All these voting posts appear to argue for the prewritten bottom line: "voting is rational". Of course, sometimes it is and sometimes it's not. Voting is like cooperating in PD, which is a good approach in some situations, but certainly not in all.
Can anyone explain to me how the "1 in 10 million chance" makes sense? I always thought the probability of a tie vote has to be based on the binomial distribution. If we take New Mexico voter population of 1,254,567, assume that the "actual" preference rate is something near even like 51%, then the probability of a tie is 7.11e-113. In general it will be astronomically low, because the binomial distribution is very narrow for large population sizes, like a whole state. Anywhere outside of the expected value, its density will be insignificant. One would need a background preference rate of nearly even, like 50.17% to get to the 1 in 10 million chance of a tie vote territory.
The linked paper said they looked at the marginal vote differences among 10,000 simulations, took the mean and standard deviation, and calculated the probability of marginal vote=0 using Student-t distribution. Didn't they use the poll results to drive the simulations? Does that mean that New Mexico's poll was exactly 50%? If voter preferences, presumably based on every individual's analysis of which choice is best for the nation as a whole, are really so break-even, then how can we as a nation then claim that any one candidate is preferrable to any other at all?
If voter preferences, presumably based on every individual's analysis of which choice is best for the nation as a whole, are really so break-even, then how can we as a nation then claim that any one candidate is preferrable to any other at all?
The general assumption seems to be that one candidate is an evil mutant, and that his supporters are either deluded or also evil mutants. Which is which depends on who you ask.
The $1555 only pays out if you vote for the right person. If you vote for the wrong person, you could end up doing $1555 worth of damage.
Here, declining to vote is analogous to overfishing -- if too many people do it, we get a tragedy of the commons.
That's the opposite of the truth. If everyone voted, the country would waste more resources on voting.
The title strikes me as strange. "Please go vote" seems like you're asking a favor, and then you justify it by saying "voting is rational", which means it shouldn't count as a favor, I should want to do it of my own accord. Unless by "rational" you mean "in keeping with the ethics of our tribe", in which case you would just be exhorting me to follow the ethics of the tribe. But that's a bad way to use "rational".
EDIT: Perhaps more likely, you're saying that voting is not a violation of the ethics of our tribe, which then justifies you asking the favor (since it would be inappropriate to ask an unethical favor).
Main Idea: Voting is rational because (1) the benefits of having your preferred candidate win might be enough to outweigh the chance you won't swing the election, (2) there are intangible benefits to voting besides swining the election, and (3) the collective action of nonvoters matters and could have swung elections.
~
Currently, about 58% of the voting age population actually vote, which in 2008 was 131 million people out of a possible 225 million people. What this means is that there were 94 million people who didn't vote in the last election despite being able to, and we can expect this election to be the same.
Is this a problem? Perhaps not. While compulsory voting has an interesting appeal, I generally think people should be free to decline to participate for whatever reason if they want to. Indeed, given people's overconfidence and under-informed approach to issues (for which I don't blame them and for which I am also at fault), many people may have a moral imperative to not vote.
All this aside, I want to tackle head on the pervailing view I've been seeing among many people, including myself, that "voting is irrational" because your vote doesn't really "count". As smart person Steven Levitt says:
Well, one good indicator of a person who’s not so smart is if they vote in a presidential election because they think their vote might actually decide which candidate wins... there has never been and there never will be a vote cast in a presidential election that could possibly be decisive.
So this leads me to wonder is voting rational? Are there good reasons to vote? In this essay, I want to reverse the argument I used to buy, change my mind, and answer with a resounding yes. Voting is rational, there are good reasons to vote, and you really should go out and vote if you can.
In the paper "What is the Probability Your Vote Will Make a Difference?" [PDF], authors Andrew Gelman, Nate Silver, and Aaron Eldin concluded that if you're voting in New Mexico, New Hampshire, Virginia, or Colorado, your vote would have about a 1 in 10 million chance of being the deciding vote in the election. The average state has a chance of 1 in 1 billion. For ll states except DC, your chance will have an upper bound of 1 in 100 billion. For DC voters, its closer to 1 in 1 trillion. (And it all ends up being the same when you take into account recounts, litigation, etc. [PDF])
Using these odds, we can think of it like a lottery ticket and analyze the expected utility. By a strict view of rationality, rational decisions are those that are expected to maximize utility, which is calculated by multiplying the utility derived by the situation multiplied by the chance of that situation coming into existence.
A win for your preferred candidate will affect the entire US population of 311 million people. As Andrew Gelman argues on his blog, if you think that your preferred candidate's policies are better enough than the other guys to be worth an additional $50 per person in improvement of life, which seems very plausible given the wide differences in economic platforms, that's a swing of +$15.55 billion. That makes your vote worth somewhere between $1555 for NM/NH/VI/CO, $15.55 for the average state, and a mere penny for the DC voter. This payoff matters to you if you care about the welfare of the US, and thus is best modelled as a social benefit (PDF).
