Should St. Louis Vote 'Strategically' on March 2? No One Knows

click to enlarge The four candidates for St. Louis mayor. From left: Board of Aldermen President Lewis Reed, Treasurer Tishaura Jones, Alderwoman Cara Spencer, Andrew Jones. - COURTESY OF CANDIDATES' CAMPAIGNS
The four candidates for St. Louis mayor. From left: Board of Aldermen President Lewis Reed, Treasurer Tishaura Jones, Alderwoman Cara Spencer, Andrew Jones.

On March 2, for the first time in the history of St. Louis elections, voters will be picking more than just their chosen candidate for mayor. A new variable has entered the mix in the form of a brief instruction at the top of the ballot: "Vote for AS MANY names as you approve of in each race."

There's no catch there, and "as many" means just that: Voters in Tuesday's dramatic primary can select from mayoral candidates Lewis Reed, Tishaura Jones, Cara Spencer, and Andrew Jones, in any combination they prefer, and without any partisan labels attached.

But with the election less than 24 hours away, the St. Louis debut of "approval voting" is being closely watched by election researchers and the campaigns themselves. No one knows precisely how voters will behave under the new system, and there's no simple answer to whether there is any strategic advantage in picking one, two, three or even four candidates vying for mayor.

It all comes down to personal preference. Still, Aaron Hamlin, the executive director of the nonprofit Center for Election Science, says there is a strong argument behind "approving" more than one candidate in the election's primary stage.

"By choosing multiple people, you increase the odds of one of the people that you really like moving on to that next round," he notes. Hamlin suggests voters settle on their top preference, but then select at least a second candidate "that they would be satisfied with to run the city."

Of course, this means a voter would have to view more than one candidate as satisfactory, or at least partially so; still, that doesn't seem so outlandish in St. Louis' current mayoral contest, which features substantial policy overlaps and multiple candidates who broadly agree on issues like police accountability and strengthening the school system.

Voting for all four mayoral candidates is allowed in the primary, though difficult to justify unless a voter would be equally happy with any of the candidates.

On the other side of the spectrum, people could stick with what they know: voting for a single candidate.

"People could still vote for one, they have that option, and that can still be a very honest expression," Hamlin acknowledges. Still, he argues that picking a single candidate on the March 2 primary "creates a lot of risk" for a voter's eventual choices on the April 6 ballot, since it will feature a matchup of the primary's top two vote-getters.

Put another way: By not making at least a second selection in the primary, a voter is giving up a chance to influence both sides of the April 6 matchup that will decide St. Louis's next mayor.

"Someone could say, 'I really like this person, and I'm not going to be satisfied with anyone else,'" Hamlin adds. "They would have to realize, 'Okay, well, I'm taking some serious risk here. It could mean that I opened up the door for someone else I don't like.'"

While the potential gamesmanship in approval voting makes for compelling commentary, Hamlin and the Center for Election Science are eager to add real data to their behavioral theories. The center, which describes itself as a nonpartisan research group "dedicated to empowering voters with voting methods that strengthen democracy," has spent the past several years advocating for approval voting to be attempted in an American city.

In July 2020, that city became Fargo, North Dakota. The result: A municipal election yielding 18,800 ballots, with both the top-two candidates coming away with more than 50 percent of approvals.

But Hamlin says that it's difficult to draw conclusions between Fargo and St. Louis. For one thing, Fargo's election system is significantly different from the one rolling out tomorrow in St. Louis: Fargo runs a single-stage municipal election that, even before the arrival of "approval voting," already featured a "top two" system that instructed voters to pick two candidates to fill open positions on the city commission.

With approval voting in place, Fargo's voters approved an average of 2.3 candidates per ballot — suggesting that voters didn't stray far from their previous election-time behavior of picking two candidates per ballot.

Even with such a limited sample, Hamlin points to some notable positives: In 2018, neither of the two winners of Fargo's election for city commission broke 18 percent of the vote, frustrating voters and leading activists to campaign for a reform of the city's voting system.

A similar path was traced in St. Louis in 2017, when a seven-candidate mayoral primary ended with the winner, Lyda Krewson, taking just 32 percent of the vote. The election sparked internal recriminations among the city's progressives and dissatisfaction among voters at the lack of a meaningful general election. (Though separate from local efforts, the Center for Election Science awarded grants to the campaigns behind the ballot initiatives both in St. Louis and Fargo.)

It hardly needs to be said, but St. Louis is a much different city and electorate than Fargo. If this were a national or state election, this would be the part of the story where we'd dive into local polling data — but there's very little water in that statistical pool. What we do have is a January 11 poll from Show Me Victories, which, among other things, asked 732 registered city voters, "Do you intend to vote for ONE candidate or multiple candidates for mayor?"

More than half the respondents, 59 percent, answered they would select just one candidate, with the remainder evenly split between "multiple candidates" and "unsure or do not know."

In that poll, voters were most closely split on their "number one choice" for mayor, with 30 percent saying they'd pick Board of Alderman President Lewis Reed, and 28 percent answering with Treasurer Tishaura Jones. However, the poll did not seek out answers about voters' possible second or third choices.

A second poll, conducted in early February by the Remington Research Group, asked 501 city voters whether each candidate "will receive one of your votes" for mayor. Again, the results showed Reed in the lead with 59 percent, and Tishaura Jones close behind at 51.

"This specific election is a big unknown," says Betsy Sinclair, a professor of political science and election researcher at Washington University who studies how different election systems are being steadily embraced by states.

As a researcher, Sinclair says she's eager to see what the March 2 data says about voting in a pandemic and the simultaneous roll-out of approval voting. "Will they feel like it was confusing or frustrating?" she wonders, rattling off questions waiting for answers. "What was the volume of approval voting? Did voters choose more than one candidate? Are voters making choices around ideology?"

All that will soon be clearer, but the results could also reveal the degree to which voters attempt to strategize their picks. Critics of approval voting, including former mayoral candidate and current St. Louis Post-Dispatch columnist Antonio French, suggest scenarios where a voter picks against their own beliefs by approving two candidates: the one they most approve of and the one they feel is the weakest opponent for that preferred pick to face in the runoff.

This strategy, known as "raiding" or "crossover voting," is something rarely seen in existing blanket primary systems in California, Sinclair says.

"My impression is the reason that people are voting in the first place has to do with the fact that they feel like they're giving voice to some deep preference they have about the way they want the world to be," she argues. "They're being sincere. They're not trying to game the system."

For researchers and election advocates, then, the election will begin to clarify the cloud of unknowns around approval voting, from the number of candidate selections to the effect of the pandemic on turnout.

But while the system has changed, the people and issues at stake remain just as critical as they ever were. The most pressing unknown of the 2021 election isn't about wonky voter strategies or hypotheticals within hypotheticals.

"I would like to know how voters are going to feel in terms of the legitimacy of the election after the election," she says. "I think that's a really interesting question. It doesn't have so much to do with the structure or the ability of this kind of system to moderate outcomes. It has to do with how voters feel."

Follow Danny Wicentowski on Twitter at
@D_Towski. E-mail the author at [email protected]
  • Sign up for our weekly newsletters to get the latest on the news, things to do and places to eat delivered right to your inbox.
  • Follow us on Facebook, Twitter and Instagram.