We Called the Deadlock: How Our Israel Election Forecast Did, Part 1

Last Friday we ran Israel’s election 50,000 times and said the most likely outcome was a deadlock: nobody gets to 61 seats. Tonight at 10 pm the exit polls came out, and that’s what they show. You can read our full analysis below or read the simpler stuff that’s available elsewhere, designed for simpletons.

One big warning before we start patting ourselves on the back: these are exit polls, not results. In 2022 the exit polls had Netanyahu’s camp 2 or 3 seats lower than the final count. That’s why this is Part 1. Part 2 comes when the official results are in.

The scorecard

Before the election we gave a seat estimate for every party, plus a “likely range”: the window the party landed in for 8 out of 10 of our simulations. Here’s how those stack up against the three mainstream exit polls (Kan 11, Channel 12 and Channel 13). We’ll get to Channel 14 separately.

	Our forecast	Our likely range	Exit polls	Verdict
Netanyahu’s camp	53	48 to 57	50	Inside the range
Eisenkot’s camp	52	47 to 56	52 to 53	Bullseye
Arab parties	15	12 to 17	17 to 18	At the edge
Likud (Netanyahu)	22	19 to 24	24 to 25	At the edge or just over
Yashar (Eisenkot)	22	19 to 25	21 to 23	Inside
Beyachad (Bennett)	14	11 to 16	12 to 14	Inside
Joint List	9	7 to 11	10 to 11	Inside
The Democrats (Golan)	9	7 to 10	9 to 10	Inside
Yisrael Beiteinu (Liberman)	7	6 to 9	7 to 9	Inside
Shas (Deri)	8	6 to 9	6 to 8	Inside
United Torah Judaism	8	6 to 9	6 to 7	Inside
Otzma Yehudit (Ben-Gvir)	8	7 to 10	6 to 7	At the edge or just under
Religious Zionism (Smotrich)	7	5 to 8	6	Inside
Ra’am (Abbas)	6	4 to 7	7	At the edge
Amcha Yisrael (Winter)	0	Misses in 77% of runs	0	Right
The Reservists (Hendel)	0	Misses in 64% of runs	0	Right

So: 9 of the 11 parties that got in are inside our range in all three exit polls, though Ra’am only just. The other two, Likud and Otzma Yehudit, are on the edge. That’s about what an honest forecast should look like. If every party had landed dead on our number, we’d have been lucky, not good.

What we got right, and what we got wrong

Right: the deadlock. This was the main call, and it’s the one that matters. In 95% of our simulations, neither camp got to 61 without the Arab parties. That’s exactly where the exit polls put us.

Right: weighting pollsters by their track record. Our method didn’t throw out Channel 14 or i24. It gave them weight based on how accurate they’d been in the past. The result landed within a seat or two of where the mainstream exit polls are tonight.

Right: the threshold. We said the two parties sitting near 3.25% would most likely both miss. That was our single most likely outcome, at 45%, and that’s what the exit polls show.

Wrong: Likud. We had Likud and Yashar tied at 22, with Yashar slightly more likely to come first. Every exit poll has Likud first with 24 or 25. Look at where those extra seats came from: not from Eisenkot’s side, which is right where we had it. They came from Likud’s own partners. Shas, United Torah Judaism, Otzma Yehudit and Religious Zionism come out about 5 seats lower than we expected, and Likud about 3 seats higher. Netanyahu’s camp got smaller overall, but its voters crowded into Likud at the last minute. The polls didn’t see that coming, so neither did we.

Wrong: the Arab parties, and this is the story of the night. We said about 15 seats. The exit polls say 17 or 18, up from 10 in the last Knesset. We’d flagged Arab turnout as the biggest blind spot in Israeli polling, and that’s exactly where we missed.

Turnout overall was huge: 74.3% by 8 pm, the highest since 1988 (Davar). But look at Arab towns, as reported by Ynet:

Town	Turnout by 8 pm tonight	Final turnout in 2022
Umm al-Fahm	70%	38%
Hura	76%	49%
Nazareth	70%	48%
Shefa-Amr	73%	52%
Rahat	80%	61%
Sakhnin	84%	68%

And those are numbers from two hours before the polls closed. The final ones will be higher.

Channel 14 vs everyone else, round two

Before the election, two outlets saw a very different race, and they share a family tree. Shlomo Filber, who ran Likud’s 2015 campaign, founded the polling company Direct Polls. In 2025 they split: Direct Polls, now run by Tzuriel Sharon, went on polling for i24 News, and Filber went on polling for Channel 14 through a firm called Next Data. Channel 14 had Netanyahu’s camp at 62. i24 had it at 58. Everyone else had it at 48 to 51.

