Analysis |

No Matter Who Wins Israel's Election, the Pollsters Will Lose

At no point in the campaign have the polling results been so disparate. Here's why

Amnon Harari
Amnon Harari
Send in e-mailSend in e-mail
Israeli soldiers voting early on a military base, April 6, 2019.
Israeli soldiers voting early on a military base, April 6, 2019.Credit: AMIR COHEN/REUTERS
Amnon Harari
Amnon Harari

On Tuesday, four months of election campaigning will come to an end. Eighty-one opinion polls were conducted in recent months in an effort to gauge the voters’ intentions and predict who the next prime minister will be. But despite all of the events that have taken place during that time, it’s actually during the last lap, with the most recent polling – in which the television stations and newspapers have invested big bucks on larger voter samples and more diverse polling methods – that the situation appears more confused than ever.

If you ask pollsters Mano Geva and Mina Tzemach, Benny Gantz’s Kahol Lavan party is expected to become the largest party with 30 seats in the 120-seat Knesset, compared to just 26 for Prime Minister Benjamin Netanyahu’s Likud. But Likud is still projected to form the next government at the head of a bloc of 63 or 64 seats. Geva and Tzemach have come to that conclusion twice, in two large opinion surveys conducted for Channel 12 and for the Yedioth Ahronoth daily.

Camil Fuchs thinks otherwise. His survey results for Channel 13 that were reported Friday project Likud and Kahol Lavan tied at 28 seats. A Haaretz poll published two days earlier actually showed Likud significantly ahead at 30 seats to Gantz’s party’s 27.

Haaretz Weekly Ep. 21Credit: Haaretz

At no point in the campaign have the polling results been so disparate, and it’s hard to explain why. One possibility is the different weight that the pollsters give to the degree of certainty expressed by survey respondents. The various polling models weigh the responses regarding various parties and the degree of certainty that the party will get their vote (or whether a respondent won’t vote at all). Even if respondents answer in a similar manner, pollsters may give different weight to the responses.

>> Read more: Netanyahu's paranoid Hail Mary may become political hara-kiri | Chemi Shalev ■ Your comprehensive guide to the Israeli electionWhy this is the most difficult election to predict since Netanyahu's first victory

Another possibility is that the minimum threshold required for representation in the Knesset under Israel’s proportional representation system – which is currently the equivalent of four Knesset seats – will mean that if a number of parties are hovering right around the threshold on Tuesday night, we won’t know then who the next prime minister will be. According to recent polls, seven parties, four on the right and three on the left, are projected to get no more than six seats. Under such circumstances, a small overestimation by pollsters who wrongly project that a party will make it into the Knesset could shift the balance between the left and right voting blocs. Overestimations could also “deprive” at least a seat from each of the large parties.

And even if the polls were spot on, it’s still difficult to predict what the vote will be in any election, all the more so this time around. The approach of Election Day does to the distribution of votes what alcohol does to a dull party. Unexpected political developments occur in record time, major sums of money are spent on aggressive advertising, and if history is a guide, shifts in voter support occur quickly. In the past four elections, the large parties picked up or lost as many as eight seats in the last few days before Election Day, after the final polling results permitted by law to be published were released. There has also been movement in the past between major blocs of parties.

    But can such shifts really change the final outcome? The short answer is probably not. An examination of the projections, when it comes, for example, to the voting blocs – right vs. center-left in current electoral parlance – shows that the pollsters have acquitted themselves relatively well. In the most recent election in 2015, the pollsters did miss a last minute Likud surge, which handed the party eight seats between the final polls and the actual vote. But the extra Likud seats came from within the right-wing bloc of parties. The final polls showed the right getting 66 seats, and at the ballot box the result was very close – 67.

    The most recent polls project about 64 seats for the right-wing and ultra-Orthodox bloc – good news for Netanyahu, who would be able to form a majority government if all the parties joined his coalition. But if we also factor in the minimum four-seat threshold, and if a party on the right that was projected to be part of the bloc fails to make it into the Knesset, it could put Gantz back in the running as potential prime minister.

    If even two of the small right-wing parties lose a quarter of their support, and the right-wing bloc loses four seats, and if Orli Levi-Abekasis’ Gesher party somehow manages to make the four-seat threshold, and Meretz and Ra’am-Balad also make it in, it could become impossible for Netanyahu to form a majority coalition. That’s a lot of stars that would have to align, looking at the latest polls, but individually none of them is impossible.

    Even if it does happen, it still doesn’t make Gantz prime minister. He and his party colleagues are currently at work on another important task – making Kahol Lavan the largest party in the Knesset by a considerable margin over Likud, on the argument that a vote for any party other than Kahol Lavan -- the party whose leader has the best chance at unseating Netanyahu -- is a vote for Netanyahu himself.

    In 2009, Kadima leader Tzipi Livni used a similar argument at the same point in the campaign. It worked tactically to boost her party’s support, from the 23 to 25 seats that the final polls projected to 28 in reality. That’s the same number of seats that Gantz is hoping to pick up.

    The Labor Party is discounting the importance of such an effort, not only recalling that the tactic fell short of making Livni prime minister, even though her party had the most seats, but also because if voters are convinced by Gantz’s argument, it could come at Labor’s expense. Such a tactic could even result in the disappearance of potential coalition partners for Kahol Lavan, but one can also understand Kahol Lavan’s rationale.

    If the support for the two blocs turns out to be particularly close this time around, more so than the final weekend polls project, a four- or five-seat advantage for Kahol Lavan over Likud could very well help the party attract coalition partners that are currently seen as members of of a Likud-led coalition government – Moshe Kahlon’s Kulanu party, for example. That the approach didn’t work for Livni in 2009 doesn’t necessarily mean it couldn’t for Gantz in 2019, especially since a decade ago Netanyahu wasn’t facing potential indictment for bribery. That could present his coalition partners with the prospect of serving in a cabinet under a prime minister who is on trial, or supporting problematic legislation giving a sitting prime minister immunity from prosecution.

    Ultimately, just two days before the election, Gantz needs a few miracles that no poll is projecting if he means to bring about a change at the top. Although he has a chance, even if theoretical – with all of the open questions, the high minimum threshold, aggressive efforts at political spin and an Israeli voter who can lead pollsters astray – in the 2019 election one thing is certain. Those poor pollsters can’t possibly win.

    Click the alert icon to follow topics:



    Automatic approval of subscriber comments.
    From $1 for the first month

    Already signed up? LOG IN


    בנימין נתניהו השקת ספר

    Netanyahu’s Israel Is About to Slam the Door on the Diaspora

    עדי שטרן

    Head of Israel’s Top Art Academy Leads a Quiet Revolution

    Charles Lindbergh addressing an America First Committee rally on October 3, 1941.

    Ken Burns’ Brilliant ‘The U.S. and the Holocaust’ Has Only One Problem

    Skyscrapers in Ramat Gan and Tel Aviv.

    Israel May Have Caught the Worst American Disease, New Research Shows

    ג'אמיל דקוור

    Why the Head of ACLU’s Human Rights Program Has Regrets About Emigrating From Israel


    Netanyahu’s Election Win Dealt a Grievous Blow to Judaism