No One Has Won Israel's Election—Yet
No One Has Won Israel's Election—Yet
And why you might have wrongly heard otherwise in major media outlets
At 4:42pm Eastern time, or 42 minutes after the polls closed in Israel, the New York Times ran a story headlined “Netanyahu has path to a majority and sixth term, exit polls show.” The actual article, which was a little less definitive, began:
Prime Minister Benjamin Netanyahu’s party held a lead in Israel’s fourth election in two years, exit polls projected Tuesday night, giving him a chance of forming a coalition to stay in power for a sixth term.
Three broadcasters’ exit polls projected that Mr. Netanyahu’s party, Likud, won from 31 to 33 seats, while his wider right-wing bloc won 53 to 54 seats — short of the 61 seats he needs to form a majority coalition in the 120-seat Parliament.
Mr. Netanyahu’s most obvious path to power now depends on Naftali Bennett, a rival right-winger whose party won seven to eight seats, and who could be a kingmaker.
All that Netanyahu needed to stay in power, suggested the Times, was to join up with another right-wing politician. Pretty good news for Bibi! The only problem was that this early assertion didn’t have much basis in reality. In rushing to get a story into the paper of record to feed the internet’s appetite, the Times had to rely on incomplete and imprecise exit polls for its coverage. And this led it to draw far more confident conclusions than warranted.
37 minutes before the Times published their story, after reading the same exit polls they had (released at 4pm ET, 10pm in Israel), I explained why we didn’t actually have any idea who’d come out ahead in Israel’s election:
Sure enough, almost three hours after the Times declared that Bibi seemed on his way to another term, all three Israeli exit polls were adjusted—against him.
Netanyahu may still win this round, in other words, but we’re not going to know for a while.
My point here is not to pick on the Times or its excellent Israel reporters, who are generally better than most, and who acknowledged the results could change, though they could have been more clear to their readers about the precariousness of their projections. My point is that today’s media has a structural problem when it comes to covering contemporary elections.
In an age of social media-driven desire for instant results and analysis, even the best outlets have been forced to publish faster and less circumspectly than they would have 20 years ago. After all, if the Times won’t publish speculation based on the latest incomplete Israeli exit polls, someone else will—and will get the traffic, hurting the NYT’s market share and possibly their centrality in the discourse.
With these incentives, and an unwillingness to level with their readers and simply say “we don’t know yet,” many major media outlets have resorted to breathlessly reporting every development in a story, rather than waiting for all the facts to be known. This is particularly problematic when it comes to election returns, as we saw in the United States, when election day in-person ballots favored Trump, but as expected, mail-in ballots that were counted afterwards favored Biden.
Trump was able to call the election result into question in part because the media insisted on reporting on incomplete results, which gave the false early impression that Trump was winning, rather than suffering a historic defeat for an incumbent.
But just as experts on Israeli exit polls knew that they weren’t definitive and probably shouldn’t be the basis for a big story, experts on the U.S. election had been projecting for months that Biden’s lead would grow as the count progressed. This did not stop media outlets from calling a dubious play-by-play anyway, and giving their audiences a misleading perception of the actual reality.
To be fair to these outlets, it’s hard to know what the solution is here. The up-to-the-minute information is out there, and the appetite for it is vast, so even if places like the Times decided not to report results until they were more solid, other less reputable places would probably step into the breach.
But to leave it at that feels defeatist to me. The internet has changed the journalism game, yes, but the media can change with it—and that includes coming up with new industry standards for how to cover unfolding news events like elections. If major television and print organizations collectively settled on a policy for covering elections that relied on final vote counts, rather than projections or incomplete exit polling, and explicitly educated readers as to why they were doing this, the national expectations surrounding these events would change.
Readers are capable of understanding the limitations of reporting; reporters just need to take their audience seriously enough to explain those limits to them. In the end, this would be a win for both parties—enhancing the credibility of today’s journalists while better informing their readers. We reporters just have to respect our audience enough to tell them the truth: “We don’t know yet.”
UPDATE, 6am ET: Remember how I said that if the smallest Arab party—a right-wing Islamist one whose leader could back Bibi or bury him—gets above the electoral threshold, you can throw all the previous takes out the window? Well:
If this holds, the kingmaker in Israel’s elections will be an Arab Islamist party that is Bibi-curious, which is going to make so many people’s heads spin. Too often, the media tends to ignore the growing role of Arab parties in Israeli politics—I’d wager most international audiences don’t even know that Israel has Arab parties in its parliament—but that’s not going to be tenable for much longer.
We discussed these very elections in our first Ask Me Anything for paid subscribers on Monday! We also tackled everything from Passover to parenthood to the forthcoming Palestinian elections. It was a great conversation, and if you’d like to get in on the next one, you can upgrade your free subscription here:
Thank you for reading this edition of my newsletter. If you liked what you read and want to support this work, please share it with your friends and subscribe: