File this in the Department of Things I Did No Know:

All the contenders will face off one nonpartisan ballot this fall, and a candidate needs to win a plurality of the vote in at least five of the nine City Council districts in order to win the contest outright. This means that, just like in a presidential election, it’s very possible for a candidate to win the mayor’s office while coming in second (or potentially even further back) in the popular vote.

If no one wins outright, then the two candidates with the most votes citywide would compete in a runoff six weeks later. However, the winner still isn’t the candidate with the most votes, it’s the candidate who wins a majority of the Council seats. If no one manages to pull this off (ie, if there’s a tie that prevents anyone from winning at least five districts), only then would the popular vote determine the winner.

This system was first put in place for the 2004 mayoral race, a contest that former Democratic Gov. Douglas Wilder won overwhelmingly, and so far, the candidate with the most votes has always won outright. In 2008, Del. Dwight Clinton Jones won 39-34 and carried five of the nine Council seats, and he was re-elected in a landslide four years later. In 2016, Stoney himself won the popular vote by a narrow 36-34, but he also avoided a second round of voting by carrying a majority of the Council districts.

So, how did this system come to be? While Richmond only started voting this way in 2004, Venugopal Katta explained in a 2017 piece for William & Mary Law School’s Election Law Society that the reasoning behind it goes back decades further. Back in 1969, the predominantly white City Council, whose members were elected citywide at the time, approved a plan to annex part of neighboring Chesterfield County. This move lowered Richmond’s black population from 52% to just 42%, and critics argued it was done to strengthen white voters at the expense of African Americans.

However, as Katta wrote, the plan ran into trouble two years later when the U.S. Supreme Court ruled that the Voting Rights Act required any changes in city lines to be approved either by the U.S. attorney general or the D.C. District Court. Richmond had completed the annexation by this time, but the attorney general’s office refused to approve it. The city finally came up with a compromise where it would elect city councilors by district, and it agreed that five of the nine seats would have black majorities.

The Supreme Court approved this new plan in 1975’s City of Richmond v. United States decision, and for decades, the City Council remained the major force in Richmond politics. The body continued to pick the mayor from among its members (now-Sen. Tim Kaine became mayor this way in 1998), but a series of corruption scandals involving councilmembers led to calls for a strong and independent chief executive.

In 2002, former Gov. Douglas Wilder and former Mayor Tom Bliley called for electing the mayor citywide. However, while Richmond was again majority black by this point, local African American leaders feared that wealthier and better organized white voters would have a greater say over who led the city. In order to ensure that a mayor couldn’t win without substantial black support, the Wilder-Bliley Commission’s plan required mayoral candidates to win five of the nine districts in order to be elected. This plan overwhelmingly passed in a 2003 referendum, and it remains in place today.