All this talk about a “wave” materializing in November has finally gotten this Long Islander to check the surf report.
Long Islanders know a few things about waves and swells. If you talk to some of my surfer friends, they’ll tell you that a wave is a small piece of a massive oceanic and meteorological story. The predictive indicator of “good waves” is often the “swell” which builds hundreds of miles away and can push up a coast to create truly epic surfing conditions or, in the case of Superstorm Sandy, complete devastation along a coastline. Tools are available to help predict when a swell is building or where a wave might break, but sometimes a surfer just gets lucky and is in the right spot at the right time. Surf’s up…or totally skunked.
Stu Rothenberg, Jay Cost, Nate Silver, Charlie Cook, and Larry Sabato have all spent considerable time in both print and television media discussing the likelihood of a strong Republican wave sweeping some otherwise weak GOP candidates into office. Granted, talking about these sorts of things is a political analyst’s stock in trade. Especially for numbers analysts like Mr. Silver, this is their busy time of year. A wave is building, it’s just that no one knows how big it will be and whether or not it will be large enough to “nationalize” the election and crash over local issues which might favor the Democrat candidates. The real headline is that there is a “slight breeze” coming up behind the notion that the Republicans are getting closer to a guarantee of winning the majority in the Senate. So why has this headline get pushed to the side? Is it as simple as the old journalistic maxim that a “sunny and 75 degree” weather forecast does not sell a lot of newspapers, or is there a more cynical reason afoot?
Writers and analysts turn to analogies as a short hand method designed to convey meaning to their readers or listeners. The shame of this practice is that members of the the media often go back to the well of really hackneyed analogies again and again. It betrays, as I noted in a previous article, a journalistic laziness that leaves fresh commentary by the wayside and instead relies on safe, predictable patterns of behavior which may not be the best way to analyze a completely new set of circumstances. As the surfers of Long Island will tell you: no two waves are the same.
The reticence of some to declare 2014 a “wave” year for the GOP is grounded, primarily, in the generic congressional ballot polling data that suggests a tie or slight lead to one party or the other. However, in looking for a wave to form out of this information one has to take note of the bigger picture – the swell behind the waves. The generic congressional ballot has consistently under-measured Republican support and over-measured Democrat support through the past three or four cycles. Mastering the intricacies of polling is better left to Mr. Silver but consumers of polling data have to ask themselves why polling a generic congressional ballot might be so problematic and what exactly analysts are analyzing if it is widely agreed that this particular sample of data is so hard to measure. The difficulty of measuring generic congressional ballots may have to do with the challenges of polling in general: how do you figure out who is a likely voter, who do you poll and how do you poll them?
Some in the media have tacitly acknowledged that the generic congressional poll is, essentially, useless. Philip Bump of the Washington Post dismisses the generic polls that are making their circuitous way around Washington parlors and bars these days by acknowledging the under-measurement of Republican support in previous years’ polls as well as the current data for 2014. Instead of dismissing the usefulness of the generic congressional ballot, why not instead ask why the generic congressional polls are not corrected to reflect a more accurate set of respondents and, therefore, more likely voters? This is commonly done in polling, and it can have dramatic effects on the poll’s accuracy.
Is it too cynical to suggest that this dismissiveness spawns from the accepted fact that, when Republicans tie or barely beat Democrats in the generic ballot question, it usually means a big Republican victory is at hand? Instead of trying to figure out if the wave will form or whether it will rush over places like New Hampshire and Michigan why not instead try to predict the swell? Pollsters know the ballot under-measures Republican support, and that has been well documented for several cycles. Why not fix the sampling or, short of that, use that fact to state in their endless analysis that a Republican wave is likely simply because the polling data is in line with previous instances of big Republican waves?
Similarly, the characterization of an election being a “wave” election can only be made after the wave has broken, or, on the day after Election Day. It is a subjective term and can be manipulated to change the meaning of what could be yet another historic election. The Republicans seem all but guaranteed at holding and expanding their majority in the House of Representatives and gaining at least six seats in the Senate, thus winning the majority. Barring any major shift in the electoral equation, this is the most probably outcome of this midterm election. If the Republicans do not manage to pick up eight or ten seats in the Senate, it would be simple for analysts to declare the water flat, without a wave having formed and by extension it being a “disappointing” or “underwhelming” night for Republicans. The table is being set now for the wave, or lack thereof, being the story and not the repudiation of President Obama and his policies which a Republican takeover of the Senate truly measures.
By October, the state of this midterm election may be more readily apparent and, as always, a major international or domestic crisis could scramble a year’s worth of polling and analysis. If the wave materializes, pushed by a heavy swell of anti-Obama sentiment, and yet again the generic congressional ballot poll fails to predict it, let’s not make the same mistake of paying it much heed in 2016.
This article was originally published on September 18, 2014.