Eastern Margins

Postmortem on my Knesset election predictions: The day before the election, this was my projection for how many mandates each party would get. In parentheses, I’ll show the difference compared to the actual results. Coalition: Likud: 23 (−4) UTJ: 8 (+1) Shas: 8 (+2) OY: 7 (=) RZP: 7 (+1) 53 seats (=) Opposition: Yashar: 21 (=) Beyachad: 15 (+3) Democrats: 8 (=) YB: 8 (+1) 52 seats (+4) Arab Parties: Joint List: 9 (−2) Ra’am: 6 (−2) 15 seats (−4) On its face, my biggest success was projecting the exact proportion of voters that would support the coalition bloc. However, I underestimated how large a share of the coalition would be comprised of Likud itself. I must admit I drank some of the kool-aid from MSM polls that showed Likud hemorrhaging support. In the end they lost 5 seats, but still gained an additional 52,000 votes compared to 2022. Likud’s relative success came at the expense of other coalition partners, most notably Shas. Due to polling data and demography, I expected the party to lose some support compared to the 11 seats it won in 2022, but still receive 8. The truth is that Shas is a dying party. It doesn’t pass the electoral threshold among first-time voters, the new generation of Mizrahi religious Zionists prefer Ben-Gvir, and many traditional voters from the periphery opted for Netanyahu. I was aware of these trends but didn’t anticipate such a significant drop. I slightly overestimated Smotrich’s performance, as I was expecting Religious Zionists to come out to vote at slightly higher proportions in Beit Shemesh, Jerusalem, and Ma’ale Adumim. Smotrich and especially Feiglin’s bases are higher propensity, so it makes sense that some polls over-indexed their support. As for the opposition, I assumed they would accrue a larger proportion of both the overall and Jewish vote. I nailed Yashar and the Democrats’ performance, but the error stems from my severe overestimation of Bennett’s performance. Again, I drank the Kool-Aid from MSM polls about a Beyachad rebound in the final two weeks of the election. Part of this can be explained by my lack of accounting for record-high Arab turnout. I must admit I thought Bennett would garner a small but nevertheless notable contingent of ’22 Netanyahu voters, primarily upscale secular center-right Jews from areas like Hod Hasharon, Ra’anana, and Kfar Saba (I know these areas have always supported the opposition, but I thought Bennett could flip some votes from the Likud in those areas). I think my error was due to a subconscious bias as I'm terminally online and Bennett did a good job promoting his list on social media. And I didn’t foresee 78% Arab turnout, neither among northern or southern Bedouins, let alone the Arabs in the Galilee or the center.