Skip to content

Ditch Your Polls

January 24, 2013

Another post from my company blog.  New content this weekend!

Small Campaigns, Big Ideas is a group of blog posts that aims to educate operatives and candidates on what works and what doesn’t work in small-scale elections.

These days, public opinion polls are an ingrained aspect of our political process.  At the very least, they give us a sense of the direction that an election is moving in.  At their best, they provide lawmakers with crucial information about how the general public feels about policy or the state of their respective district, state, or country.  Some pundits and polling advocates even suggest that they can shape the results of elections.  Polls seem like an asset in campaigns.  So why am I telling you to ditch your pollster during campaign season?  Because those polls are expensive, time-consuming, and worthless.

Your election is small, local, and likely not at the forefront of the minds of your constituents.  That’s just a fact.  The drop-off of interest after presidential and state-wide elections is tremendous.  Sure, a poll can slightly boost your name recognition insomuch that the people selected for the poll will hear your name once, but is that really worth the cost, especially when you could spend the time and resources on delivering a message to these people?  The quicker that you realize that opinion polling in small elections is more about vanity than results, the sooner you will save your campaign budget from taking an insurmountable hit.

We saw a perfect example of this in the last election cycle.  We had a guy that we were circling for some paid work for almost the entire election.  The first time we talked, he mentioned that he had run a poll in an attempt to gauge his opponent’s approval rating early on in the campaign.  That would have been fine if our candidate was flush with cash and simply attempting to gauge rather or not his challenge would be worth the time and effort.  Unfortunately, he was down 10:1 in fundraising and he ran the poll after he had already started the campaign process.  So he essentially flushed 10K down the toilet to figure out that the people in his constituency didn’t really like the guy they had elected.  The problem was, his opponent had a ton of cash and just smeared our guy relentlessly the last 2 weeks.  That 10 grand would have been huge in mounting a comeback, but he was broke and reeling down the stretch.  He lost by half a percentage point.  If he had had something, ANYTHING, in those last few weeks, he would have won.  Money for a commercial, a website debunking the smear rumors, a mail piece, Google or Facebook Ads, anything could have pushed 200 votes in his direction.  But we didn’t even end up getting paid our measly hourly rate for our design work or opp research.  The research we did would have blown his opponent up, but once again, he didn’t have the money to get it out.

There’s also bigger issues with polls besides the fact that the really nick your campaign bank account.  In local races, the math can be terrible.  As a standard rule of thumb, a poll needs at least 1,000 respondents to achieve a margin of error of 3.  In other words, when a poll conducted with sound methodology surveys 1,000 individuals, 95% of the time the results will be within 3 percentage points of where public opinion truly sits.  But in a constituency of 4-10K voters, good luck getting to 1,000 likely voters with a public opinion poll.  Suppose you can get 3 or 4 hundred to answer the poll (much more likely).  You’re margin of error then moves to 5-6 percentage points at best.  Is it really worth it to empty your bank account when the results aren’t even within a margin of error that you can trust?  How many local races aren’t decided by 6 points or fewer?  And even if you do run a good poll and get good information, what are you supposed to do with it if you’re broke?  What if the poll tells you that you’re down 4 points? You’ve paid a hefty price to be told that you are likely to lose without more groundwork and money.

A lot of people talk about polls keeping candidates on message or shaping public opinion.  But like I said earlier, no one really cares about your local race.  Not to say that there don’t exist engaged voters in small elections.  But a Gallup poll that shows Obama over Romney is going to register with far more people than a local internal poll detailing a state senate race.  In fact, good luck getting local news outlets to even publish those results.  A good portion of local outlets stay out of smaller races for one reason or another.  There’s a good chance that your constituents will never see the results or that they will dismiss them.  In regards to a candidate’s message, the fact of the matter is that you should know your message for a small race from day one.  Deviating from your core beliefs that you are running on during a campaign just means that your campaign is being run poorly, plain and simple.  Small elections revolve around your plans to bring pork to your constituency and maybe a couple state-wide issues.  You’re not dealing with scandals, national budget or security issues, or tiny elements of your past being picked apart daily by media outlets.  You’re running on a few core ideas and your reputation.  If those things suck, paying someone a bunch of money to have people in the community tell you that isn’t worth much.

Is this all to say that polls are completely worthless in small elections? No.  there are no absolutes in politics.  If you’ve got more money than you could ever spend, then poll away.  But if you’re trying to run a lean campaign and still come out on top? Ditch the polls.


Leave a Reply

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

You are commenting using your 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: