We asked a mathematician how to best use your vote.

Bill Adler / Flickr

On November 6, D.C. voters will have the chance to vote for up to two candidates for at-large councilmember.

The ballot has six options. There are the two incumbents—Democrat Anita Bonds and independent incumbent Elissa Silverman—as well as challengers in the form of independent Dionne Reeder, independent Rustin Lewis, State Party Green candidate David Schwartzman, and Republican Ralph Chittams. Of a voter’s two options, only one can be from the majority party (aka a Democrat), but if they so choose, voters don’t have to elect any Democrat at all.

While common wisdom says that Bonds will likely sail to reelection, the contest for that second seat has become the marquee race of an otherwise snoozy cycle. Mayor Muriel Bowser has decided to boost Reeder over Silverman, giving the Ward 8 business owner a fundraising edge over her opponent. While a poll commissioned by the Silverman campaign at the end of September found both incumbents with a comfortable lead over the rest of the candidates, that was before Bowser hosted a get-out-the-vote rally with hundreds of her supporters alongside Reeder and Bonds.

At a recent debate, an audience member asked the candidates (aside from Bonds, who wasn’t there) who their second vote was going to. Here’s how they responded:

But this got us thinking: if they really want to win, should the candidates really be voting for two people? And if you bleed Reeder, or you’re a huge Silverman fan, or, for you it’s Chittams or high water, should you? Or would it be more strategic to be a one-and-done voter? (The technical term is bullet voting.)

We chatted with Daniel Ullman, a professor of mathematics at George Washington University and the co-author of The Mathematics of Politics, to get some insight.

Ullman says that this decision is an example of game theory at work: “You have to make a decision and there are other people that have to make decisions too, and what you do depends on what they do and what they do depends on what you do.”

If this sounds complicated, that’s because it is. “Game theory is very, very difficult with even three players,” he says. “So with 200-300,000 players in this game, it’s very hard to know.” During D.C.’s 2014 general election, the last midterm cycle, there were a total of 177,377 ballots cast.

The best way to proceed is to learn as much as you can about the plans of your opponents (in game theory, that means everyone else who is playing). “After I read the polls, I will know where my vote is most needed,” Ullman says. “But of course, everyone else is reading the polls, too.” Plus, he adds, they aren’t always accurate.

Ullman says that bullet voting is “a sensible strategy if it’s a close race and you want one particular candidate to win. Then, you might be hesitant to vote for another candidate because they might pass yours.”

A person might also opt not to vote for a clear favorite (in this race, that’d be Bonds), “if you have two of the others that you like a lot.” Of course, if everyone decided to do that, then that candidate would lose. That’s game theory for you!

So ultimately, how should voters play this game? Ullman demurs. “That’s a question of values, not mathematics.”

Filed Under: at-large race, dcision2018,