AI generated this summary from the Hebrew recording. No human has checked it, and it may contain errors. The recording is the authoritative source.
In this episode, Dr. Boris Gorelik, a data researcher and the podcast’s host, analyzes the results of the recent Israeli elections himself. According to him, raw election data is not a simple X-ray of the voters’ will, but a complex picture that requires interpretation through models. He argues that two researchers can look at the same data and reach completely different conclusions, depending on the underlying assumptions of the model they use. Gorelik demonstrates this by analyzing three cases: the supposed defection of Likud voters, the “Abu Yair” campaign in the Arab sector, and voting patterns in the Haredi sector.
Read the full episode summary
The supposed defection from Likud
Dr. Gorelik begins by discussing an analysis by data professional Elad Sidbon, who claimed that about 91,000 Likud voters from 2022 abandoned the party in the last election, mainly for Religious Zionism and Otzma Yehudit. Gorelik conducted his own analysis, in which he cross-referenced data from thousands of polling stations between the two elections. His model shows a completely different picture: only about a thousand voters left Likud, and 92% of its 2022 voters remained loyal to the party.
He explains that the huge gap between the two analyses stems from different underlying assumptions. Any analysis of voter transfers between parties is based on a statistical model and suffers from what is known as the “ecological fallacy.” Since the data is at the polling station level and not the individual level, it is impossible to know for certain whether a change in voting patterns at a specific polling station is due to voters switching parties or to population turnover. Gorelik’s model, based on Bayesian statistics, gives significant weight to inertia, meaning the assumption that voters tend to remain loyal to their party unless the data clearly shows otherwise. According to him, this approach prevents the “invention of chaos” where stability can explain the results. The conclusion is that any graph of vote transfers is not a foregone conclusion, but an argument based on the model’s assumptions.
The success and shelving of the “Abu Yair” campaign
On the second topic, Gorelik analyzes Benjamin Netanyahu’s “Abu Yair” campaign in the Arab sector in the 2021 elections. He examined polling stations in Arab localities (excluding Druze ones) where there was clear voting for the Arab parties. While support for Likud in these localities was historically less than one percent, the “Abu Yair” campaign doubled the rate of support. In cities like Nazareth and Rahat, the number of Likud voters grew threefold and fourfold. Gorelik defines this as a stunning success. His analysis shows that the new votes did not come from voters of the Arab parties, but from voters of other Jewish parties and from citizens who had not voted at all before.
If the campaign was successful, why was it shelved? Gorelik suggests three possible reasons. The first is the internal contradiction, as it is difficult to court Arab votes while simultaneously attacking political rivals for cooperating with Arab parties. The second is a simple cost-benefit calculation: the campaign brought in barely a quarter of a mandate, and it may have been decided to invest the resources elsewhere. The third explanation, and a cynical one according to him, is that the campaign’s real success was not in mobilizing votes, but in keeping Arab voters at home. By presenting Netanyahu as an alternative, the campaign may have damaged the motivation of Arab party voters to turn out and vote, thereby weakening their power.
The Haredi fortress
In the final part, Gorelik examines voting patterns in the Haredi sector. Despite expectations that parties like Otzma Yehudit, Noam, or the “Haredi Public” party would succeed in breaching the walls, his analysis shows a picture of almost absolute stability. While Shas lost a few votes within the Haredi core to Likud and Ben-Gvir, United Torah Judaism proved to be a real fortress. 95.5% of its voters remained loyal, and the number of votes it received in Haredi cities grew by 13%, exactly the rate of natural growth of eligible voters.
Gorelik argues that voting in the Haredi sector is not an expression of a personal and rational decision in the Western sense, but a demonstration of institutional power and obedience to rabbinic authority. According to him, it is almost impossible to penetrate this sector and take votes without the consent of the rabbinic leadership. He quotes a Haredi journalist who said that Haredim have the right to vote, but not the right to choose. Loyalty and defection in this sector do not reflect a personal decision, but an institutional one.
The original episode, in Hebrew: ישראל 2026 מהקלפי