If you have ever researched a mutual fund or paid someone to do it for you, there is a good chance that you’ve seen fund ratings from Morningstar. Much of the information on a Morningstar report is useful, but a recent study by the Wall Street Journal found that Morningstar’s much-heralded and quoted fund star ratings are of no help in predicting which funds will perform well in the future.

The shortcomings of the Morningstar star ratings system arise from its reliance on something called the Capital Asset Pricing Model (CAPM), a flawed academic theory. CAPM produces a statistic known as beta, a magic number that purports to precisely calculate the risk of almost any investment. Beware of false precision, especially in the case of something as subjective as risk.

The Morningstar ratings system is based on historic volatility and performance. An evaluation system based on past performance contains an inherent design flaw: recency bias. Extrapolating past returns doesn’t work with stocks, nor does it work with mutual funds – which are simply baskets of stocks. How predictive are the Morningstar ratings? The Wall Street Journal found that only 12% of five-star funds earned that top rating over the subsequent five-year period. An embarrassing 10% of five-star funds were awarded only a single star five years later.

The human tendency to favor recent winners, combined with a statistical system that mysteriously does the same thing, explains the seemingly contradictory finding that most mutual fund investors underperform the overall market. A separate Morningstar study looked at the decade’s best performing stock fund (up 18% annually), and found that the average investor in that fund lost 11% annually. How? By moving out of the fund following periods of poor performance, then back into the fund after it went up. That is, buying high and selling low.

If you shouldn’t rely on Morningstar ratings, how should you select funds within your 401(k) plan? I own individual stocks in my retirement plan, but here is how I would choose mutual funds if that were my only choice.

I am not recommending that you do this, because I don’t want to be sued if you aren’t happy with the results.

Discard your fund options in the bottom half of ten-year performance and those without a ten-year history. Then discard the funds with managers on the job less than seven years and those with more than 100 stocks in their portfolios. I would buy one or two of the remaining funds. Those funds have experienced, competent managers and have a chance of outperforming the market. A fund that owns 200 or more stocks is simply an index fund in disguise. If you want to own an index fund, buy an index fund. It probably has a 5-star rating, anyway.

David Moon is president of Moon Capital Management. A version of this piece originally appeared in the USA TODAY NETWORK.