The late Benjamin Graham, globally renowned value investor and author, was of the opinion that past performance is useful in calculating the value of a mutual fund only so far as it is indicative of what is to come in the future. The key word being ‘indicative’.Think about it. A student who receives straight A’s throughout his entire academic career isn’t guaranteed to repeat this performance going forward. Having admitted that, it does give us some clue as to what one might expect in the future. In the same way, past performance plays a very important role in providing useful information in identifying future winners. However, it is absolutely no guarantee for future success. Performance data is remarkably informative but its predictive power is limited.

When looking at past performance, pay attention to consistency. Don’t get swayed by a sporadic burst in numbers. Go for funds that put up a good performance on a consistent basis, rather than those that have a great run that immediately followed a dismal one.When looking at returns, check to see how much it diverges from its peers and from its benchmark. All in all, this will give you a very realistic view of your fund’s ability to maintain a good performance.

To get an accurate picture of a fund’s consistency and its ability to perform under different market conditions, don’t stick to the trailing returns. Trailing returns can obscure a lot of action under the surface. They could get skewed with a good one-year number. Back up your analysis with annual returns.For instance, a fund may be topping the charts with a great 10-year return that might cloak some dismal years along the way. Would you want to live with a fund that can plummet during a year and then climb the charts furiously later? Look at how the fund did in 2007 and 2009 to see how it performed during market rallies and in periods such as 2008 to figure how it got battered during a bear phase. If the fund holds up well on the downside, it could hint at conservatism. On the other hand, it could be a perennial bull run winner that collapses during downturns.

Past returns are undoubtedly useful in assessing the effectiveness of a fund manager or strategy, but they must be put in context.What do the pattern of returns say about the fund manager’s investment style? Can the numbers be reconciled with the fund’s stated investment approach? How has the fund performed during sharp rallies and market downdrafts? How much has been the divergence of returns when compared with the benchmark and its peers?Past performance has turned out to be the most common selection criteria and does play a confirmatory role in the fund selection process but is not a divining rod. Using it in isolation can be harmful. A good fund may experience years of bad returns during to its strategy. A value fund may be languishing when growth stocks are ruling. Conversely, a non-so-impressive fund may have a good run simply because the market is favouring the stocks it owns.Just as very strong recent performance often doesn't foretell strong returns ahead, the opposite is also true: The fund with very weak recent returns can end up on top.

What risk did the fund manager take to achieve that return? Did he just get lucky with his top holdings? One of the easiest ways for a fund to climb to the top of the return charts is to make a sizable bet on one or two sectors—or even a handful of individual stocks—that subsequently take off.

But risk-taking can cut both ways, and boldly positioned funds frequently experience long spells of poor performance after their hot streaks. If a fund is at the top of the charts during several trailing time periods, it's more important than ever to ask questions about what kind of risks it might be taking rather than assuming the manager can keep up the winning streak.