To Nate Silver:
Congratulations to you. You have gained a lot of attention and are on the threshold of attaining even more. With these well wishes, from someone who has also achieved a good deal, be alerted to your new responsibilities. To date you have many fans. But the real scrutiny is just beginning and some fans are ephemeral. Here is some advice from someone who has been where you are today.
Don't Create Standards You Will Find Hard to Maintain Yourself. You are hot right now - using an aggregate of other people's work, you got 49 of 50 states right in 2008. I know how it is to feel exhilarated. I get the states right a lot too. But remember that you are one election away from being a mere mortal like the rest of us. We very good pollsters have missed some. They tell me you blew the Academy Awards and your projections in the 2010 U.K. elections were a tad squidgy. So be humble and continue to hone your craft. Be aware that some of your legions who adore you today and hang on your every word will turn their guns on you in a minute. Hey, I have been right within a few tenths of a percent - but you are a probabilities guy and even a 95% confidence level and a margin of sampling error are not enough for some.
Be Honest. You take other people's polls, compare records for predictions, add in some purely arbitrary (and not transparent) weights, then make your own projections and rankings. But your efforts include more than just statistics. There is some edginess: you don't like numbers when they don't agree with your preconceptions. One night you took a one sentence comment I made to a Boston Herald reporter that my partial poll results showed President Obama with about a "50-50 rating," and you headlined your piece, "The Worst Pollster in the World Strikes Again." Was that due diligence? Did you realize that less than a week later, as is often the case, most other polls were similar to mine?
Then there are your now famous rankings of pollsters based on 2008 presidential election "predictions." Aren't you even a little embarrassed that Zogby Interactive's last national online poll was published on October 4 - one month before the election? (And the numbers were right in the middle of the Real Clear Politics average). Or that the results of the several states we polled online were published October 21? Most states were at or near the RCP average. What point do you need to make by using numbers so long before an election? Those of us doing this work for decades understand that so much happens in the closing weeks, days, and hours of a campaign. As many as 4% to 10% of likely voters tell us they make up their minds the day of the election. Some of my colleagues suggest that you are being disingenuous when you knowingly use this data; others say you have a personal axe to grind. But repeating these errors over and over will not make them true. (Apparently though, you did rate my telephone election polls in 2008 as one of the three most accurate in a speech to Fordham University).
Understand That There's Much More to Being a Good Pollster. To the degree that we "predict" an outcome, the "prediction" must be based on results collected as close to the election as possible. But we pollsters are data-based problem-solvers. We work with clients to solve problems, plan the future, project trends, and test effective messages and models. This involves lots of people skills, a passion to get it right and do right by people who trust us. We are so much more than where we stand on election day. Your ratings come with and generate a lot of vitriol. How does that make our world a better place?
Appreciate Innovation. We have mastered polling by landline but today's landline penetration is where it was in 1963. We, at Zogby International, have been testing and perfecting new models and methods for years. We did well in our national online polls in 2004 and 2008. In both years, as well as 2006, we slipped in only a few states - but were right on the money in most- and we found where the errors were and continue to upgrade our sampling methods. We have had the courage to be the only pollsters to get it right (many times all alone) and to get it wrong (a few times). But remember, innovation is a necessity itself and an added value. Columbus never got to India. Roentgen studied physics and never expected to discover the x-ray. But they learned so much more. Interactive polling offers so many new opportunities for cluster sampling, image-testing, and reaching people we simply can't reach by traditional methods. You may want to take a little time and learn about interactive polling and the industry. Do Some Polling. Janet Maslin and Robert Bianco are terrific reviewers and critics but they never wrote or directed a movie or stage production. You are a statistician - a very good one - but you are not a pollster. You should conduct some polls and learn that the rest of us good pollsters survey people, not statistics. The numbers tell the story; preconceived ideologies and fuzzy-math statistical models do not.
Just some advice from an old warrior. A mutual fan suggested I take you to lunch some time. Good idea. Let me know when you'll be in Utica.
John Zogby is chairman of Zogby International and author of The Way We'll Be: The Zogby Report on the Transformation of the American Dream (Random House, 2008).