Before you buy a fund, you should know what it will do to the overall risk of your entire portfolio of funds. Here’s a canny performance measure to help you judge.

It’s a fundamental principle of investing that raw performance doesn’t tell you how good a stock picker is. You have to know how much risk he took to get that return. In a bull market a subpar player can have a terrific gain just by buying on margin–or buying high-risk stocks. In the late 1990s some mediocre Wall Streeters looked for a while like geniuses just by going heavily into Nasdaq stocks. You know what happened to them.

This principle of comparing risk and reward is enshrined in the various rating systems for mutual funds. This magazine, for example, grades funds separately for bull- and bear-market performance; if a seemingly hotshot fund boasts of a terrific ten-year gain but gets a FORBES D or F in down markets, then it was taking big risks with customers’ money and doesn’t deserve any of yours. Morningstar rates funds with a formula that penalizes gamblers and rewards the steady performers. A similar methodology is incorporated in the widely used Sharpe ratio, named after Nobel economist

William Sharpe

. Sharpe divides a portfolio’s excess return (return less the riskless T bill return) by its volatility.

Now we’d like to add to your arsenal of fund-picking tools a less well known measure called the Treynor ratio. Its particular virtue is that it puts a fund’s risk in context. It attempts to answer this question: Does a chosen fund deliver good performance relative not to its own volatility, but relative to the volatility it is likely to add to your overall portfolio? The results are (on occasion) a bit surprising. Shopping for a Treynor ratio winner may give you a fund that jumps around a lot by itself but probably won’t make your portfolio jump around. Don’t use Treynor numbers if you intend to own just one fund. Do use them if you have a diversified collection of funds and/or stocks.

Example: State Street Research Global Resources. All by itself, it’s risky indeed. One quarter it’s up by 9%, the next it’s down 14%, then it’s back up again. Naturally the State Street fund gets faulted for riskiness. Morningstar gives the A shares just two stars. Investors scared away by that rating, though, would miss some nice performance. The fund is up 4.2% this year and an annualized 13.8% over the past three, beating most others in its category (natural resources) and the S&P 500.

There’s a reason for the energy fund’s good results during a bear market and, if you can stomach its 5.75% front-end load, for adding it to a mix of other stock market investments. Namely, when oil prices shoot up, the fund’s energy stocks are likely to do well, just when fears of inflation and depressed corporate profits damage most other stocks. In other words, a fund like this is a nice counterbalance to a fund that owns Cisco Systems and Delta Air Lines. especially if you can buy it with no load through your re-tirement plan. Gold stocks (or funds that own them) would offer similar benefits. It would be a big mistake to own a gold fund and nothing else, but there’s something to be said for putting 15% of your equity money in gold and the rest in mainstream funds.

The Treynor ratio is named for

Jack L. Treynor

, one of the fathers of modern portfolio theory and for years editor of the Financial Analysts Journal. (At 72, he is still a trustee of several Eaton Vance mutual funds.) Treynor’s formula divides a portfolio’s excess return by its “beta.” Beta is the widely used measure of market-related risk in a stock or collection of stocks. If a stock has a beta of 0.5, it tends to move up (or down) with the market but only half as far. Cisco has a beta (as calculated by Value Line ) of 1.45–meaning it’s likely to go up 14.5% in a month when the market climbs 10%.

Note: Beta is not, as is so often claimed by naive market-watchers, a measure of volatility. Newmont Mining has one of the lowest betas in the Value Line universe (0.35), but this gold stock is no low-volatility stock. Quite the contrary. The low beta simply tells you that Newmont does not march to the S&P 500′s drummer. This is a useful trait if you want to offset the market’s gyrations.

The idea behind the Treynor ratio is that market risk–measured by beta–can’t be diversified away through investing in a panoply of funds, and so ought to be penalized. Treynor adherents shrug off that part of a particular fund’s volatility that’s not related to the broader market, since you will overcome this risk by diversifying.

Yale School of Management professor

William Goetzmann

suggests using Treynor ratios this way: Look at funds category by category (small-company value, small-company growth, large-company value, etc.), and go for one of the high scorers in each category. A further refinement is to avoid expensive funds, no matter how nice their performance measures. The table shows high Treynor scorers with reasonable expenses in various categories.

One is

James Gipson

‘s Clipper Fund, which we featured in our Feb. 4 issue. Gipson has done well betting on tobacco-tainted Philip Morris and now looks for a Tyco comeback. FMI Focus, a champ in small-company growth highlighted in our Sept. 16 issue, has made a bundle in retail with the Jos. A. Bank men’s clothing chain. Other hits include Tollgrade Communications and Quest Diagnostics .

You can get Treynor ratios for funds on Yahoo’s Web site or calculate them yourself (see box ). One big caution before you dive into any fund-rating formula: Past performance is no guarantee of future results. That oft-heard disclaimer is an understatement. Any fund-rating system is based on historical performance, and this is only weakly correlated with future performance. That’s one reason we keep harping on the expense ratio. This is one of the few things about a fund that is predictable, and high expenses will definitely make you poorer over time.

Sources: Morningstar; Yahoo Finance.