Their exit polls tonight are closer to everyone else’s, but the gap is still there:

	Final poll before the election	Exit poll tonight
Channel 14	62	60
i24 News	58	55
Kan, Channel 12, Channel 13	49 to 51	50

Channel 14’s exit poll has Likud at 30 and Liberman at just 4. Everyone else has Likud at 24 or 25 and Liberman at 7 to 9. Those can’t all be right.

So is Channel 14 just biased? Filber is not a neutral figure. He ran Likud’s 2015 campaign, and Netanyahu then made him director-general of the Communications Ministry. Later he turned state’s witness against Netanyahu in Case 4000, one of the corruption cases against him. Today he polls for a right-wing channel.

But leaning right and being wrong are not the same thing. Here’s how Direct Polls did on Netanyahu’s camp in its final polls while Filber ran it:

Election	Direct Polls said	Actual result	Everyone else said
2020	59	58	56 to 58
2021	53	52	49 to 51
2022	62	64	60 to 61

In 2021 and 2022, most pollsters underestimated the right, and Filber’s lean was a correction, not a mistake. That’s why our model gave Direct Polls one of the highest weights of any pollster. What his record doesn’t include is a gap like this year’s: 8 to 12 seats from everyone else instead of 1 to 3, and about 10 in tonight’s exit polls. If the final count moves toward Channel 14 the way it did in 2022, Filber gets the last laugh again. If it doesn’t, the lean finally caught up with him.

One more detail worth knowing. On election morning, Filber went on Channel 14 and said Netanyahu’s camp would get at least 60 seats. He also named the one thing that could break his forecast: Arab turnout. If Arab voters surged, he said, “that’s a different story.” He dismissed Ahmad Tibi’s claim of 70% Arab turnout as “fiction” (ICE). By 8 pm, Umm al-Fahm was at 70%.

Why these numbers aren’t final

Exit polls are surveys of people walking out of polling stations. They’re good, but they miss. Soldiers’ ballots and other special votes are counted last, and the Central Elections Committee says counting them could run until Friday morning. Winter’s and Hendel’s parties are below the threshold in all five exit polls, but no one has published their exact percentages yet (Maariv).

The biggest thing that could still change is the threshold. To get into the Knesset, a party needs 3.25% of the vote, about 4 seats. Winter’s Amcha Yisrael and Hendel’s Reservists both appear to have missed, but the exit polls don’t say by how much. If either one is at 3% or so, the real count could still push it over.

That would matter, because 4 seats would come from somewhere. Using our simulations, here’s roughly where they’d come from, applied to tonight’s mainstream exit polls:

If…	Netanyahu’s camp	Eisenkot’s camp	Arab parties	Hendel
Nobody else gets in (exit polls as they stand)	50	52 to 53	17 to 18	0
Winter gets in	about 53	about 50	about 17	0
Hendel gets in	about 47 to 48	about 51 to 52	about 16 to 17	4
Both get in	about 50	about 49 to 50	about 16 to 17	4

The logic: In our simulations, the nights Winter got in were good nights for the right overall, so his 4 seats were a net gain for Netanyahu’s camp and most of the cost fell on Eisenkot’s side. Hendel’s seats come mostly out of Likud, and he hasn’t committed to either camp. Neither scenario gets anyone to 61.

Who can govern now

On Friday we lined up the seat math against what every leader has sworn. Here it is again with tonight’s numbers from Kan, Channel 12 and Channel 13:

Possible government	Seats in the exit polls	Gets to 61?	What stands in the way
Netanyahu’s current coalition	50	No	The numbers
Eisenkot, Bennett, Golan and Liberman	52 to 53	No	The numbers
The same, plus Ra’am, like the 2021 Bennett-Lapid government	59 to 60	No, even with Ra’am	Bennett and Liberman have ruled out Ra’am anyway
Eisenkot’s camp as a minority government, with the Arab parties backing it from outside	69 to 71	Yes	Liberman says no minority government, not even a temporary one
Likud, Yashar, Beyachad and Yisrael Beiteinu (a unity government)	67 to 69	Yes	Eisenkot, Bennett and Liberman all refuse to sit with Netanyahu
Eisenkot’s camp plus Shas	59 to 60	No	Liberman says no Shas

Tonight’s new detail: on Friday, the 2021-style coalition with Ra’am had 61 seats in about a third of our simulations. On these exit polls it’s short even with Ra’am. The options that have the numbers are the ones that require someone to break a promise.

What they said tonight

• Eisenkot said the exit polls show “total no-confidence” in Netanyahu, and that he’ll move immediately to form a “Zionist draft government,” meaning a government that drafts ultra-Orthodox men (Ynet). Earlier this month he said that even without 61 seats, he would bring his government to a Knesset vote, and he wouldn’t ask the ultra-Orthodox or Arab parties to vote for it (Ynet).
• Liberman rejected a minority government again tonight (Ynet).
• Mansour Abbas said this morning that Ra’am could be the piece that completes a second “change government,” like the one he joined in 2021.

