When you check a mutual fund’s factsheet, you’ll usually find three-year, five-year or 10-year returns. These are point-to-point figures that show the growth of the fund between two fixed dates. But while easy to read, they can be misleading. A single start or end point can distort the picture depending on whether the market was booming or crashing during those dates.
Consider this: a fund that shows 20% annualized returns over three years might have benefited from a market low three years ago. Another that shows 5% might have started from a market peak. Does that make the first fund better? Not necessarily. It just had better timing, and that is exactly what you want to avoid when picking a fund based purely on past performance.
Rolling returns
Rolling returns measure performance over multiple overlapping periods, for example, three-year returns calculated daily over several years. Each calculation is “rolled” forward by one day, capturing every possible investment window within that timeframe. The result is a more comprehensive and fair assessment of how a fund performs across market cycles. In other words, this reflects a more consistent performance of the fund.
Take a ten-year period, for example, 2015 to 2025. If you look at the three-year rolling returns for each day over that period, you’ll see not just one number but a range of returns, perhaps from -4% to 50%. Now, when these multiple return observations are averaged out, you get a clearer picture of how the fund performed over this long period. You can also zoom in to see the best and worst of the fund’s performance, as well as how many times it delivered returns above the benchmark, thereby gauging its consistency.
The rolling returns can be further sliced and diced to see how many observations are above 12%, 15%, or 20%, to gauge how frequently the fund has delivered higher returns.
Take the example of the popular Parag Parikh Flexicap Fund. Over a 10-year period, the fund has delivered annualized returns of 18.8% as of 24 October 2025. However, if you examine the rolling return analysis, the average one-year return of the fund, calculated daily over the 10-year period, is slightly higher at 19.56%. Upon further examination of the data, across the ten-year period, 39% of the one-year return observations exceed 20%, indicating steady performance.
Real picture
Rolling returns help investors answer a more realistic question: what are the chances of earning decent returns, regardless of the starting point of the investment? It aims to mitigate the impact of timing. Point-to-point returns are limited as they only tell you how an hypothetical investor fared if he had entered at one point and exited at another point.
For fund analysis, rolling returns are particularly useful when comparing two schemes within the same category. A scheme that consistently beats its benchmark on a rolling basis is more dependable than one that outperforms occasionally. It filters out noise caused by market timing and focuses on performance across market phases—bull, bear, and sideways markets.
The longer the period over which returns are rolled, the more reliable the data becomes, as it reflects how the fund performs across an entire market cycle. You can even look at the frequency of negative observations to get a sense of how long you’d need to stay invested to avoid capital loss.
However, remember that any analysis based on past data cannot exactly predict future performance. It can only be used to make an assessment. You can get rolling returns data from sites such as primeinvestor.in, mfotline.co.in, as advisorkhoj.com.