More money has been traded on Yossi Cohen becoming Israel's next prime minister than on Gadi Eisenkot. Cohen is priced at a tenth of a cent. Eisenkot is the favourite.
Three and a half million dollars on Cohen. Just over three million on Eisenkot. And Eisenkot is sitting at fifty-four cents, which means the market thinks he's more likely than not to be sworn in.
That's the market Daniel wants us to read. Israel votes a week on Tuesday, and he's asking what Polymarket is actually predicting. Not the vibes around it, not the think pieces about whether betting markets are smarter than pollsters. The prices themselves. What the money says about who wins, who governs, and how sure it is.
There's a lot more of it than people realise. Everyone quotes the prime minister market, but there are about twenty-five live markets on this election. Seat counts for every major party, bloc totals, turnout, a hung parliament, whether Trump endorses anyone, whether the vote even happens on the day.
You've downloaded all of them.
I've downloaded all of them.
I could hear the fan on your laptop from the kitchen.
Prices as of this morning, Sunday the eighteenth. The headline market is the one with forty-seven and a half million dollars in it, which for a parliamentary election in a country of ten million people is a slightly absurd number. Eisenkot fifty-four. Netanyahu twenty-seven and a half. Bennett thirteen and a half.
And everybody else?
Liberman at two and a half, and then a long graveyard of names at a tenth of a cent. Ben-Gvir, Lapid, Golan, Ohana, Barkat, Feiglin. All of them at the floor price.
A tenth of a cent is the market's way of saying no without being rude.
It's the minimum tick, so it's as close to zero as the screen will show you. Now, the thing I'd want a listener to notice first is how tight the Eisenkot price is. The gap between what buyers will pay and what sellers want is two tenths of a cent. That's a properly liquid market. You could put a few thousand dollars in without moving it.
Tight isn't the same as right, though.
No. Tight just means lots of people agree on the number. Whether they're right is the rest of this episode.
Here's what I'd notice first. There's a second market, on which party wins the most seats, and it's more confident than the prime minister market. Yashar is at sixty-eight. Likud thirty-one and a half.
That gap is the most important thing in the whole set of prices. Sixty-eight percent that Eisenkot's party finishes first. Fifty-four percent that Eisenkot becomes prime minister. The market is telling you, in fourteen points, that coming first in Israel doesn't make you prime minister.
Which every Israeli knows, and every headline will forget on the twenty-eighth.
The president doesn't hand the mandate to the biggest party. Herzog consults every list that gets in, and he gives it to whoever he thinks is most likely to form a government, based on who the parties recommend. Then that person has twenty-eight days, plus up to fourteen more. If that fails, someone else gets a turn, and then the Knesset itself gets three weeks to find sixty-one signatures for anyone at all.
Win the night, lose the job.
Easily. And the market's pricing that. If you subtract, there's roughly a one in seven chance, in the market's mind, that Yashar comes first and Eisenkot still doesn't get sworn in.
Herman, can I guess where that one in seven goes?
Go on.
Bennett. He's the one the opposition might prefer if Liberman won't sit with Eisenkot's arithmetic.
Partly. But look at Netanyahu's price against Likud's. Likud is thirty-one and a half to come first, Netanyahu is twenty-seven and a half to be prime minister. They're close. The market thinks if Likud comes first, Netanyahu probably gets there. If Yashar comes first, it's much less sure about Eisenkot.
That's a market that thinks the right knows how to form a government and the centre doesn't.
That's one way to read it. The other is that the anti-Netanyahu side is three parties led by three people who all want the job.
That's the same reading, Herman.
Fine, it's the same reading.
How did it get here, though? This market's been open since last November. I assume it didn't start with a retired chief of staff in the lead.
It started with Eisenkot at one or two cents. For months. In December he was basically at zero. Netanyahu opened at about fifty-two and drifted up, and in early March he peaked at sixty-three cents.
Sixty-three. So at one point the market had Netanyahu as a clear favourite to win again.
Nearly two in three. And then within about two weeks he was at forty. He's never been back above fifty since. Bennett is the other half of the story. He was the alternative all winter, at thirty-six, thirty-eight cents, and he touched forty again in May.
Then he merged with Lapid. I remember the press conference. They looked like two men sharing an umbrella.
