Oakland School Board Election in Chaos as Different Winner is Declared After Results Are Certified
“The entire race has now been thrown up in the air, as the new results go against the certified results”
A school board race in Oakland, California, is a total mess due to an error in the confusing process of ranked-choice voting.
The person everyone thought came in third ended up with the most votes. But the declared winner refuses to concede, and the election has already been certified.
This means that it would take a judge to overturn the results. What a nightmare.
A reader named Spencer Klein, who maintains a blog called Neutrino Hunting in Antarctica, wrote about it:
The complete debacle in the recent (Nov. 8th, 2022) election for the Oakland School Board (OSB) led me to think more about elections. For those of you who are from outside the San Francisco Bay Area, after tabulating and announcing the election results, the Alameda County registrar found an error in how ballots were counting, and On Dec. 28th announced a new top vote-getter, less than two weeks before inauguration day.
Oakland is a city of about 440,000 people just south of Berkeley (where I live). The OSB is important, since Oakland schools are facing many problems, including declining enrollment, educational recovery from Covid closures, and financial problems. The OSB election used rank choice voting, so counting took time; the results were announced by early December. Nick Resnick won in District 5. This result was duly certified.
Then, on December 28th, the registrar dropped a bombshell. There was a mistake in tabulating ranked choice votes, and a different candidate, Mike Hutchinson, was the actual top vote getter.
Ranked choice voting sounds horrible.
The California Globe has more on this:
Alameda County Registrar of Voters Announces Counting Error For Oakland School Board Race
The Alameda County Registrar of Voters announced on Thursday that the ranked-choice voting system used by the County was not properly configured for the Fourth District Oakland Unified School District Board race in November, resulting in a new winner for the race and increased doubts over the county’s voting method.
In a statement, the Alameda County Registrar of Voters noted that “The ROV learned that its RCV tally system was not configured properly for the November 2022 General Election. It should have been configured to advance ballots to the next ranking immediately when no candidate was selected for a particular round. This means that if no candidate was selected in the first round on the ballot, then the second-round ranking would count as the first-round ranking, the third-round ranking would count as the second round ranking, and so on…
The entire race has now been thrown up in the air, as the new results go against the certified results, with candidates being told now that a judge would have to overturn the certified results in order to have the now-winning candidate be certified instead.
“Without being cynical, I now believe in holiday miracles. So it was very shocking to wake up this morning and receive a phone call at 10:30 a.m. from the Alameda County head of elections informing me that I had actually won the election,” said Hutchinson.
People are now questioning the city’s mayoral election.
With the school board election now in chaos weeks after it was supposed to be over with, and lingering concern over how the Mayoral vote was tallied, many in Oakland are now calling for an end of ranked-choice voting. For several election cycles now, the concern over ranked choice voting, a system that allows voters to vote for up to three candidates, but must rank them under a first, second, or third vote, with the end votes then being tallied up, has grown considerably.
Here’s a video report from NBC Bay Area:
There’s a straightforward way to make sure this doesn’t happen again. Return to traditional voting practices.
Just to give you an idea of how confusing ranked choice voting is, check out this report from Mercury News:
Why didn’t Oakland’s ranked choice ballot follow city charter?
The ongoing dispute over Oakland’s November school board election has opened a floodgate of questions about how Alameda County conducted its ranked choice voting system, and officials so far have no quick or easy answers.
A key question that officials have not yet answered: Should voters in Oakland’s mayoral race have been allowed to choose between all 10 candidates who ran for the office and not merely five?
According to the city charter, they should have. The charter reads: “The ranked choice voting ballot shall allow voters to rank as many choices as there are candidates.”
The charter does specify elsewhere that ballots could be limited to no fewer than three choices if the available voting equipment “cannot feasibly accommodate” the total number of candidates running.
But the Dominion Voting Systems election software used by Alameda County did, in fact, allow for ballots to include 10 choices, according to Registrar of Voters Tim Dupuis.
Prior to the November election, Dupuis met with the previous Oakland city clerk, and they opted to raise the number of choices on the election ballot from three to five — a policy shift for which there was no public announcement.
Featured image via YouTube.
DONATE
Donations tax deductible
to the full extent allowed by law.
Comments
“Ranked choice voting sounds horrible.”
