There's a major problem with Modern Portfolio Theory. Don't get me wrong. The mathematics are elegant and the statistical mechanism is downright brilliant, but, truth be told, the darn thing just doesn't work. Worse, it's probably misled people more than it's helped them. We can blame its reliance on standard deviation as the definition of risk as the root cause of its failure. In fact, the whole concept of the relevance of risk creates a fallacy that has so often led to dissatisfaction among investors.

How does this manifest itself for 401(k) plan sponsors? These fiduciaries often hire other professionals based on the service providers keen understanding of all those fancy investment statistics excreted from Modern Portfolio Theory formulae. These statistics purport to quantify some facet of risk and are almost always based on the standard deviation (or its related kin, variance), of a universe of returns. That this universe uses real return data only reinforces the apparent credibility of these numbers.

Why caution those who use standard deviation as a proxy for risk? Disregarding the fact standard deviation only works in normal distributions and the distribution of real returns is far from normal, the very trait it measures is irrelevant to most investors. In the simplest of terms, it measures the consistency of events, in this case, returns. A return universe with a small standard deviation offers most consistent returns – and thus is defined as "less risky" – than a return universe with a large standard deviation.

But does this make sense? Let's return to our pigskin metaphor for a moment. Pretend there's a coach of a team on the cusp of the Superbowl. The only ingredient missing is a consistent field goal kicker. He's got two in camp. He measures the standard deviation of each kicker. The standard deviation will give him a range in which two-thirds of the kicker's field goal attempts will fly. The first kicker has a standard deviation of six feet. This compares very favorably to the width between the field goal posts (18 feet and six inches). The second kicker has a standard deviation of twenty feet, meaning it's wider than the width of the field goals.

The coach cuts the first kicker and signs the second kicker. 

"What?" you ask incredulously.

Well, it turns out that, although he's more consistent, the first kicker is less accurate. His average kick is wide right by three feet, meaning he only has roughly a one-in-five chance of scoring. In contrast, the second kicker's average kick is dead on – it splits the uprights – meaning he has roughly a three-in-five chance of scoring.

The coach cannot use the standard deviation in a vacuum. He needs to know two other data points: the average kick and the size of the goal posts 

Likewise, a 401(k) investor doesn't need to focus on risk (or volatility or standard deviation or Sharp Ratio or whatever you want to call it). The 401(k) investor needs to focus on the goal and what needs to be done to get there. How much money is he starting with? How much is he contributing annually? How much does he need to retirement at his desired lifestyle? What is his Goal-Oriented Target?

None of these questions involve risk or volatility. They involve real numbers, three out of four which can be controlled with precision by the investor.

In the end, these are the only numbers 401(k) investors should pay attention to. Any undo emphasis on "risk" or "risk tolerance" may only lead to desperate employees throwing a Hail Mary with their 401(k) investments late in their career.