The Greek cynic Diogenes once roamed the streets of Athens carrying a lantern, supposedly in search of an honest man. Like Diogenes, minus lantern, I have been trying to determine if the presidential election polls are honest. Even though many of these polls use turn-out models to adjust the percentages of various voting blocks, focus on likely voters, and try to keep their surveys within a small statistical margin of error, they often differ quite a bit. The best you can get is a trend, as in Trump up/Hillary down or vice-versa.

The problem with any poll is you have to trust what the voters are telling you. While I certainly do not believe every voter polled is lying, I do believe some of them have more reason than ever before to shade the truth, and I believe I have proof to support my hypothesis.

This started about a month ago during a conversation with a Trump supporter. I asked him if he was going to put up a Trump yard sign. To paraphrase his reply, he said there was no point. It [the sign] would get torn up or worse. I then began to wonder if other people felt the same way.

The only way to get a handle on this, apart from doing my own poll, would be to take a look, which might be more difficult than taking a poll. Fortunately, in the past three weeks I have had to make five trips by car to five separate and widely spaced destinations. So, I decided to count the number of Clinton/Trump/Johnson/Stein bumper stickers and yard signs I saw along the way.

Before I get to the results, let me give you some background information. The mileage for all five trips, there and back again, came to approximately 1,860 miles, which took about 32 hours of driving time. The routes included parts of two red states, a blue state, and a battleground state in the Midwest, plus parts of four major metropolitan areas, and some travel through three college towns. The three longer trips involved significant time on interstate highways while the shorter two were mostly on state roads. All involved travel on city streets at some point.

The results were truly surprising. I was expecting to see a combination of maybe 40-50 Hillary signs and bumper stickers and maybe half to two-thirds of that for Trump, but my estimate was not even close. Altogether, I saw two Trump yard signs, two Hillary signs, and one Bernie Sanders bumper sticker. I did not see a single bumper sticker for any of the four active candidates. I mean zero.

Obviously, my random study has no statistical relevance. However, my memory on bumper stickers and yard signs goes back to Reagan vs Carter, and I cannot remember a single presidential election from there to this point where you would be unable to find at least a small handful of stickers or signs, even if you were just casually driving around town.

Okay -- so what does this mean, if anything?

There are several different possibilities. One of which is voters simply do not care enough to display a preference. This is unlikely due to polls and other reports indicating high interest in this year’s election.

Another possibility is voters are so unhappy with the major candidates they simply do not wish to identify with them. This has some credence as the Real Clear Politics (RCP) Favorable/Unfavorable poll average shows both Clinton and Trump with double-digit negative ratings. Yet, according to the latest RCP numbers, Hillary has a 42% average favorable rating and Trump has a 38% favorable rating. This means roughly 40% of the electorate likes both together. If this is true, where is the signage to back those numbers up?

This brings me to the third possibility I picked up on during my earlier conversation with the Trump supporter. Voters may be afraid to identify with the candidates. This sounds a bit far-fetched, but it could be true.

In Trump’s case, some of his policy statements, plus his off-message comments and gaffes have allowed the media to brand him as a Nazi, racist, homophobe, etc. So, if you place a Trump sign in your yard or a bumper sticker on your car, you get smeared by the same media narrative. Add that to the images of Black Lives Matter protests, leftists harassing Trump supporters at rallies, attacks on police officers, plus a string of home-grown jihadi attacks, and you can see why people might want to avoid identifying themselves as a Trump supporters.

Surprisingly, Hillary has a similar problem. If you identify as a Hillary supporter you get to wear her mantle of unfavorability; weighed down by the email scandal, the Clinton Foundation issues, and Benghazi, to name a few. Also, it is possible the media has vilified Trump supporters to the point where Hillary’s people are afraid to put a sign in their yard or a sticker on their car for fear of being set upon by a horde of peasants with pitchforks, or fear of simple vandalism.

The most reasonable conclusion I can offer for the paucity of election signage is voters on both sides have gone into stealth mode out of fear and/or personal discomfort of being associated with either major candidate. Even if the risk is small that you could be harmed in any way for your political affiliation, why make yourself a target? The same goes for public condemnation. Why endure scorn for your choice if you don’t have to?

The big unanswered question is, has reluctance to identify with a candidate for either reason carried over to the polls? Let’s imagine you are an average voter who has been asked to participate in a poll. Would you really be honest with a total stranger? Would you feel comfortable telling the pollster, or anyone else, who you intend to vote for if you thought your choice was unpopular? Would you risk telling the truth if you thought that information could be tagged to your phone number and shared outside the polling organization? You might feel it is easier to shade the truth and not worry.

The easy out for anyone who is uncomfortable with declaring they are for Hillary or Trump is to claim to be undecided, or pick one of the third party candidates, either Johnson or Stein. Gary Johnson cannot win, but he comes across as non-threatening and honest. Thus, he may be the perfect hiding place for college-educated Republicans who are Trump supporters, and moderate Democrats who are going to vote for Hillary, but don’t want to admit it.

I cannot prove a stealth vote for Trump or Hillary exists. However, the signs, or lack thereof, say something else. Most of the folks in the houses and cars I passed on the highway are going to vote for someone. They are just not comfortable saying who.

As the television character Dr. House was fond of saying, “Everybody lies.” But in the voting booth they tell the truth. If a stealth vote exists, we will find out on November 8th when, at last, we get an honest poll.