The merger was announced at the end of April, Together, with Bennett at the top. For a few weeks the price held. Then through June it fell off a cliff. Twenty-seven, fifteen, twelve. By August Bennett was at six cents.
The umbrella did not help.
That's the old Israeli pattern, two parties merged poll worse than their separate totals. And the money that left Bennett went straight to Eisenkot. He crossed forty cents at the end of June and fifty in late July, and he's basically been between forty-four and fifty-four ever since.
Wait. So the big move in this race happened in the summer.
In the prime minister market, yes. The last three months have been a flat line with wobbles. The seats market is different, and that's where this week has been wild. On the twelfth of October, Likud was a slight favourite to come first, fifty-two cents. By Friday it was at thirty-two.
Twenty points in five days. What happened?
I can tell you what happened in the world in those five days. I can't tell you which of it the traders were reacting to. Channel Twelve had Yashar on twenty-three and Likud on twenty. Channel Thirteen had Yashar twenty, Likud eighteen. Ofer Winter's party, Amcha Yisrael, slipped below the threshold in a couple of polls and lost a candidate. And there was a report about Netanyahu's decisions in the weeks before October seventh.
Any one of those could do it.
Any one of them. Amcha is the sneaky one, though. If Winter misses the threshold, that's three or four seats the right loses, and the Likud bloc's ceiling drops. A market on who finishes first shouldn't care about a small party's threshold. But the traders in it are the same people.
Here's my problem with all of this. You keep saying the market thinks. There's no market. There's a few thousand people with crypto wallets.
Do you wake up at night unsure who will win the elections in your own home? Introducing Ballot Box Monthly! Every month, a beautiful cardboard box arrives containing one prediction, hand-picked by our panel of experts, about who will be prime minister of your kitchen! Bronze members receive a prediction. Silver members receive a prediction and a coalition. Gold members receive a prediction, a coalition, and a second box containing the opposite prediction, so you are always right! Every prediction is priced in cents, and every cent is non-refundable! To cancel, simply visit our office in person on the day after the next election, in a year we will choose, and recite your account number to a man who does not like you. Ballot Box Monthly. The future, delivered late.
The future, delivered late.
Please don't read those back. It's worse in your voice.
You were telling me there is a market.
There is, and you're half right about the crypto wallets. Here's what the money looks like on Netanyahu. The biggest single holder of yes on him has about half a million shares. At twenty-eight cents, that's around a hundred and fifty thousand dollars of conviction. One of the other big holders calls themselves Only Bibi Twenty Twenty-Six.
Very balanced research desk, then.
Then there's this one, which I love. The single largest holder of no shares on Eisenkot is the same wallet that's the largest holder of no on Netanyahu and the largest holder of no on Bennett. The same amount on each, about four hundred and fifty thousand shares.
Someone's betting against all three favourites?
Well... that's what it looks like on the screen, but it almost certainly isn't a view. At most one of those three can win. So if you hold no on all three, at least two of your positions pay out a dollar whatever happens. My guess is it's a market maker who created complete sets, everyone at once, sold the yes side into demand, and is sitting on the leftover no. That's a liquidity business, not a prediction.
The biggest player on all three candidates has no opinion about the election. Lovely.
Probably not, no. And that's the thing about reading these. The size of the money is a terrible guide to the strength of a belief.
Which is your Yossi Cohen thing.
Which is the Yossi Cohen thing. Cohen, the former Mossad chief, has three and a half million dollars of cumulative volume. Gideon Sa'ar has over two million. Ayelet Shaked has two million. All of them are at the floor. That volume isn't people thinking Shaked will be prime minister. It's a year of people buying no at ninety-nine point something cents and grinding out a tiny return, and of market makers being paid to keep quotes up.
Volume, then, is mostly a measure of how long a name has been in the graveyard.
Mostly, yes. If you want the market's conviction, look at the price and the spread. If you want to see where the action is right now, look at the last week. Two point three million dollars traded on the prime minister market in seven days. Half a million yesterday alone.
That's real money in the last week. That's not graveyard grinding.
That's people with a view. Now, Corn, you're the sceptic in this house. Do you trust any of these numbers more than the polls?
