This piece was originally published in the New York Times on May 26, 2020.

Do vice-presidential picks matter? Conventional wisdom argues they have limited electoral impact.

But a vice-presidential pick does matter in a particular way to elections. It suggests the strategy and tactics a campaign will pursue. The pick might complement the ticket, like Al Gore did for Bill Clinton in 1992, or the pick might balance the ticket, like Mike Pence did for Donald Trump.

So Joe Biden’s electoral fate may well hinge on this decision. In our polarized era, where turnout determines election victors and each party’s coalition has become more locked in, ticket-balancing picks for vice president can be helpful in mending primary wounds and generating excitement for the coalition in the general election.

That is why Mr. Biden should select for his running mate a ticket balancer.

Now, the temptation for Mr. Biden to pick a ticket “complementer” will be high. All the conventional wisdom suggests that ticket complementers “do no harm” because they are, essentially, prototypes of the presidential nominee.

By contrast, ticket balancers offer voters something the main nominee lacks and often are meant to motivate a group within the coalition with which the nominee has struggled to gain traction. Balancers are perceived to be riskier, especially since John McCain’s selection of Sarah Palin in 2008.

Hillary Clinton is often castigated as running a terrible, horrible, no good, doomed campaign. She actually ran a perfectly fine campaign — but strategically speaking, it was the wrong kind of campaign. It was based on the flawed assumption that a significant portion of American conservatives would not, simply could not, vote for Donald Trump.

But on Election Day, 90 percent of Republicans voted for him; Mrs. Clinton also failed to carry independents, despite a campaign structured mostly on winning them over.

Mrs. Clinton’s “do no harm” pick — Senator Tim Kaine of Virginia — ended up doing her considerable harm. The strategy built around the pick ended up leaving the party’s progressive flank vulnerable to, among other things, a sophisticated Russian propaganda and disinformation campaign.

In addition, so-called protest balloting in 2016 was three to five times higher than normal in the swing states. In states like Wisconsin, which was decided by less than a percentage point, nearly 6 percent of the electorate cast protest ballots. For all the attention placed on white, working-class voters and their continued realignment away from Democrats, protest balloting affected the outcome of every swing state contest and played a pivotal role in Mr. Trump’s destruction of the Democrat’s Midwest “blue wall.”

Yet Mr. Trump still has a plurality problem, and his campaign manager, Brad Parscale, knows it. Even before the pandemic came along to destroy his best argument for re-election, the economy, Mr. Trump was unpopular among independents and has hardly ever (aside from a brief moment early in the coronavirus crisis) hit 50 percent approval rating nationally and rarely in swing states.

The only way to re-elect a plurality president is to make a plurality vote share sufficient enough to win. And the best way to do that is to replicate Russia’s playbook of targeting parts of the Democratic coalition — like progressives and young black voters — to turn them against voting for Mr. Biden.