Wouldn’t it be nice if there were a magical way to invest the cash in your IRA or 401(k) and know the return you will get in advance? When it comes to equities, returns are extremely difficult to predict. That’s because returns on stocks or stock funds can vary widely from year to year. If the only data you can find when comparing such investments is historical, how can you go about making a rational choice from among the thousands of mutual funds and ETFs available to you?

A good starting point is to understand the asset class to which each fund belongs. Wikipedia defines asset class as “a group of securities that have similar financial characteristics, behave similarly in the marketplace, and are subject to the same laws and regulations.” Because funds in different asset classes have very different investment characteristics, you can’t really compare them at all. A fund that invests in U.S. small company stocks, for example, is not likely to behave in any way similarly to one that invests in municipal bonds. So before even thinking about specific investments, you should first determine in which asset classes you would like to invest and how much of your overall portfolio should be allocated to each. And that decision should be based on how much you need to grow your savings in order to support all your future goals balanced against your own personal tolerance for investment risk. In other words, you need a plan for your future before you should even get started picking investments. Surprisingly, not many people actually do that.

But back to investing. Once you’ve made the asset class allocation decision, it is now possible to compare funds within each asset class. Since past performance is such a poor predictor of future returns, is there any other measure that might be useful in helping you choose between funds? You might consider the variability of the fund’s returns over time, as measured by its standard deviation. This is a common measure of investment risk because funds with smaller historical standard deviations have been less volatile (and consequently more certain of generating the returns their investors expected) than funds whose deviations are higher. As with performance, while there’s no guarantee that historical standard deviations can predict future standard deviations, this measure does tend to be more relatively consistent over time than returns.

Even better, there is a measure that integrates both risk and return. It’s called the Sharpe ratio, after Bill Sharpe, one of the originators of the Capital Asset Pricing Model for which he and others received the Nobel Prize in Economics. The Sharpe ratio divides an investment’s historical return (actually its excess return above what’s called the risk free rate) by its standard deviation to measure its risk-adjusted return. The number will tell you two things. In absolute terms, if an investment has a Sharpe ratio greater than 1, it means that the investor is being adequately compensated for the risk he/she is taking. If less than 1, you might conclude that the investment is too risky to be worth the return it has generated. When comparing different investments, the higher the number, the better the fund has performed relative to its volatility or risk.

For example, here are two mutual funds that invest in diversified emerging country stocks: American Funds New World fund (NEWFX) and JPMorgan Emerging Markets Equity fund (IJPIX). (Note that I do not hold either of these funds nor am I endorsing them in any way. I found them simply by searching through Morningstar.com.) The Sharpe ratio for NEWFX was 0.42 over the last five years while it was 0.20 for IJPIX. That means the return you got for the risk you accepted was better for NEWFX, but not really justified for either fund. The same was true over the last three years, where the difference between the two funds was even greater.

A variant of the Sharpe ratio is the Sortino ratio, named for Frank Sortino but actually created by Brian Rom. It takes into account only the downside volatility (or semi-deviation) on the grounds that investors really only care about the risk of lower returns. It also allows a target return to be specified. When the average historical return of an investment is close to its median return the two measures provide similar results. But when there is more skewness in the performance of an investment over time, the Sortino ratio can give you more insight into the downside risk.

Both the Sharpe and Sortino ratios are reported in Morningstar.com. Although there are other criteria I additionally utilize to determine whether or not a particular fund deserves inclusion in a particular investment portfolio, these two measures are much better at indicating which of several funds might be the better choice than simply looking at past returns.