I trust them about as much as the polls, for a different reason. The polls are a few hundred Israelis answering the phone, and the house effects are enormous. The market is a few thousand people, mostly not Israelis, mostly reading the same polls. The price is a weighted average of how people read polls.
With one difference. The polls don't lose money when they're wrong.
Neither does Channel Fourteen, apparently.
Ha. Fair. But that's where I'd push back on you. A market's price can be more than an average of polls, because the people who disagree with the polls put money on it. And here's the interesting test. If the market were just echoing mainstream polls, its seat counts would match them. They don't.
Go on.
Every major party has its own seats market. Likud, Yashar, Together, the Democrats, Shas, all the way down to Amcha Yisrael. Each one is a set of brackets, under twenty, twenty to twenty-four, and so on. So you can work out what the market thinks each party's expected seat count is.
You did this at two in the morning, didn't you.
I did. Likud comes out at about twenty-three seats.
The mainstream polls have Likud at eighteen to twenty.
Eighteen in Channel Thirteen and Maariv this week, twenty in Channel Twelve and Kan. The market is three to five seats more generous to Likud than every mainstream pollster. It's sitting between them and Channel Fourteen, which has Likud on twenty-nine.
The market's betting the polls are under-counting the right again.
That's what twenty-three means, yes. And that's not crazy. In twenty twenty-two the final polls under-counted Netanyahu's side by three or four seats. The Likud bracket market is the only seats market with real money in it, more than half a million dollars. Twenty to twenty-four is the favourite at forty-two and a half percent. Under twenty, where the mainstream polls are, is only twenty-four.
And the chance Likud gets thirty or more?
Under ten percent. Channel Fourteen's world gets about one chance in ten.
What does it say about Yashar?
Twenty-three, about the same as Likud. Which is a little odd, given the same traders put Yashar at sixty-eight percent to finish first.
That's not odd. That's a contradiction.
It's... no, you're right, it's a contradiction, though a mild one. Yashar's distribution is wide. There's real weight on twenty-six or more, but also a fifth of the money on under twenty. Expected values can be close while the chance of finishing ahead is lopsided. But here's the bigger problem with these seat markets. Add them up.
Add up what?
The expected seats for every party with a market. Twelve parties. You should get a bit under a hundred and twenty, because Hendel's list doesn't have a market. I get about a hundred and twenty-three.
The market has invented a bigger Knesset.
The market has invented three extra seats, and it's left Hendel out entirely. And it's because inside several of those little markets the brackets don't even sum to a hundred percent. Ra'am's brackets add up to a hundred and nineteen. Yashar's to a hundred and eighteen. The Democrats a hundred and fourteen.
Herman, a market where the odds add up to a hundred and nineteen percent is a bookmaker.
It's a thin market that nobody's bothered to arbitrage. Most of these seat markets have five to twenty thousand dollars of total volume. The Joint List's has five thousand. For comparison, the prime minister market moved half a million yesterday. So the seat markets are a sketch drawn by a few dozen people, and you should read them that way.
Then I'll ask you the question that matters in a sketch. Does anybody get to sixty-one?
There's a market for that too. It defines the blocs explicitly. Netanyahu's side is Likud, Otzma, Smotrich and Feiglin's list, Amcha, UTJ and Shas. The other side is just four parties: the Democrats, Yashar, Together and Liberman. Arab parties and Hendel don't count for anyone.
Hang on, the opposition bloc in that market can't include Ra'am?
Can't include Ra'am, can't include the Joint List. Which matters, because the realistic anti-Netanyahu government probably needs one of them. The market asks if either of those defined groups reaches sixty-one on its own. And the market says twenty-four and a half percent.
Three in four, nobody does.
Three in four. And there's a sister market asking which of those two groups gets more seats. Netanyahu's bloc is the favourite, sixty cents to forty.
Hold on. The market has Netanyahu's bloc out-seating the opposition, and Eisenkot as the favourite for prime minister.
Both. And they're consistent, which surprised me. Add up the market's own seat estimates for each bloc. Netanyahu's six parties come to about fifty-five and a half. The opposition's four come to about fifty-five. It's a dead heat, with a nudge to the right.
And the difference between fifty-five and sixty-one is where Eisenkot lives.
