us as it’s about the world. Biden actually has a 100% or a 0% chance, but we (unlike an omniscient deity) don’t know which of those is right. However, the tools of forecasting allow us to estimate how much knowledge we have–with precision. In fact, Nate Silver’s estimate rises and falls by the hour.

According to this subjective interpretation of probability, when Silver’s estimate moves from 85%-86%, he has not invalidated his previous prediction but has updated his assessment of the best possible level of knowledge at the present time. Once the election is over, our knowledge will become complete, and we will rightly say that the odds are 100% in favor of what actually happened.

Two problems occur to me about this interpretation. First, a prediction is not falsifiable in the usual way (and falsifiability is a hallmark of science).

Second, how much knowledge is “possible” is relative to circumstances. Anyone who could see all the current, private, survey data at the congressional-district level would have more knowledge than Nate Silver has. But he knows a whole lot more than I do. His estimates seem to be measures of how much certainty he is entitled to, based on the work he has done and money he has spent. If I say that Biden has an 87% chance because that’s what I read on FiveThirtyEight, I am really saying that I believe Nate Silver’s claim that he has an 87% level of confidence. But how could I test whether that estimate is correct? How can we know that he is right to raise or lower the estimate by a point? Certainly not by waiting to find out what happens next week.

affective polarization is symmetrical

Leave a reply

Shanto Iyengar and colleagues write, “Democrats and Republicans both say that the other party’s members are hypocritical, selfish, and closed-minded, and they are unwilling to socialize across party lines, or even to partner with opponents in a variety of other activities. This phenomenon of animosity between the parties is known as affective polarization.”*

Although affective polarization is far from our only problem, it does make public deliberation more difficult and hence undermines democracy. It is a particular problem for people who would like to build up public institutions: progressives. A minimal state doesn’t require much public comity, but an ambitious one does.

Iyengar and colleagues treat affective polarization as symmetrical, depicting both parties as equally affected. Based on the graphs below (which I contribute), I think that is fair.

The first one shows that in the late 1970s, Democrats and Republicans both rated the other party at almost 50 on a 1-100 scale (the ANES’ “feeling thermometer”). You could describe their feelings about the other side as neutral. Now both sides’ ratings are down to the mid-twenties on the same scale. You could call that hostility. The downward trend is pretty consistent from 1990-2016, across Clinton, Bush, and Obama.

This second graph is more dramatic still. It shows the percentage of Democrats and Republicans who chose to rate the opposite party at zero on the 0-100 scale. For both sides, that proportion has risen from very low in the late 1970s to about one in four people today. Again, the trend is linear all the way through the Bush and Obama years. More Democrats have rated Republicans at zero than vice-versa.

These graphs show ratings of the parties, not of party members. Ratings of people are better for this purpose, but ANES stopped asking that in the 1980s and now asks only about the parties. However, during the period when they asked the same respondents about both parties and party members, the correlations were high (>.75). Therefore, I think these lines are good proxy measures for how people feel about other people across party lines.

Independents–including leaners–began this period rating both parties above 50 on a 0-100 scale. Their ratings have fallen in parallel, although not as steeply as the partisans’ ratings of each other; and they have viewed Democrats more favorably than Republicans.

*Iyengar, S., Lelkes, Y., Levendusky, M., Malhotra, N. and Westwood, S.J., 2019. The origins and consequences of affective polarization in the United States. Annual Review of Political Science, 22, pp.129-146. See also: promoting democracy and reducing polarization; marginalizing views in a time of polarization; empathy boosts polarization; what is polarization and when is it bad?; civic education in a time of inequality and polarization, etc.

citizens against domination

Leave a reply

This review-essay was recently published and is available for free: Peter Levine, “Citizens against Domination: A Critical Reading of Ian Shapiro,” The Good Society, vol. 28, No. 1-2 (2019), pp. 1-8.

