As the 2020 election nears we are constantly being bombarded with polls, telling us who’s winning and who’s losing. It can get very confusing and downright annoying. I owned my own Democratic political consulting business for twenty-five years in Annapolis, Maryland, and I am going to try my best explain polling, so it makes sense and doesn’t drive you crazy. This election is already extremely agitating, and we don’t want to be fooled again, like in 2016. So, let’s get it right this time.
We will begin with the 2016 election. All my friends refuse to trust the current polling data that shows Biden comfortably ahead in the popular and electoral vote count because, “They got it all wrong in 2016. The pollsters said Hillary Clinton would win and she didn’t. Why should this time be any different?”
Well, because the pollsters didn’t really didn’t get it wrong in the last presidential election. Clinton won the popular vote by over two million votes. But she lost by razor thin margins in three key states she was predicted to win: Pennsylvania, Wisconsin, and Michigan.
Every poll has a margin of error — usually something like plus or minus 3-4 percent. And the Rust Belt states where Clinton lost, easily fell within the margin of error.
- Pennsylvania – 1.2%
- Wisconsin – 1%
- Michigan – 0.3% (11,612 votes!)
All pollsters are not created equal. Many work directly for one of the political parties and they are not for real. In fact, their sole purpose is to confuse and paint a false picture. To help sort out the good, the bad, and the ugly, I went to the King of the Pollsters, Nate Silver at 538, who has assigned a grade and a bias score for virtually every political polling company in America.
Here’s the pecking order of the phoniest pollsters on the market, along with who they shill for:
- TCJ Research (R)
- Strategic Vision LLC (R)
- Research 2000 (D)
- Blumenthal Research Daily (R)
- Pharos Research (D)
- Big Data Poll (R)
- Overtime Politics (D)
- KG Polling (D)
- CSP Polling (D)
- Our Progress (Online)
And let’s not forget the 49 companies that may or may not be biased, but either way, generally suck at polling because they do not follow the standard protocols or cut corners:
- Survey Monkey (D)
- Google Surveys (Online)
- Lucid (D)
- ccAdvertising (R)
- McLaughlin & Associates (R)
- Swayable (Online)
- Critical Insights (Online)
- Brown University (R)
- Humphrey Institute (D)
- Nielson Brothers Polling (D)
- Datamar (R)
- Targoz Market Research (D)
- Baydoun Consulting (R)
- Point Blank Political (Online)
- Fort Hays State University (R)
- Target Insyght (R)
- St. Leo University (D)
- Carroll Strategies (R)
- Hampton University (R)
- Elucd (Online)
- Rand – American Life Panel (Online)
- Brigham Young University (D)
- Indiana University-Purdue University Fort Wayne (Live)
- University of California Berkeley (R)
- Zia Poll (R)
- Dane & Associates (R)
- TP Research (R)
- Glascock Group (D))
- CPEC (R)
- Saguaro Strategies (D)
- BWB Global (R)
- University of Colorado (D)
- Baldwin Wallace University (D)
- GaPundit.com (D)
- Zata3 (R)
- ResearchNow (R)
- MarblePort (R)
- CrossTarget (R)
- Jayhawk Consulting (D)
- Great Lakes Strategies Group (R)
- New Frontier Strategy (Online)
- GCR & Associates (Online)
- Slingshot Strategies (D)
- Strategy Research (R)
- Rossman Group (D)
- Triumph Campaigns (R)
- Liberty Opinion Research (Online)
- Iona College (D)
- co/efficient (D)
- TeleResearch Corp (Online)
And then there are the 373 polling companies — which seems like a helluva lot to me — who, on average, do an okay, nothing special, job. Many like to use rapper spelling for their company name or the words: Strategies, Research, and Associates. And both parties like to engage in this fairly sketchy exercise in spinning public opinion. Most polls are essentially crap.
Keep this in mind: poll is an old German word that first pops up in the English language in the 13th century, and its original meaning was derived from the word “poleaxe”, an axe used for striking at the head and knocking people to the ground with a sudden blow. I would suggest the current meaning of the word has pretty much stayed true to form.
So, it pays to know who the good polling companies are, and that’s easy because our straight-shooting buddy Nate Silver says there are only fourteen who get a solid grade of A, in the following order:
- Selzer & Company
- Muhlenberg
- Sienna/Ney York Times/Upshot
- Marist College
- ABC News/Washington Post
- Monmouth
- Survey USA
- CBS News/New York Times
- Sufflok
- RKM Research
- Emerson College
- NBC News/Wall Street Journal
- Sienna College
- Fox/Beacon/Shaw
All the other polls should be thrown into a big hat and are part of what I like to call the “bullshit aggregate“. In other words, an average of a broad spectrum, good and bad. And we all need to be very careful when it comes to figuring out what it truly means when a poll — even one from a reputable outfit — says that Joe Biden is winning the presidential race over Donald Trump 50.3% to 43.1%. Because there’s a trick to making sense of those numbers, and in my next blog, I will teach you how to master the polling game.