The other deal on the table: a rotation with Netanyahu?

The unity government in our table has a reported price tag. On Sunday, Israel Hayom reported that the Trump administration is pushing for a broad government of Netanyahu and Eisenkot (Israel Hayom). According to the report:

• Netanyahu would serve about two years, end the wars, push a regional deal built on the Abraham Accords, and then retire.
• Ben-Gvir’s party would be left out.
• Eisenkot’s camp would want a written commitment that Netanyahu retires, the judicial overhaul cancelled, and a state commission of inquiry into October 7.

Treat this with caution. None of it is confirmed. Eisenkot said just last week that he won’t serve under Netanyahu, “not even in a rotation agreement” (JPost). Herzog’s office denied the part of the report about a pardon for Netanyahu. But if the deadlock drags on, this is the deal people will be talking about.

The long shot: Shas, and Israel’s first Mizrahi prime minister

There’s one more scenario worth talking about: Shas, the ultra-Orthodox party of Sephardi and Mizrahi Jews, crosses over to Eisenkot and makes him Israel’s first Mizrahi prime minister.

Before you get excited, here’s what Shas leader Aryeh Deri has actually said. In July he said Eisenkot “leads a camp of hatred against everything in Judaism” (Ynet). Two weeks ago he softened, telling Kan’s religious radio station: “I’m in favor of a broad government with Eisenkot, led by Netanyahu” (Srugim). So Deri is open to Eisenkot. Just not as prime minister.

Still, it isn’t impossible:

• Shas has done this before. Under Deri, it sat in Rabin‘s government in 1992 and Barak‘s in 1999. Deri goes where the deal is.
• There’s a deal to be had on the draft. Deri says Shas won’t join any government unless the status of yeshiva students is “regulated” by law (Times of Israel). He wants legal certainty, not necessarily zero draft. Eisenkot wants to end the blanket exemptions but has said he’d accept exemptions for up to 3% of yeshiva students (Wikipedia). That’s a number you can negotiate around.
• The Mizrahi angle is real. Eisenkot was born in Tiberias to parents from Morocco. Every prime minister Israel has ever had came from an Ashkenazi family. Shas was founded by Rabbi Ovadia Yosef on a promise to “restore the crown to its former glory” for Mizrahi Jews. Putting the first Moroccan son in the prime minister’s office is a story Deri could sell. Former Shas MK Yigal Guetta already said it on Channel 12 this summer: “Yes, a Moroccan prime minister!” (Point of No Return).
• Shas’s alternatives look bad. At 6 to 8 seats, with Netanyahu’s camp at 50, Shas is looking at opposition or yet another election.

And here’s why it probably won’t happen:

• Shas alone doesn’t get Eisenkot to 61. Eisenkot’s camp plus Shas is 59 or 60 in the exit polls.
• Liberman says “no Shas” (JPost). If he walks, Eisenkot plus Shas drops to about 51 or 52, and adding United Torah Judaism only gets to 57 to 59.
• UTJ wants a draft law “without sanctions and without quotas” (AIJAC). A Shas deal on quotas means breaking with UTJ and taking on the rabbis whose word moves Shas voters.
• Shas voters lean right, and tonight’s exit polls show right-wing voters crowding into Likud, not drifting away from it.

Our guess: about 1 in 20. That’s a judgment call, not a model output, and it’s lower than the 1 in 10 we’d have said before reading Deri’s own words. What would raise it: a final count that moves a seat or two toward Eisenkot, Liberman agreeing to sit with Shas as long as UTJ stays out, or Likud starting to talk about life after Netanyahu. Watch Deri this week. If he talks about a broad government and stops adding “led by Netanyahu,” pay attention.

Coming in Part 2

When the official results are in, we’ll do the real scorecard: every party against our forecast, which pollster came closest, whether Channel 14 or the mainstream channels got it right, and whether Winter or Hendel snuck in at the last minute. The full model and data from Friday are still available (just email us at ck at jewlicious dot com) if you want to check our work.

Sources

• Exit polls (Kan 11, Channel 12, Channel 13, Channel 14): Israel Hayom
• Turnout at 2 pm: Times of Israel; turnout by 8 pm: Davar; Arab-town turnout: Ynet
• Threshold parties in the exit polls: Maariv
• Filber on election morning: ICE
• Our pre-election forecast: Jewlicious
• Eisenkot tonight and Liberman’s response: Ynet; Eisenkot on a vote without 61: Ynet
• The reported rotation deal: Israel Hayom
• Eisenkot’s background and draft position: Wikipedia
• First Mizrahi PM and the Yigal Guetta quote: Point of No Return
• Deri on Eisenkot: Ynet, July; Srugim, October
• Deri’s condition: Times of Israel
• Liberman’s red lines: Jerusalem Post
• UTJ’s position: AIJAC