Ra'am, the Joint List, maybe Hendel. The Arab parties are worth about thirteen seats in the market's estimates. Netanyahu can't use them, and the opposition might. So the right can win the bloc count and still not have a government, and that's why both prices can be true.
You want to know the odd thing about that for me? We put our own forecast out last week. You had the coalition at fifty-one, and Yashar twenty-one, Likud twenty.
I did. And the market is to the right of me. I had Likud at twenty and the market says twenty-three. I had Amcha out, the market gives it about two and a half seats on average.
You said forty-seven million dollars of mostly foreign money was underpricing the right, and it turns out the money agrees with you more than the pollsters do.
On seats, yes. On the prime minister, I was less confident in Eisenkot than the market, forty-five to their fifty-four. So the market thinks the right does better than I expected and Eisenkot still gets there more often. Which only works if the market's very confident about Ra'am or the Joint List being in play.
Or confident Liberman bends.
Or that. What do you think the market's getting wrong, Corn? You've been sitting on something.
The hung parliament market. That's the number I'd look at. It asks whether a new election gets called before any government is formed. It's at thirty-seven and a half.
Higher than my one in four.
Higher than your one in four, and higher than I'd have guessed. Now put it next to the prime minister market. If there's a new election, Netanyahu stays on as caretaker, and caretakers don't count for that market. It resolves to whoever is properly sworn in. So nearly forty percent of the time, the people holding Eisenkot at fifty-four cents are waiting for a second election next year before they get paid.
Oh, that's good. The market resolves to other only if nobody's sworn in by the end of twenty twenty-seven.
A fifty-four cent ticket might be a fifty-four cent ticket you hold for fourteen months. That's a price with a lot of patience baked into it.
Patience is expensive. Money sitting in a contract until late next year could be earning interest somewhere else. That pushes long-dated favourites down and long shots up. It might be part of why Netanyahu holds twenty-seven cents when his bloc is short.
Or he's Netanyahu, and twenty-seven cents is what you charge for betting against a man who's survived every forecast for thirty years.
There's that too.
What about the other little markets? You said there's one on whether the vote even happens.
Eighty-eight percent that nationwide voting takes place on the twenty-seventh.
Nine days out. Twelve percent it doesn't happen.
Which sounds high until you remember where we live. The chief of staff reportedly warned that renewed fighting could force a postponement. Trump says he won't resume strikes on Iran before the American midterms. And Netanyahu's office has denied a report that he pressed Trump to start again. The market is pricing the version of next week where something happens.
A twelve percent chance of no election is the most Israeli number on the board.
Then there's turnout. The favourite bracket is seventy-two to seventy-four percent, at thirty-eight cents. Seventy-four plus is next. Last time turnout was about seventy point six. So the market expects a higher turnout than twenty twenty-two.
Which helps who?
Depends who turns out. That's the whole election, isn't it? And the Trump market, which I enjoyed. Will Trump endorse a party? No endorsement is sixty-four and a half. Likud twenty-eight, and up seventeen points in a week.
The market thinks Trump probably stays out, but if he jumps in, it's not for Eisenkot.
Yashar's at about five cents. For a president who called Herzog weak and pathetic over the pardon, staying out would be very disciplined.
I'd take the other side of no endorsement.
You would? At sixty-four cents?
He's endorsed in Hungary, he's endorsed in Argentina. Nine days is a long time for him not to post something.
That's not a forecast, that's a character reference.
Same thing, with him.
Last thing I'd flag on who's trading, because it's local. In March, an Israeli reservist and a civilian were indicted for allegedly using classified information to bet on Polymarket about the strikes on Iran last year, reportedly making more than a hundred and fifty thousand dollars. And the Israel Democracy Institute has proposed treating these prices legally as polls, which would bring them under the ban on publishing polls in the final days.
In some future version of this country, this episode goes out on Friday and gets us arrested.
It's a proposal. We're fine. But it tells you the people who worry about elections here think these numbers move voters.
Do they?
I genuinely don't know. Bandwagon effects from polls are real but small. A market price is a cleaner number, fifty-four cents, and it gets quoted as a probability. My guess is it matters less to voters than to journalists, who love a single number.
Which leaves me with the thing I don't get. Everyone treats the market price as the probability. Eisenkot is fifty-four percent. But a price is just what the last buyer and the last seller agreed. It's not a probability, it's a handshake.
