Analysis: Why pollsters have a Labor blind spot

One gets the impression over the radio that they're suffering from collective trauma.

Barak smiles 298.88 (photo credit: Ariel Jerozolimski)
Barak smiles 298.88
(photo credit: Ariel Jerozolimski)
Listening to Israeli pollsters on the radio over the last few days on the subject of the Labor primaries, you get the impression that they are suffering from collective trauma and have all but lost confidence in their own profession. An interviewer only has to ask them what they think might emerge from the balloting Monday and he's instantaneously covered in a pile of caveats. Of course, we're told, polls don't predict, it's not an exact science, it only indicates trends and proportions and precious little else and it's especially fickle when it comes to trying to make head or tail out of what the Labor members are thinking. This is no laughing matter. The pollsters' credibility is really at risk. They exaggerated Kadima's lead in last year's general elections, and missed the surprise showing of the Pensioners Party and the near-obliteration of Netanyahu's Likud. But the biggest recent failure was that none of the pollsters foresaw Amir Peretz's razor-thin victory in the last Labor primaries a year and a half ago, all instead buying into the illusion of yet another Shimon Peres "victory." As much as it might be fun to ridicule the pollsters, there are objective reasons for their Labor-blindness. It's virtually impossible to sample a group of 100,000 members that changes annually, losing tens of thousands who stop paying their membership dues and gaining replacements who are somehow persuaded to join. How could any research company monitor the shifting demographics and changing tribal loyalties in this disparate group? But still they make the effort, and as usual find comfort in the company of their rivals. That way, if the pollsters are wrong, at least they are all in the same boat. This time, too, none has gone out on a limb with an unorthodox prediction. All have similar results: Ami Ayalon and Ehud Barak polling a bit more than 30 percent each, with Ayalon slightly in the lead; Amir Peretz trailing them on something between 15 and 20%; Ophir Paz-Pines trying to scrape together double-digits with little success, and, somewhere, a fifth candidate we seem to have forgotten. So what are the chances of an upset? Since turn-out is expected to be no higher than 60% (according to the polls again), relatively small numbers moving from side to side - a couple of thousands, even a few hundreds - could have a major effect. In this scenario, good organization and old loyalties should play a part. This doesn't bode well for front-runner Ayalon, less well established than his rivals. A key question, then, is whether Peretz's well-oiled apparatus will reassert itself and perhaps give him a shot at the second round. Barak, backed by most of the ministers - chief among them Binyamin (Fuad) Ben-Eliezer - should also notionally enjoy a solid base among their supporters. But Fuad has been looking rather wobbly lately. And if he was confident of delivering the goods, wouldn't he be running himself? Ayalon might suffer from a lack of organization and an established base but he also enjoys a relative absence of hostility, especially when measured against Barak and Peretz. He would probably get most of the anti-Barak and anti-Peretz vote in the second round. So it may be that some of those voters who are now telling the polls they support Paz-Pines's principled but hopeless campaign might, at the last moment, prefer to ensure Ayalon makes it into round two, rather than seeing a dreaded Barak-Peretz showdown. This fear could even give Ayalon the necessary surge to the 40% threshold and a first round victory. On the other hand, many of his potential supporters might vote for Paz-Pines out of sympathy in the first round, confident that they can switch to Ayalon in the second, triggering an even larger upset. And we have yet to factor in the alleged forgery of ballots already being planned in the Arab sector. Looks like the biggest surprise will be if the real results actually mirror the polls.