Evidence continues to mount that shows former U.S. Senator Bill Nelson’s race for re-election in 2018 was hampered by a ballot design quirk.

Florida has already enacted a law that would standardize ballots to avoid a repeat of what happened in part of heavily-Democratic Broward County: a ballot design that made the Senate race “easy to overlook.”

It was immediately clear after the Nov. 6 election that far more voters than expected were leaving the Senate box blank. Many of those voters supported other Democrats, like the losing candidate for Governor, Andrew Gillum.

In a new academic paper presented Thursday at the Election Sciences, Reform, & Administration Conference, two researchers looked beyond vote totals, drilling down to the individual ballot level.

What they found was this: In Broward County, voters who skipped the Senate race likely did so by accident, rather than purposely avoiding the race.

Broward County ballots put voting instructions in the first column, rather than stripped across the top. That design pushed the Senate race far down the page, isolating it from other marquee contests.

Broward County's ballot, on the left, isolated the U.S. Senate race. By comparison, Orange County's ballot had the race grouped with others.
Broward County's ballot, on the left, isolated the U.S. Senate race. By comparison, Orange County's ballot had the race grouped with others.

Shortly after the election, several voters in the part of the county overlapping with Florida’s 24th Congressional District told the Times their ballots were “confusing,” and many reported nearly missing the Senate box entirely.

Generally, voters who choose one major-party candidate vote for others in the same party as well. On the flip side of the coin, voters who decline to vote for a Democrat or Republican in one race (either by picking a third-party or write-in choice, or by leaving it blank) are more likely to do so in other races on their ballot.

The academic paper’s authors, Marc Meredith at the University of Pennsylvania and Michael Morse at Harvard University, looked at individual “undervoters” — people who voted, but not for senator.

According to the research, “about 33 percent of Senate undervoters in other counties also voted for a third-party or undervoted in the gubernatorial race, consistent with a pattern of more intentional undervoting.”

But that wasn’t the case in Broward. There, “only about 7 percent of Broward voters who undervoted in the Senate race also voted for a third-party or undervoted again in the gubernatorial race.” That indicates they skipped the race unintentionally.

But there’s no definitive evidence that the quirk mattered enough to flip the race.

Even if everyone in Broward who voted Democrat for Governor also cast votes for Nelson (and the opposite did for Scott), Nelson would have picked up 9,658 votes, and lost by 375.

Another paper presented at the same conference tried to estimate undervotes by comparing the Senate race to all four statewide races (including those for Attorney General, Commissioner of Agriculture and Chief Financial Officer). Some of those would have flipped the race, but some wouldn’t, leaving the authors unable “to determine whether Broward County’s ballot was pivotal or not.”