It's not a share. You both keep saying shares. A share, you get a vote at the meeting. I've been to the meetings.
In fairness, Hilbert, Polymarket calls them shares.
Polymarket can call them what it likes. I went next door while you were on the Trump bit. Yehuda Kalderon, flat four, across the landing. He runs a book on the building. Has done for years. Mostly on the committee, who gets the parking space by the bins, whether the lift is fixed before Pesach. He does it on a chalkboard by his door and he settles in jars of pickled lemons.
Pickled lemons.
His wife makes them. Very good. That's the currency. One jar is a dollar, roughly, though he moves it about. Anyway, he'd seen the Knesset prices on his phone, so he's opened an Israeli election market. On the chalkboard. Eisenkot at fifty-four, same as yours. He copies them every morning. Except he's added a line nobody else has. Whether the Gutmans in flat two vote at all. They didn't last time. Their dog was ill. He's got that at sixty.
That's a turnout market. He's pricing turnout one household at a time.
He's pricing the Gutmans' dog, Herman. The dog's been well since August. I said so. He didn't move the price. I bought eleven jars of yes on the Gutmans voting. That's where it went wrong. Kalderon doesn't let you hold a position. He holds it for you. In his storeroom. And when the price goes against you, he takes a lemon out of the jar, as margin. Physically. He opens the jar, takes a lemon, closes the jar.
Your position is shrinking in real time.
It's lost four lemons since you two started on the seat counts. The Gutmans' daughter came home from Haifa this morning and Kalderon thinks she votes differently from her parents, and he's decided that lowers the chance any of them vote. I don't follow it either. Then the bloc market. He's got one for the building. Which side of the landing gets more votes. Our side is ahead, because Mrs. Peretz in flat six has registered her late husband's sister, who lives in Ashdod, to the building for the purpose of the book. Not for the election. Just the book. The election doesn't know.
He's running his own version of the threshold, then. Who counts as a voter in his market isn't who counts in the actual election.
I'm out of it now, anyway. I had one jar of my own. Not from the book. A jar his wife gave me at Rosh Hashanah, for fixing her radio. Kalderon took it as collateral for the Gutmans position. It's on his shelf now, with my name on a bit of masking tape. Sixty cents, he says it's worth. It was a present.
The most valuable position you held in this market was the one you didn't bet.
It was the one I'd been given. He priced it this morning. He'd never priced it before.
I'll tell you the thing, though. Corn said a minute ago, the price is just what the last two people agreed. The chalkboard by Kalderon's door. I've walked past it for six years and read it as what's going to happen in the building. It's what Kalderon will give you for lemons.
Then he's the only market maker in Jerusalem with a liquidity problem you can eat.
And he's doing what the big wallet on Polymarket does, Hilbert. He holds both sides. Whatever happens with the Gutmans, he still has the lemons.
He doesn't hold both sides. He holds the storeroom. That's a different thing. He settles at the price on the day you collect. Not the day you bet. You want your lemons back after the election, you get whatever a lemon's worth that morning.
Which, after a hung parliament, could be some time next year.
Fine. Next year. Tell the Gutmans.
The market's telling you something real, then, but not the thing the headlines say. A coin that's slightly weighted towards Eisenkot, sitting on top of an election that's a dead heat in seats.
To the right of the pollsters, though.
It's to the right of the pollsters on seats, and to the left of them on the prime minister. Which only holds together if the market believes the opposition can put together a government using the Arab parties, or Liberman's flexibility, or both. That's the real bet in all of this. Not who comes first, but whether the anti-Netanyahu side can do something it's only done once, in twenty twenty-one.
And whether that government lasts longer than the last one.
There's no market on that yet.
Give it a day.
The thing I'll take from this is when the money actually gets paid. Fifty-four cents on Eisenkot is a bet on who gets sworn in, and nearly four times in ten the market thinks that's another election away. Everyone will read Tuesday's prices as the result. For a lot of these tickets, Tuesday is barely the start.
Thanks as always to our producer, Hilbert Flumingtop, who keeps this show running from behind the desk.
This has been My Weird Prompts. You can find us on our Telegram channel, where every new episode lands as soon as it's out.
We'll be back soon.
