Why The “But I poll better” defense is pretty much nonsense.

     On both the Republican and Democratic side, the candidate who received the most votes during the primary season is almost certainly going to be the party’s nominee. Both these candidates also have quite high unfavorable ratings (though Trump’s are clearly worse). This led John Kasich to run around the country citing poll numbers that showed him besting Democrats in head-to-head comparisons and arguing “the party will ultimately chose me” even though voters hadn’t. Bernie Sanders is now making a somewhat similar claim in the hopes of getting Super Delegates to flip.  

    Neither argument went, or is going, particularly well. There is a good reason for this. Both Kasich and, to a slightly less degree, Sanders are comparative unknowns. Thus both candidates’ polling reflects a preference for someone other than their opponent, rather than a strong commitment to them. This kind of soft support is not to be trusted. 

      In addition, arguments based on general election polling create a perverse incentive. If polling better were a good enough reason to overturn the voters then campaigns would have every reason to try to destroy their opponents.  The Trump and Clinton camps would be forced to launch horrible negative attacks to try to fix this problem. That would turn a process that often makes all the candidates unpopular into one that guarantees that result. 

     It could be argued that anyone who cared about winning as a party and not just as an individual would not go so negative even if it meant losing. But we’ve already demonstrated here just how far candidates’ selfishness goes in wanting to win for themselves.  Leaving that aside, candidates can always rationalize these attacks, insisting they are needed because the other party will certainly make them in the general election. The candidate assumes he or she has a duty to the party to show the downsides of a primary challenger rather than allow the opponent to be nominated based on the upsides.

    

    On the Democratic side, Super Delegates represent 15% of all delegates. Thus for a candidate to win without them he or she would need 59% of pledged delegates, which doesn’t happen when a race is remotely close. That means Super Delegates are often the deciders. Do we really want Super Delegates to play the “let’s see what the polling says” game? If we do, then we have to be comfortable with 714 people overriding the votes of up to 30 million people. And if we do, we might as well let these 714 people make the decision without conducting primaries at all. 

           Letting 30 million voters pick a nominee seems a much better idea than letting polling data, which can change on a dime with new information, be the arbiter. Polling better is a fine argument to make to the voters but if the voters don’t care (and they certainly didn’t for Kasich) then general election polling is not a leg to stand on, and it certainly should not be a hill to die on.