It IS horrible. Any voting system which has a great tendency to confuse the voters is good for one particular political group (which pushes the system) and bad for actual voting results. There are a couple of multi-millionaires (who’s names escape me at the moment) who fund pushes for states to convert to ranked choice voting. IIRC they funded the push for Alaska and South(?) Dakota to adopt the system. They barely won in Alaska (50.5%) and lost in SD. 60% of the first ballot in the AK House race was Republican, but ranked choice voting installed the first Democrat in years – and, as outfits like VOX singing the praises of ranked choice voting prove, that is the desired outcome.
I was going to say the same thing.
It is horrible.
Any voting system that forces you to vote for someone you don’t want to or lose your vote is horrible.
That’s what ranked choice does.
Your top choice doesn’t get it? Then vote for someone you don’t want or don’t get a vote.
Doesn’t matter if your top choice was leading.
Any red state that adopts it is verifiably nuts.
Every system forces you to do that. On the contrary, ranked choice is the only system that lets you vote for whom you really like first, and only when all the candidates you actually like are eliminated do you have to choose between the ones you don’t like.
Bullshit. If your top choice is leading then your vote stays with that choice. Your second choice only comes into effect if your top choice is out of the running, and your third choice only comes into effect if both your top choices are out of the running, etc.
In the old system if your top choice is out, you get no further say. So in order to have a say you can’t vote for whom you really want, and are forced to predict which two candidates will be the final alternatives and choose between them, even if you hate them both almost (but not quite) equally.
Ranked choice voting requires confidence that the administrators of the election can handle the administration of the formula. Given the difficulties in many locations of handling the basics (registrations, counting, etc.), that seems to be a bit of stretch.
Anyone can handle the formula. And if the count is transparent, as it should be, then anyone can verify that it was conducted correctly. The fact that these people screwed it up speaks to their incompetence, not the difficulty of the voting method.
And the exact same thing applies to all counting methods! Incompetent administrators can just as easily screw up counting a preferential vote as they can screw up the count of a first-past-the-post vote.
That is nonsense. Nobody was confused in Alaska, unless they were morons who would be confused by any system at all. Yes, 60% of the first preferences were for either Palin or Begich; but Begich couldn’t win, and when faced with the choice of either Palin or Peltola a clear majority of those who had a preference at all preferred Peltola, so it was right and just that she should win. If a majority had preferred any Republican to a Democrat, then Palin would have won.
Hah. You fool yourself. Alaska got a Governor a majority of people did not *want*.
What are you talking about? Mike Dunleavy got an outright majority of the vote. 50.29% of voters preferred him to all the other candidates.
As for the House race, again an outright majority preferred Peltola over Palin. Of the 266,797 votes cast, 51.45% said that of those two they preferred Peltola, 42.16% said they preferred Palin, and 6.4% said they had no preference between them and if it came down to those two they would not vote. That gave Peltola 54.96% of those who did express a preference between them, but even counting the abstainers as a vote for “none of the above” Peltola still got a clear majority.
It’s not horrible to everbody. I am very pleased with how things turned out.
Nevada has a far better system. Every office on the ballot has the choice of “none of the above”.
That’s a meaningless gesture. NOTA votes are ignored in the actual count. Even if NOTA gets an outright majority, the candidate with the most votes still wins. Leaving the office vacant and calling a new election would be nice, but that’s not what they do. Just leaving it vacant for a full term is also an interesting idea, but for some offices it’s necessary to fill them.
See L. Neil Smith’s Probability Broach, in which NOTA won the 1972 presidential election, so in the presidential portrait gallery the 1973-77 term is represented by an empty frame.
“There are a couple of multi-millionaires (who’s names escape me at the moment) who fund pushes for states to convert to ranked choice voting.”
Smoke and mirrors. RCV is not the problem here, the problem is that at the same time they warp the entire underlying election process and claim it’s a necessary part of RCV. It isn’t.
RCV allows you to cast votes for as many candidates as you want, in a preferential order. “I really like Ron Paul, but if he really is unelectable, then I want my vote to go to George Bush, rather than have my vote for Ron starve George and help get Al Gore elected instead.” Makes all sorts of sense.
The millionaire’s group pushing “RCV” is really pushing a cross-party primary system. Instead of casting your ranked vote for your party’s nominees in your party’s own primary, then casting another ranked vote in the general, there is an animal-ball “top two” free-for-all primary including all nominees of all parties, that everybody votes in. The results get skewed by the number of nominees in each party. If one party has way more nominees than the other, chances are that the top two winners will be from that party, and when the general election comes around, you may as well stay home because there is nobody remotely reasonable to vote for.
