The urge to believe there is a magic formula for success, and that it can be deduced from studying past hits, is powerful. Strategic Communications Group CEO Mark Hausman believes he has distilled The 3 Hallmarks of Exceptional Content. Columnist Marcel Williams is convinced he knows The Essential Features of a Hit Record. Their two cents on the subject may be worth a nickel, but as Yogi Berra observed, “a nickel ain’t worth a dime anymore.”

These checklists, while almost intuitively obvious and rightly containing features correlated with past successes, fail immediately as predictors of the future by being neithernecessary nor sufficient. Their authors cannot prove that all the characteristics are necessary for success; i.e. eliminating any one of them will guarantee failure. (Mr Williams even admits as much.) Nor can they prove that their list is sufficient for success; i.e. a work possessing all these characteristics may still be a failure for lacking other elements.

The lists highlight statistical regularities—good for documenting the past, lousy for predicting the future. In The Financial Regulators’ Dilemma economist Edwin G. Dolan illustrates how statistical regularities do not always reflect causal relationships using the so-called Nickels Paradox:

Suppose the Federal Reserve observes a strong past correlation between the number of nickels issued by the U.S. mint and the rate of inflation. Does that mean restricting the issue of nickels would be a sufficient instrument to control inflation? Of course not—not if pennies, dimes, bank balances, and money in all other forms were issued in the same quantities as before. All that would happen is that the previously observed correlation of nickels with inflation would disappear.

How does the Nickels Paradox apply here? Suppose all music producers take Mr. Williams advice to heart, and 100% of future tracks can be characterized by: catchiness, timeliness, a strong vocal, and accomplished performance. There would still be only 100 releases on The Billboard Hot 100 in a given week, and the vagaries of public taste would still decide who gets there. Or, taking Mr. Hausman at his word, suppose all future web content has: a unique perspective, great “get” and/or brilliantly constructed prose. All that would happen is these would cease being differentiating features in attracting readership.

Don’t stop critiquing your past work, learning from others, applying rigor and discipline, or nurturing creativity in your quest for greatness. Just don’t mistake performance metrics forpredictive ones. And use that advice for what it’s worth.