Imagine you’re living in a very conservative town with a small liberal minority.
This election, you have three candidates to pick from: Angela Adams (a conservative), Barbara Baker (also conservative) and Charles Connor (an extreme liberal).
As a conservative yourself, you slightly prefer Adams, but Baker is your second choice. Both are preferable to Connor.
On Election Day, Adams gets 32 percent of the vote and Baker gets 33 percent. Surprisingly, Connor wins the race — with just 35 percent.
Now, your town has a mayor that 65 percent of voters actively didn’t want.
Does this sound like democracy functioning properly?
Sure, the person with the most votes won. But in this not-so-implausible scenario, the outcome has far more to do with a fluke technicality — i.e., the “spoiler effect” — than the will of the governed.
In fact, a spoiled race is a very natural outcome of a competitive election involving three-plus candidates.
That’s because in New York, we use the “first-past-the-post” (FPTP) system, where all you need to do to win is get the most votes — even if it’s not a majority.
On its face, “most votes wins” sounds reasonable.
But as our hypothetical demonstrates, FPTP breaks down in elections involving more than two candidates: In our town, the far-left candidate didn’t win on his own merits — he won just because the conservative vote was split.
Our scenario has all three candidates running roughly neck-and-neck.
But an election can also be dramatically spoiled by a very minor candidate.
As one of many examples, many Democrats blame Ralph Nader for siphoning off just enough liberal votes in Florida to cost Al Gore the 2000 election.
The recent real-life primary in New York’s 19th Congressional District took our town election’s conundrum and made it more complex — seven people were running for the Democratic nomination.
The winner, Antonio Delgado, only got 22.0 percent of the vote; the two runners-up won 17.9 and 17.7 percent, respectively.
Clearly, just tallying up the votes can’t give you any sense of who NY-19 Democrats on the whole wanted most – it just meant that Delgado happened to have a larger base of core supporters.
Hypothetically, how can we know that the Delgado wasn’t the last choice of the 78 percent who voted for other candidates?
I’m definitely not trying to delegitimize Delgado’s win. This issue goes beyond NY-19.
Can we really say that anyone who wins an election without a majority has the consent of the electorate?
As it stands today, it’s literally impossible to know.
We just accept that FTTP is the way democracy has to work. But there are actually several other ways to determine election winners.
One way is having runoffs, which is actually how many states and localities hold their primaries.
Unfortunately, the first round in a runoff election still suffers from the same spoiler effects as in regular elections. Runoffs also require two or more election dates — not a good way to increase voter participation.
Fortunately, there is a way to fix these problems and eliminate spoiler effects.
The solution is a system called instant runoff voting (IRV), and it’s already used in U.S. cities like Minneapolis and San Francisco, the state of Maine and even countries like Australia.
IRV is based on the simple principle that in order to win an election, you have to have get a majority of votes – not just a plurality. As we’ve seen, that doesn’t always happen when you just poll voters’ first choices.
What IRV does is it automatically runs multiple runoff election rounds at the same time — all while using one ballot per voter. Instead of filling in ovals as is done now, voters rank-order their preferred candidates on an IRV ballot.
At the end of each election, votes are tallied up as usual. If a candidate has a majority, they win.
But if no candidate has a majority, the candidate who received the fewest ballots is eliminated. The votes that person won are automatically transferred to the other candidates according to each ballot’s next-ranked choice.
This process repeats until one candidate has a majority of votes — and emerges victorious.
For example, in our imaginary town, you might rank your choices: No. 1 for Adams, No. 2 for Baker and No. 3 for Connor.
On Election Day, none of the candidates get a majority in the first round.
Recall that Adams got the fewest raw votes, which means she would be knocked out of the running and her votes distributed accordingly. On the IRV ballots, 90 percent of her supporters have Baker as their second choice, and 10 percent have Connor.
So at the conclusion of Round 2, Connor would get 38.2 percent of votes, and Baker would get 61.8 percent – and win the election with a broader mandate.
We’ll obviously be using the FPTP system for our real-life Nov. 6 election and the Sept. 13 state and local primaries. And as usual, after the votes are counted, you can bet that there will be races where the winners fail to get a majority of votes.
Without IRV, this will almost certainly happen in the Democratic nomination for state attorney general, where four candidates are running.
Of course, IRV isn’t flawless. Any electoral system involves subjective tradeoffs.
But I think in order for an election to be truly democratic, the victor needs to have the consent of the majority — not just the largest bunch of immediate supporters.
In a sense, IRV requires winners to be the best at bringing in a majority of the public. To do this, many candidates would have to appeal to voters beyond their immediate base — including third-party voters — to register them as their second choice.
On the whole, IRV is straightforward, automatic, completely opt-in and, frankly, elegant.
Looking forward, New York’s local communities should try it out. And if it works, New York should strongly consider using it across the entire state.
Steve Keller is a regular contributor to the Sunday Opinion section.