To Poll or not to Poll ……

10/17/2020

As curious people, we all love getting answers to questions and if we can get the answers ahead of time or before anyone else then so much the better.  Given this, it makes complete sense that we really want to know who is going to be elected this November 3rd, hence our reliance on the polls and our desire to believe polls.  However, our experience in the last 2 elections leaves us questioning our resolve and the accuracy of these vaunted polls.

How can we rely on the polls in the “Age of Trump”?  It was a forgone conclusion that Queen Hillary would be crowned our most high leader in 2016 according to the polls and to the dismay of some of us, the polls told us we would lose the House of Representatives in 2018.  The polls were 99% wrong in 2016, (one poll had it right), but in 2018 the majority of polls were correct – Who Knew?  What’s it going to be this time around, is the “Trump factor” going to come into play or like 2018 the Trump factor is now old hat and the polls are back on track again?  When listening to the pollsters and the pundits, always begin your review by understanding the motivations of the pundits or polling companies and what goes into the poll itself.  The two questions I try to understand is what does the polling company want to accomplish and what outcome is the company who hires the polling company want to see.  As a petroleum engineer, I was always looking to predict the most accurate oil or gas reserves possible and just as important the timing for recovering those reserves.  I genuinely believe most people look at it from this same perspective, namely “How can I determine the outcome with the highest accuracy.”  There’s a bit of competitiveness included in that and perhaps a bit of self-centeredness, wanting and thinking I can do it better than others.  It was quite a revelation when I realized there might be another motivation when dealing with the polls.  As the CEO of Moody’s admitted in an interview this week, the polls are overrepresented by registered Democrat voters, in part because there has always been more registered Democrats than Republican until lately.  But also, there is a bias towards wanting to influence the potential voter that their candidate is so far behind in the potential vote count that there is no point in voting, or for some voters it is a lack of desire to do any research and then they’ll just vote for who the popular person is, after all they reason, that many people couldn’t be wrong.  This is likely the motivation behind the organization contracting the polling company.  What’s the motivation behind the polling companies?  Clearly, with the number of polling outfits in existence they are competing for a limited number of $, so without completely sacrificing their integrity they will find ways to meet certain metrics and still give the appearance of legitimacy.  This provides an answer that probably fits within the outer limits of accuracy but still can be called reasonably accurate.  Keep in mind this doesn’t require any grand conspiracy theory or collusion among the various entities in order for this kind of polling suppression to have occurred.  

Motivation of the pollsters and contracting companies is one issue, another is the questions asked and how they’re asked within the polls.  These factors all contribute to an embedded bias in the polls.  Lately though we’ve been hearing of a different approach to the polling.  Recognizing there may be a “Trump factor” that results in people not being honest with the pollsters, one company – Trafalgar Group, is asking a different set of questions.  They are asking who the respondents think their neighbor will be voting for.  They are limiting the length of the polls recognizing that most people don’t have time to spend 15-30 minutes answering questions on the phone.  The ideas here are people will tend to project their opinions on their neighbors and, you can get a different set of people responding to a 2-5 min survey when they would just hang up if faced with 15-30 minutes of questions.  In 2016 Trafalgar successfully predicted Trump the winner as well as the House of Representatives flipping to the Democrats in 2018.

Given these questions, we could just ignore the polls altogether and guess, but in reality, we like to plan and prepare for what might lie ahead.  Move our cash into a defensive position in the market, move our 401K funds into a more aggressive posture, buy additional ammo or stock up on supplies, buy additional storage facilities for our business or maybe just move out of the country because we don’t like the results of the election.  For any of these scenarios we should try to anticipate what’ll happen next in order to properly prepare ourselves.  Polls are not the only thing to look at in order to gauge overall public sentiment.  We can always listen to the noisy few who protest and riot in order to attract the press for a story, but really those are truly a minority of people who, despite appearances to the contrary, are very well organized.  A better way of determining what might happen would be looking at the stock market.  The stock market does not like uncertainty plus it’s always focused on the future, so when you see it decline then there is probably a change in policy coming in the near future.  This is especially true with such a diverse outlook on how best to help the working class as our two candidates have expressed.  Another data point in considering how much to trust the polls, is asking the question made famous by Ronald Reagan: “Are you better off now than you were 4 years ago?”  During GW Bush’s 1st term as he was running to be re-elected in 2004 for a 2nd term, 47% of the people believed they were better off then, than 4 years prior.  In 2012, when Obama was running for his 2nd term, 45% of the people believed they were better off.  The answer to this same question, as of last week, showed 56% of the people believe they are better off now than 4 years prior, even while in the midst of a pandemic.  Look at a comparison of similar situations, past and present in specific battleground states, to give further clarity. For example, these same polls showed Hillary up 6 points in Wi, 8.6 pts in Pa, and 2.4 pts in NC in October 2016 whereas, VP Biden is up 5.5, 6.7 and 1.4 pts in each state respectively.  The consumer confidence is also resting comfortably at 81.2 signaling people are comfortable in their current situation.

Surveys aside, you can always look at the public gatherings of the political candidates and surrogates to get a feel for which way the winds are blowing.  When the news media chooses to show it, there are very large numbers of people showing up for the Trump rallies and according to the figures, there is roughly 18-20% of the attendees who are registered democrats.  This time around, (as opposed to 2016), there are significantly more voters registering as republicans.  What does all this tell you?  You probably can figure out which way I fall on the subject but listen to your own sources for information and if they are contradictory, then discard them and find new sources because the old ones are definitely doing you a disservice.

Remember, Polls represent a snapshot in time of what people might be thinking, if they are honest.  Watching them over time could show the trend of people’s attitude and the positive impact one candidate could have over the other candidate.  Or it could mean the voters are becoming more involved in the process as it comes nearer to the decision time.  Some people actually have a life outside politics unlike some of us.  So, ignore the polls at your own peril but do so with all the information available and remember to look at the evidence honestly.  The point of this article is to exercise what I’ve discussed in previous articles – Critical Thinking.  Whenever you are faced with competing ideas that seem to have legitimacy don’t let someone else tell you what to think, look at the information available and make your determination based on evidence at hand directly, but also indirectly.  AND have fun with it!!!!

4 thoughts on “To Poll or not to Poll ……

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s