I admire and recommend Shapiro’s book, Politics against Domination, but I use the review as an opportunity to push two positions that I frequently advocate:

The state should not be sharply distinguished from other institutions; it is not uniquely capable of dominating people [or preventing domination]; and
The salient question is not how to design a state to prevent domination–because none of us are really state-designers–but how we should prevent domination by working through the state and the other institutions that we can influence.

The rest of the special issue is valuable. I have been particularly eager to see Brooke Ackerly’s essay, “Rage, Resistance, and Politics against Oppression,” in print. She explores the overlap and the differences between domination–the keyword in modern republican political theory–and oppression, a fundamental term for much of the left, especially for intersectional social movements. That contrast is valuable for anyone to consider.

promoting democracy and reducing polarization

Leave a reply

Some of the threats to democracy in the USA involve bias or injustice in our political system: barriers to registration and voting, gerrymandered districts, the filibuster, money in politics, the Electoral College, and an unrepresentative Senate. With some important exceptions, reforms that address those problems will tend to benefit Democrats and progressive causes.

A particular kind of polarization also threatens democracy. The problem is not disagreement. Our elected representatives should hold diverse political views–maybe more diverse than they do now. Debate and contestation are valuable. It is good to have at least two parties with sharply distinct approaches, so that people can choose between them.

However, our Constitution was written by people who disliked parties and didn’t anticipate them. It is badly designed for partisan polarization. Whenever the branches are controlled by different parties that have homogeneous ideologies, all the incentives favor mutually destructive game-playing.

I can accept the argument that Republicans have played harder and less ethically than Democrats, although I am not certain that’s true, because my prior assumptions and my media stream are biased. But even if it’s true, all the incentives now predict that Democrats will play tit-for-tat.

Besides, affective partisanship–disliking fellow citizens because of their political identities–is the kind of divide that prevents human groups from governing themselves. In order to deliberate, you need not agree, but you must believe that it’s valuable to discuss common issues with the other side. Affective polarization blocks discussion. In the absence of a reasonable level of goodwill, arguments and reasons have no purchase.

The main alternative is simply power: amassing more votes than the other side has. That is preferable to amassing more dollars or more guns than the other side, but it is still raw power. It is government by accident and force, not by reflection and choice, to paraphrase Federalist #1. I suppose if mobilizing more votes consistently generated good outcomes, it might be OK–but it doesn’t.

Democrats and progressives (if they take power) and nonpartisan reformers face a real dilemma. Political reforms will be seen as partisan because they have unequal effects, but partisan polarization is also a problem–and an obstacle to achieving progressive goals.

This is what I would recommend:

First, push political reforms despite their partisan impact if they’re the right things to do. Not only are equity and accountability very high priorities, but our current system actually worsens polarization by giving the parties too many opportunities to pass or block policies without building consensus. I’d vote for expanding the Senate and the Supreme Court, but those reforms are so unlikely that perhaps it’s a mistake to focus much attention on them. More to the point would be voter protection, districting reform, and ending the filibuster.

Second, include political reforms that are bad for Democrats if those are also the right things to do. Redistricting reform meets that criterion in some states. Ranked-choice voting might qualify. Constraining the executive branch is an example, if Biden is president.

An often-overlooked problem involves the dates of municipal elections. Many mayors and city councils are elected in odd-numbered years or in different months than November. Democratic incumbents and municipal employees’ unions benefit because the electorate shrinks to hard-core party and union members. But turnout is often miserably low. I’d move municipal elections to November of even-numbered years and fight Democratic city halls to get that done.

Third, take explicit actions to reduce affective polarization by promoting cross-partisan dialogue and deliberation.

Civic education can help if it focuses on discussions of contested issues.
Private efforts like Braver Angels have a role.
The federal government could organize official public deliberations on controversial matters. The UK Climate Assembly might be a worthy recent model, but the same method could be used to discuss whether to open schools during the pandemic.
After the election, Biden and/or Harris could attend listening sessions with people who might be alienated from them–including right-wing groups but also marginalized people of color. It would be easy to mobilize partisans to distort these sessions. Biden/Harris could turn them into manipulative symbolic events. However, skillful advance work could make them work out better. For example, imagine Biden visiting a conservative evangelical church in the South without prior public notice (maybe he’d coordinate in advance with the pastor) and listening to congregants behind the closed church doors before emerging to make some remarks. If that were done with sincerity, it could make a difference.

