5 ways to judge (or misjudge) an election campaign poll
Between now and Aug. 2, you can expect to hear a lot about campaign polls — internal polls, external polls, internet polls, and so on.
When you read about these polls, you are going to need more than a grain of salt. Jimmy Buffett sang about his “lost shaker of salt.” Better find that lost salt shaker.
In the meantime, here are a few grains to take with your polls.
The most reliable polls are those you never see. Those who know,
Rule No. 1:
don’t say. Those who say, don’t know. The purpose of a poll is to give the campaign information that helps in making decisions about how to allocate scarce resources like money or the candidate’s time.
For that reason, polls have to be accurate and reliable. Critical decisions are made on the basis of that information. Bad polling leads to bad decisions. Bad decisions lead to bad results.
Media-sponsored polls are not more reliable than internal polls. In fact, they may be less so.
Unlike the campaign, a media outlet has no skin in the game. It doesn’t really
Rule No. 2:
matter to the press whether their poll is accurate or not. They are not allocating any resources based on the results. They are just generating a story about the horse race aspect of the election.
If the poll is right, they will tell you: “As we predicted ... . ” If their poll is wrong, they will describe the result as “an upset.” Either way, they get a headline and a story. That doesn’t mean that mediasponsored polls are all bad. They just require a high degree of skepticism.
Beware of any poll that gets leaked to the press (see Rule No. 1). Any internal poll that gets leaked to the press is leaked for a reason. The reason may be to encourage the supporters, discourage the opposition or raise money. In any event, a leaked poll should not be viewed as a real description of voter sentiment, but as an effort to influence voter sentiment.
Rule No. 3: Rule No. 4:
Remember that the campaigns have better information than you do. Watch what they do, not what they say. The campaigns are sending the candidate to the places their polling indicates that the candidate needs to go geographically and rhetorically.
Remember in November 2016 when Donald Trump was going to Michigan and Hillary Clinton was not? His polling suggested that he had a chance there.
Rule No. 5:
Watch for movement rather than numbers. What is the trend for a candidate: Is he or she moving up or down? But don’t compare apples and oranges. Their methodologies are different, and their samples are different.
Trends can’t be determined by comparing the results from two different polling organizations, like Mason-Dixon poll and PPP poll. It’s important to compare numbers from the same source, with the same methodology.
That is the layman’s guide to looking at the polls. There are a lot of good polling firms out there, doing quality work. There are also some that are slapdash, putting out a quick number that can be used for a purpose other than truly measuring voter sentiment.
Take them all with a grain — or maybe a whole shaker — of salt. As every candidate who is on the losing side of an opinion poll says: “The only poll that matters is the one on Election Day.”
John Ryder is a Memphis attorney who serves as Chairman of the Republican National Lawyers Association. He previously served as General Counsel to the Republican National Committee. He can be reached at ryderontheriver@gmail.com.