By Robert Miller
In 1986 Sam LeBlanc, a New Orleans progressive and eight-year state legislator, ran for mayor, and I came out of political retirement to help him. Before a speech I had written, he told me it needed a point. It was a list, not a commitment.
I asked him why he was running.
He said: “pretty good government.”
This guide holds candidates to that standard the way a gaming regulator holds a license applicant: Will they follow the rules of the house? I scored 15 candidates on five questions: litmus tests, pledges, experience, working with others and compliance. The scores, the method and the stories behind them are in the companion piece, “Why DO THIS.” But first, here is what to do.
Our slate
District 3: Kellie Torres and Steve Novick.
District 4: Olivia Clark, Eric Zimmerman, and Eli Arnold.
Novick and Torres, Clark and Zimmerman are the top two scorers in their districts. Arnold is on the slate because he can win: in 2024 he drew 8,123 first choices, more than any other District 4 challenger, and lost the last seat on transfers.
What you do
In District 3, Novick will draw first choices on his own. Ranking Torres first is how she wins a seat alongside him. We stop at four. Rank more only if you would rather see one of the others win than none of them.
In District 4, the birthday rule splits first choices evenly between Clark and Zimmerman, so neither is knocked out early. Arnold is third on every card. Which order you use for the first two matters less than ranking all three.
The order on your ballot matters.
Your ballot isn’t one vote. It’s a list of backups. If your first choice wins easily, or has no chance, your vote moves to your second choice. Then your third. It keeps moving until it helps someone win. But it can only move to people you ranked. When it runs out of names, it stops counting.
That happened to thousands of voters in 2024. When Eli Arnold was eliminated in District 4, his 12,900 votes moved to each voter’s next choice. About 6,459 went to Eric Zimmerman and elected him. The other 6,440 went nowhere. Those voters hadn’t ranked anyone still in the race.
By the end of the count, about 19,175 votes’ worth of District 4 ballots had stopped counting. A seat takes 19,180. Almost a full seat’s worth of voters ran out of choices before the count was done.
A card can’t create votes that aren’t there.
Seats come from first choices. What a card does is make sure no vote is wasted:
• Finish your list. Don’t let your ballot drop out before the last seat is decided.
• Rank the whole slate, not just your favorite. In District 3, only 22 percent of Novick voters ranked the next-strongest candidate in their lane.
• Put pretty good candidates first. That’s the whole game. If you like a candidate on this list, rank them first and tell your neighbors to.
Sam’s speech needed a point, because a list is not a commitment. Your ballot is a list. Ranking it with care is the commitment.
Every judgment and every score is mine, and so is every mistake. Every score, the method behind it, and the sources: [Why Do This]
About the author
I am president of a gaming technology company, and I have spent five decades in the gaming industry. I may bring business proposals to the city in the future. I have no financial relationship with any candidate in this guide or with the mayor’s office. I have publicly called on Mayor Keith Wilson to build an election machine; I hold no role in one. I have discussed the District 4 race with people who support some of these candidates. I used AI tools, Claude and ChatGPT, to gather public records, run the numbers and check facts, including the recount of the 2024 ballots.



Who is the author of this article?
I’m not convinced by the birthday rule. I prefer the “Would I hire this person?” rule.
Suppose I’m hiring for a position and 12 people apply. After reviewing their résumés and talking briefly with each, I decide there are three people I would be willing to hire. I wouldn’t hire the other nine.
So I rank the three in my actual order of preference:
1. Clark
2. Zimmerman
3. Arnold
Then I stop. I don’t rank someone I wouldn’t be willing to hire.
If I wanted to make my ranking more strategic, I might put Evanstar #4 and Jeremy Beausoleil Smith #5. I’d call those defensive rankings: not affirmative choices, but backups I still prefer to an unranked candidate if my higher-ranked choices can no longer use that portion of my vote.