This is the basic warping of the system that we need to fight. Every qualifying party should have SOMEONE on the ballot. Plus, who decided that “two candidates” is all anyone ever needs to choose between? That’s authoritarian bullshit. Might as well just offer one, Soviet-style.
“Top two” is how California is locking out Republicans in state races.
Henry, what you describe is a problem, but RCV solves it. Provided that a voter really does prefer any Republican to any Democrat, and expresses that preference in his primary vote, his vote will be transferred among Republicans as they are eliminated one by one, until it reaches the one who can use it. At no point will it help a Democrat.
The Alaska system produces not a “top 2” but a “top 4”. Hence the House race was between two Republicans and one Democrat, since the fourth primary winner dropped out.
RCV is one part of the issue. It seems purposefully confusing and introduces many more possible “human errors” as the case seems to be here.
I think we should also, though, look at the “certification” process. It seems pretty useless, a forced rubber-stamp to add legitimacy and propaganda value. What does it mean that results are certified?
Your impression might be that all the equipment was rigorously looked at for problems, ballots were inspected, systems were double-checked, etc.
The reality is that it just means that election officials signed off, and note that they don’t really have a choice. Mohave County, AZ, tried to delay certification due to election issues in ’22 but were threatened with felonies if they didn’t sign of as well as having the county’s results not included in statewide totals.
If there is no alternative but to sign off on botched elections, what good is certification other than propaganda value?
Mohave County was forced to certify their results because there were no issues in that county, and there was no doubt whatsoever that the results were correct. Had they been having an issue and could not determine that their results were right, they could and would have refused to certify them. Of course if they couldn’t fix it by the state deadline then the state would have had to go ahead without those results; what alternative would there be? Just as if a state doesn’t send in any electors, then the electoral college count proceeds without it. Or if a state doesn’t choose any representatives Congress proceeds without them.
There’s nothing confusing about preferential voting, which they’re now suddenly calling “ranked choice”, as if it’s some new thing. A small child can easily understand how to vote properly, and a middle-schooler can easily understand how to count the votes by hand. The problem here was not confusion, but a programmer not bothering to read the fine manual, or in this case to read the election rules before implementing them in code.
The system has been used in Australia for about a century, at every election on every level, right down to elections in elementary school (class president or whatever), and nobody is confused by it.
(If you want to see something at least slightly confusing, try the system used to count votes for the Australian senate; but even that is easily understood by anyone of reasonable intelligence. At any rate how to cast a vote in a senate election is dead simple; you don’t even have to rank the candidates, you can choose to let one of the political parties do it for you.)
Dear Milhouse,
Not everyone has your Einsteinian IQ and prodigious ability to understand the finer points of “ranked choice” voting (and because the Australians understand it so well only means that we, too, could find ourselves saddled with an unwanted Socialist govenment).
Voting should be as simple as “choose only one from this list for position one; choose only one from this list for position two, ad inf.
This way even a small child can easily understand how to vote properly,
You are wrong. You don’t need even an average IQ to understand that you look through the candidates and vote for the one you truly like best and would like to win in an ideal world. Then you look again and decide, if you can’t have that one, which of the rest is second best. Then repeat as often as you like, until you’re down to a bunch of candidates you hate equally, and if it comes down to them you don’t care which one wins. That’s when you stop.
Your system means that before voting a person has to go through all kinds of calculations of what other people are likely to do. Or else they end up voting for someone with no chance, and the worst candidate (from their point of view) ends up winning. So they’re forced to vote for the second-worst candidate and give up any hope of getting someone who is actually good, even if in actuality a majority would prefer that person.
Its so NOT confusing that the Oakland result ISNT a total mess right!!
This is the problem when the winner doesnt take all…it leaves the door open for people who actually got less votes OUTRIGHT to actually win.
I know, I know…we aren’t all blessed with a brain the size of the planet like yours Justice Milhouse but the results and the ensuing chaos speak for themselves.
It is not at all confusing. The Oakland people were just incompetent, and would have screwed up any count.
And yes, the whole point is that the person who got the most votes may actually be opposed by a majority, who all prefer someone else. In a first-past-the-post system they all have to vote for someone they don’t actually want, just to prevent the worst candidate from winning. With preferential voting the winner is the one whom a majority expressly prefer to the only viable alternative.
I am not stupid so no way will I ever like or want rank chose voting. Rank chose is a con game.