On a narrow view of the costs and benefits of voting, one would compare the expected social benefit expressed in monetary return (poor DC!) against the estimated cost of voting, and that would be that. However, that's not all there is to voting. Indeed, there are many additional benefits, beyond an expectation of impacting the election.
As Adam Lee points out, voting might be a social good merely by continuing to express support for, legitimize, and perpetuate democracy as a broad institution. Democracy must be renewed in order to exist and the more people who vote, the more representative and legitimate the system is. Lee further goes on to argue that by adding your vote, you're increasing your chosen candidate's mandate by that slim amount, and doing your part in ensuring that the government stops playing solely to the extreme fringes that do vote.
There are other social "intangible" reasons to vote as well. For many, including me, it feels good to be engaged and consider yourself a part of the process. Going to your polling place, at least as a college student, is lots of fun. And working to vote informed keeps myself motivated to be educated, in a way that the more abstract land of the classroom often does not. And as a political science / psychology student, it gives a fascinating realm of human behavior to study.
Voting is also a grand opportunity to signal to yourself and others what kind of person you are. Like it or not, we define ourselves by how we vote (or choose not to vote). Whether you take up the mantle of passionate progressivism, skeptical conservatism, somewhere in between, or on some other political axis altogether, you get to stake yourself on a political identity (or a lack of one).
From Gelman, Silver, and Eldin's model, it's clear your vote always has a chance of counting, even if that chance is a mere 1 in 1 trillion. But your vote matters in another critical sense -- imagine that you decided to not vote because you think your vote won't matter. Now imagine that everyone else also had that same reasoning -- everyone else also agreed voting is irrational. All of a sudden, no-one turns out to the polling place, and the country falls apart into chaos.
Certainly your choice has no causal impact on the choices of others -- the entire country isn't going to stop voting just because you choose not to. But what we really have here is a classic collective action problem. Here's an example:
The fishermen can either choose to fish normally or overfish. If all the fishermen overfish, they stand to deplete the lake and all fishermen lose their jobs. However, if just a few fishermen overfish, they get the benefit of added fish to sell, and the lake can handle the slight increase in load. So this tension is to be the fisherman that wins most by personally overfishing, while not collectively depleting the entire lake. Such problems are called collective action problems — people do well individually by defecting but do worse collectively if everyone defects. The result of a collective action problem ending in disaster is called the tragedy of the commons.
Here, declining to vote is analogous to overfishing -- if too many people do it, we get a tragedy of the commons. But if enough people vote, you can "free ride" on their work by not voting yourself. And by refraining from voting, you hurt the group just that little bit in a collective action sense.
On the flipside, no matter where you vote, you're vote does matter in the same collective action sense. Sure, you may be a voter in DC either redundantly supplementing or trying hopelessly fight the dominant liberal tide. But either way, your vote is there, and it has a marginal effect on the election that could affect the outcome.
For a good example about how things could be different in a collective action sense, imagine what would be different if all the nonvoters decided to vote instead. We know that there are going to be approximately 90 million non-voters. According to Pew Research, these undecided voters, if they were to vote, would add another 53 million votes for Obama and another 21.6 million votes for Romney. Going off of the USA Today - Suffolk survey, we'd get +38.7M for Obama and +18M for Romney.
If everyone voted, that would put Obama ahead by somewhere between 31.6M and 20.7M votes. If we go with Nate Silver's predictions (as of Nov 3) for a 50.6% Obama / 48.3% Romney split, that would project that a pool of 131 million people voters would break 66.3M for Obama and 63.3M for Romney. Add in the undecided pool, and it becomes 97.9M vs. 84M -- a total trounce rather than a "close-ish" match. While it may not change the outcome in Obama/Romney, given that nonvoters are characteristically more liberal, this could have made a huge changed-the-outcome difference in Bush/Gore or Bush/Kerry.
So hopefully between (a) the expectation of voting as a lottery ticket with actually high expected value, (b) the additional "intangible" benefits of voting, and (c) the collective action approach, voting is not <i>intrinsically</i> irrational. However, we haven't really taken into account the costs of voting.
If people are voting out of a desire to help others, they might feel insufficiently confident to vote in a way that will actually help others. That sounds like a fair worry to me, though I'd suggest that if they even have the slightest hunch one way or another, it might be worth casting your vote just to balance out the batshit crazy people.
Furthermore, they might legitimately have something better to do that's likely to impact the world far more than their vote. While the new availability of absentee voting would thus require you to be so busy with high-value activities that you just can't have time to fill it out, though, I suppose this is possible.
Or maybe voting comes to you at an unexpectedly high hardship. Here, I think there are good resources to help you, but that's also understandable.
But at the end of the day, voting is not only potentially and legitimately high reward, it's also surprisingly low cost for many people. Based on the money approach alone, you're spending like two hours max for a potential $1555 if you're in a solid swing state. That's damn good pay.
Thus, in conclusion, I ask you to please find some way to be informed on the issues and strongly consider voting for whoever your preferred candidate is. It's truly the rational thing to do.
(Edit: I didn't realize this would be one of some dozen essays on voting; I wrote it independently of seeing all the others. Sorry if there is any redundancy.)