The extreme polarization of the news and information sphere may actually help. As Barack Obama learned, it doesn’t really matter how you act; the right-wing news cycle will treat you as an anti-American extremist. Under these circumstances, why not push for genuinely worthwhile reforms, while also making sincere efforts to combat polarization?

overestimating the impact of leaders

Leave a reply

A perennial question is the relative importance of influential decision-makers (“leaders”) versus other factors that can cause social outcomes, such as the structure of institutions, mass opinion and behavior, demographic and ecological change, or sheer accident.

Reporters have a professional bias to overestimate the impact of leaders, because it’s easier to write about individuals than abstractions. Video journalists have the strongest bias, because they must put human beings on screen.

Headhunters and search firms have a similar bias. Like journalists, they are indispensable. I wouldn’t recommend trying to find a nonprofit or college executive without their assistance. But they do tend to overestimate the impact of the kind of people they help to hire. They will tell you, for instance, that Dean So-and-So “raised the ranking” of her college. Any Dean was, at most, only a contributing factor to a change in reputation.

I don’t know much about corporate boards, but I suspect they also overestimate (and hence overpay) top executives. That is partly because their explicit task includes hiring, compensating, and assessing CEOs. It’s partly because they tend to be corporate executives themselves. And it is simply easier for anyone to visualize the impact of a person than an abstraction. But if a board believes that the boss personally doubled the company’s earnings, they are not thinking clearly about causality.

Near the end of War and Peace, Tolstoy offers the opposite view–surely exaggerated, but worth considering as a corrective. He argues that of all the people who fought in the Napoleonic Wars, Napoleon was the least consequential. In a battle, each soldier could decide to stand or run. If most of them stood, Napoleon became a genius. If they broke and ran, he was a defeated fool. His fate was entirely in their hands.

And it was not Napoleon who directed the course of the battle, for none of his orders were executed and during the battle he did not know what was going on before him. So the way in which these people killed one another was not decided by Napoleon’s will but occurred independently of him, in accord with the will of hundreds of thousands of people who took part in the common action. It only seemed to Napoleon that it all took place by his will.

At a larger scale, Tolstoy says, the Napoleonic Wars were the result of an underlying historical current that caused great masses of Europeans to spill beyond their national borders. Napoleon was carried by this current all the way to Moscow. In subsequent decades, Europeans would instead fight their own countrymen in revolutions (and, although Tolstoy doesn’t say so, would conquer other continents). Napoleon was not a cause but an unwitting product of his time.

In her classic work The Thirty Years War, C.V. (Veronica) Wedgewood asks why so many people in the 1600s were so interested in dynastic politics: royal marriages, successions, and usurpations. Commoners were willing to die to ensure that one family prevailed over another. She says the reason was “the faulty transmission of news” and poor “diffusion of knowledge.” People just didn’t know about aspects of politics apart from royal persons. “The public acts and private character of individual statesmen thus assumed disproportionate significance, and dynastic ambitious governed the diplomatic relations of Europe.”

In those days, people would naturally explain important social developments as the consequence of leaders’ actions. For instance, many would have said that south-central Europe was consolidating because of the Hapsburgs’ fortunate marriages or that England had remained Protestant because James VI of Scotland (a Protestant) had inherited the throne. But these were myopic explanations. Much more likely, the Hapsburgs married well because south-central Europe was consolidating, and Jacobean England was Protestant because of the strength of reform movements in northern Europe.

We should be able to fix this problem today by gathering more information and analyzing it better. However, we now have the opposite problem: too much data. So much information is available that we cannot process it, and one common response is to return to understanding the world in terms of the behavior of a few famous or infamous individuals. To the list of people who overestimate the impact of leaders, I would add: most of us voters.