A belief, once firmly held, is hard to change, even when the evidence and data overwhelmingly refute that belief.

Take male hand size.   We all know it correlates with…lower fumble rate among NFL Quarterbacks (why, what were you thinking?)

Every year at this time the NFL evaluates college players and their pro worthiness with something called the “combine”.  One of the more time-honored metrics employed beyond the player’s speed at the 40-yard dash and bench press repetitions is the measurement of a Quarterback’s hand size.

Why?  Because It is firmly believed that larger hands correlate with lower fumble rates by Quarterbacks.  Professional scouts whose job it is to evaluate and rank NFL talent have been quoted as saying hand size is more important than arm strength.  Their opinion impacts draft position, which, in turn, impacts financial futures to the tune of millions of dollars.

There’s just one problem.  This belief is complete and utter nonsense– mythology perpetuated by anecdote and the unflappable confirmation bias that causes us to discount anything that runs counter to what we already believe.

Based on a huge dataset of NFL quarterback hand measurements and actual fumble rates, there is a near-zero correlation.  In fact, there is a much higher correlation with the number of letters in the Quarterback’s last name and fumble rate than hand size and fumble rate.

To add insult to measurement, there is a physiological link between hand grip strength and fumble rate (but hand size and grip strength have no relationship).  And, there is a $30 device called a dynamometer that has been around for decades that measures grip strength with extreme precision.

This boils down to one inextricable fact: the NFL, which is a $15 billion dollar business, has been making a choice about the most important position on the field (Quarterback), based in part, on a measurement that doesn’t measure anything.

The Fundraising Equivalents to Hand Size

What is the fundraising/marketing equivalent of beliefs that simply won’t die because of availability bias (forming beliefs based on skewed, limited experiences) and confirmation bias (ignoring data that runs counter to existing beliefs?)  Here are but a few.

• The more we ask, the more we get. We’ve written about this until we are blue in the face.  You can find it here , here,here, here and here.
• Donors are all the same. This isn’t spoken but look at how most of the sector operates and it seems to be the default, unspoken assumption.  One-size-fits-all treatment.  Or, the random nth test that implicitly assumes that everyone in the test group is the same and that our treatment idea and its effect is so strong that it will outweigh all the inherent, fundamental differences between the people sitting your “they are all the same” test group.  This is part of the of the reason most tests fail – different reaction within the test group to your idea that gets lost in the average.  Said differently, lots of false negatives – i.e. failure that is in fact hidden success (with certain subsets within the test group).    Of course the other reason tests fail is the lack of rigor in the thinking – random never works except randomly.
• That “statistical significance” is some sort of gold standard. There is an enormous amount of noise in any direct marketing test that significance testing is not sensitized too.  Make the sample large enough and any difference, however small, can be flagged as significant.  One of our colleagues ran a 15-way split test in the mail.  There were 5 different “big ideas” each with three variations to execute on said idea.  It mailed, results came back and there were big, statistical differences across the 15 treatments.  The agency involved had an instant narrative to fit the results and leaned heavily on this all be real because, you know, it’s statistically significant.  The problem?  At it turns out, there was a rather major screw up along the way and everybody got the exact same mailing.    Lots and lots of noise.  Especially if the test ideas are random.
• That people give because of emotion.   Emotion is the goal, not the cause.  People do not give because we made them sad in the appeal.  They give because if we made them sad, their goal is to not be sad.  If they believe (typically based on past experiences) that giving is the best way to feel better they will give to achieve that emotional goal.  But, just making people feel something with a solicitation is one hand clapping – at best.
• Generational differences are real and important. Another blue in the face moment here.  We’ve debunked this generational mumbo jumbo here, here and here.  But, frankly it is Don Quixote tilting at windmills.

The late Senator Patrick Moynihan once quipped that “everyone is entitled to their own opinion, just not their own facts.”  It seems folks have their own limited “fact” base that forms an opinion and that opinion is unmovable despite overwhelming, counterfactuals.

Hell, maybe the hand size of the organization’s CEO  correlates with response rate…

Kevin