It seems too confusing to election officials to set it up properly, that’s the whole reason we’re talking about it.
Nope. The Oakland people were just incompetent.
Leave it to California to come up with a Byzantine, muddled voting process and then have the shocked Pikachu face when it all goes wrong.
It is neither Byzantine nor muddled, and California did not come up with it. It’s been around since the 19th century.
You must be trolling at this point.
Run a *primary*. Then furthest past the post. If you want people to vote for a different person, let them *vote* again. Don’t leave the system open to be gamed.
That is exactly what preferential voting does. It lets people vote again, right away. It is the only system that can’t be gamed, and doesn’t require people to game it.
So the city charter issue complicates the already tricky issue by forcing Hutchinson to appeal the certification based on a ranked-choice process that didn’t comply with law. And who knows what ballot security and tracing issues exist when the Registrar just pops up weeks later and says “hey, I think I got the math wrong — but trust me this time.” Ouch.
As a procedural matter, unless there is some undisclosed loophole here, I think the city charter problem creates a significant hurdle to directed relief — how does a Court order Hutchinson to be declared the winner based on a voting process that didn’t comply with the law? Smells like a revote situation (rare, but it does happen).
This is not a new problem to the computer industry. It’s called failure mode testing, and you’re SUPPOSED to do it on everything you release… otherwise you end up having your paying users beta-testing your code, and they get ornery about that.
You have someone with some brains (in the “QA Department”) set up a large number of test cases, some simple, some tricky, some exploiting known weak spots in other implementations. Then you run the programmer’s code against that data and make sure the correct result is achieved each time. (It’s even better if you are testing two programmers’ packages at the same time, since if the results ever differ, it’s obvious that one of them is wrong.)
If this yutz’s code screwed up the very first set of results, it’s a big red flag that nobody bothered to do this.
That may not be the issue. What I understand to have happened here is that the programmer was working with a different set of rules from a different jurisdiction, and some incompetent person went and used the same program for this election, which had different rules from the one it was written for.
Or it may be that the program was written for these rules, but neither the programmer nor the testers bothered reading them carefully first, so the test cases were set up for the wrong result.
“having your paying users beta-testing your code” is SOP. Microsoft has been doing it for decades. They don’t care if their users get “ornery”.
If the votes are public then anyone can recreate the count for themselves and verify that the original result was wrong and the new one is right. That may even be how the error was discovered in the first place.
Ranked choice voting is, IMO, a solution in search of a problem. Many eligible do not vote, which is fine, the govt shouldn’t make it a requirement. Our electorate relies in many States on the option of pulling the party lever, these folks do not evaluate the various candidates for each particular office.
If we must introduce coerced voting and a ranked choice system is just that then do it like an AP football top 20. Each voter puts the candidates in order of preference with points assigned. If five candidates then top choice gets 5 points, fifth choice gets 1 point. Then add the points for each candidate. Far simpler, less room for error and most people already understand how it works.
Ranked choice does not coerce anyone to vote. It doesn’t even coerce anyone to keep making choices once they’ve reached the point where they have no further preference. That’s what happened in Alaska, where 6.4% of voters said that if it came down to Peltola or Palin they didn’t want to vote for either of them.
From the voter’s point of view, you have described exactly what preferential voting does. Each voter puts the candidates in order of preference. No points involved. How does using points make the voter’s task any easier? But point systems can be gamed; the single transferable vote (which is what we’re discussing) can’t.
Milhouse, are you drunk, stoned, or just insane? RCV does coerce a person to vote if their ballot is thrown out because they didn’t rank all choices. If I’m voting and there is only one candidate that I want to get my vote, then requiring me to vote in any way for anyone else is coercion. There is no question about it!
“Ranked choice voting sounds horrible.”
Ranked choice voting is wonderful, if it’s done right. For one thing, it eliminates the “I really like candidate A, but I’d be throwing away my vote” FUD problem.
No, what you’re seeing here is the result of ranked equity diploma granting.
I grew up in an age of men capable enough to send three colleagues to the moon and back on a handful of reserved radio frequencies and the computing power of a Pac-Man game. The “adults in charge’ today have a global Internet and Amazon server farms the size of the moon, and they can’t count a bushel of potatoes the same way twice.
The average voter barely pays enough attention to be able to pick a candidate to vote for. Recall how many “undecideds” there have been in the past few elections as the election is on the doorstep.
Now ask these same people to rank the candidates in order and vote that way.
Hopeless. People will just vote 1 for their choice and then mostly randomly rank all the other candidates from their party #2-x.
Maybe something like this works on a small scale with an educated and involved electorate, but we don’t have that,
That’s a valid choice. They’re saying that after their #1 choice, all the remaining candidates from their party are equally acceptable to them, but they all rank ahead of the other party’s candidates. Had all the Alaskans who put Begich as their #1 choice felt that way, they would have put Palin second last and Peltola last, and Palin would have won. Peltola won because when it came down to those two a majority said they actually wanted her.
In any case, the way it usually works is that each party prints up and distributes a “how to vote” card, and most voters just bring that into the booth with them and copy it on to the ballot. That way they’re saying they agree with whatever their party decided. Those who disagree with their party have the option of changing it, and some do.
Nothing forces you to “rank the candidate” beyond the people that you know.
“I like Joe. If Joe doesn’t win, Bob wouldn’t be bad. Other than that, screw ’em all.”
That’s a perfectly valid RCV vote.
You’re not “ranking the candidates,” you’re “ranking your preferences.”
Once you’ve expressed those, you’re done.
But I’ve been told all our elections are clean, reliable, and fair. Any claim otherwise is just conspiracy theory agitprop from the Orange Man Bad.
For all we know, this election may well have been clean, reliable, and fair. All that we know is that whoever was in charge of programming the count screwed up.
Well then, DEFINITELY not reliable.
The count and the election are two different things. The election may well have been reliable. The count was wrong. That is fixed simply by recounting the votes and making sure you do it right this time. An unreliable election can’t be fixed, because the votes are unreliable, so it doesn’t matter how many times you count them, the result will still be unreliable. E.g. Georgia in 2020.
To expand on that: It turned out that in Georgia in 2020 the count was clean, reliable, and fair. All the allegations about shenanigans in the counting turned out to be false. The votes were recounted by hand, with adequate supervision, and the result was the same, thus proving that the original count was good. No alleged suitcases of ballots being counted twice, no Dominion kraken, everything was above board. What was not clean, reliable, and fair, was the voting. There are an unknown number of invalid ballots in the mix; they cannot be identified and removed, and every time you count the votes they will be counted too. Thus the clean, reliable, and fair count yields an unclean, unreliable, and unfair result.
Are you really that stupid? The count is part of the election process and cannot be separated. In other words, the counting process is integral to the whole election process.
Honest question.
So what if I don’t want my vote to go for anyone but the person I chose first, period. Would my vote be given to someone I don’t want to have my vote if I chose not to put anyone else in the rankings. The reason I ask is because the right to vote is not only the right to choose to give your vote to candidate A, but also the right to chose to withhold your vote from candidate B.
Sure. You vote for a #1 choice, then nobody else, just like today.
Somebody on this list might say, “I want Trump. Otherwise, screw em all.”
Then as Sanders is taking the oath, he might say, “Gee, I wish I had specified DeSantis as my second choice.”
And yet other countries manage to use ranked voting, some giving voters the choice of over 100 candidates to rank.
Do their electoral systems work?
The ones with over 100 candidates only work because there are workarounds. If you’re thinking of the Australian senate, most people vote “above the line”, checking a box indicating that they are accepting a specific political party’s recommendations. (These are submitted in advance and published, and available for perusal at any polling place.) Those who aren’t happy with any of the parties’ recommendations remain free to number all those candidates themselves, or as many as they care about. Filling in the ballot can take a good ten minutes, if you do that.
This is the Peoples’ Democrat Republic of Alta California. Votes are not cast in order to determine the wishes of the people or to indicate what the “consent of the governed” is for or against. They are cast, singly or in multiples, legally or otherwise, by citizens, non-citizens, or fictional people in order to provide a sufficiency of “ballots” to allow the Nomenklatura to declare that the person they want, won. Legitimacy is no longer considered to be necessary in our system of government. And fewer and fewer people believe in its existence.
Subotai Bahadur
I first heard of ranked choice in 1962 or 1963 at a Junior Statesmen convention. As best I can remember, Australia was using it in some elections. I have no idea if it is still in use there. May be a good subject for research by a graduate student.
It sounded like a good idea to us high school students, but there was no discussion of its advantages and disadvantages.
The Alameda county recorder should have provided enough transparency so that anyone could have checked their math.
With all due respect, transparency? In an election? In Leftist California??
Subotai Bahadur