I had never heard of a “near-miss solution” until I read an article on Mother Jones a while back called: “How the Simpsons Have Secretly Been Teaching You Math.” And yes… we are talking about that Simpsons – the long-running cartoon that was initially a short on the fledgling Fox network’s “Tracey Ullman Show.” It became a Fox network show in 1989 and has been quietly printing money since then.

While the entire article is about the level of math that the writers for the Simpsons seem to inject into their episodes the start of the post referenced an episode where it seemed as if Homer had solved Fermat’s Last Theorem. (If you want to dig deeper into Fermat’s Last Theorem see the end of this article.) The discussion point in the MJ article was that the solution that Homer put up on the blackboard – when run through your phone’s calculator – seemed to solve Fermat’s Theorem. But we all know that Homer isn’t that bright. The problem is that when you run the equation your normal calculator can’t take the solution out the required number of decimal points needed to see the problem with the solution on the blackboard.

The author’s called this a near-miss solution.

It looks right but it really isn’t.

The tool used for measurement isn’t precise enough to really show the answer. It shows AN answer that “seems” right – but it’s just not the “right” answer.

Employee Engagement Is Just as Imprecise

That idea of having an answer you think is right – or seems right because the tool used to do the calculation isn’t precise enough to get to the point where you either see the real answer – or the real error – sounds a lot like the whole issue of employee engagement.

We can measure a lot of the things that contribute to engagement. We can measure all around it but we can’t really measure “it” accurately because we don’t have a real definition and real set of criteria that have hard and fast measurement points.

We have soft criteria. We have multiple definitions. We have relative measures. We have measures that are temporally influence, economically influenced, correlated but not causal. We have a variety of surveys measuring everything from friends at work to number of people in your network who are engaged vs. disengaged.

All of this CAN show AN answer but I wonder if they are all falling victim to the problem of a near-miss solution?

Are human beings and how they react to and interact with others just too variable be boiled down to 5 things, or 3 things – or even 12 things? Are we asking our surveys and our mathematics to answer questions that can’t be solved for with the tools we currently have?
I ask because with all the investment companies have made relative to engagement we’re just not seeing the needle move to the degree you’d think it would. Every consultant can provide their answer to engagement with their proof. In fact, there is no shortage of solution sets – each including something different – that address the issue of employee engagement. Many different experts have weighed in and each has shown their solution “work.”

But let me ask this… when we have that many possible solutions do we really have any solution at all?

I don’t know.

I do know this however.

You cannot go wrong treating people with respect and treating them like valuable human beings. Listening. Responding. Validating. Recognizing. Thanking.
I don’t need to measure that stuff. It’s just the right thing to do – precision be damned.

Fermat and Homer – The Solution

Fermat’s Last Theorem states that while there are an “infinite number of solutions” to this equation: x² + y² = z², it is impossible to find any solutions (that are three different whole numbers) for this equation x³ + y³ = z³ or for this equation x4 + y4 = z4 or for any other equation with the same structure, but whole number exponents greater than 2.

On the blackboard there is an equation where it seems that Homer has done it (see image below.) According to the author of the post, if you do the math on your phone calculator (3,98712 + 4,36512 = 4,47212), you will see that what Homer wrote down is apparently the solution.