Most fund advertisements and comparison tables show returns over fixed periods ending today: one year, three years, five years. These are point-to-point returns, and they can change dramatically depending on where the start and end dates happen to fall.
Rolling returns look at every possible period of the same length over a longer history, giving a much fuller picture of how consistent a fund has been. This article explains the difference, works through an illustration, and shows how to use rolling returns when comparing funds.
The problem with point-to-point returns
A point-to-point return compares the value on one date with the value on another. The trailing returns shown on factsheets are point-to-point returns ending on the latest date.
The trouble is endpoint sensitivity. If the start date falls in a market crash, the return looks spectacular, because the fund is being measured from a low point. If the start date falls at a market peak, the same fund looks poor.
Illustration: a fund's five-year return measured from just after a sharp fall might show 18 percent a year. Measured from six months earlier, before the fall, the same fund might show 10 percent. Nothing about the fund changed; only the start date.
The same problem affects comparisons. A fund that happens to look best on today's trailing table may simply have had a favourable start date.
What rolling returns are
Rolling-period returns calculate the return over a fixed length, such as three or five years, starting on every day, week or month across a long history, and then summarise all those results.
For example, three-year rolling figures over a fifteen-year history would include the three years starting January 2011, the three years starting February 2011, and so on, up to the most recent completed three-year period. That gives well over a hundred overlapping periods instead of one.
The results are usually summarised as:
- Average or median rolling return.
- Minimum and maximum: the worst and best periods.
- Percentage of periods above certain thresholds, such as above 0 percent, above 8 percent or above 12 percent.
- Percentage of periods in which the fund beat its benchmark.
A worked illustration
Illustrative figures for two equity funds, using five-year rolling figures over a long history.
| Fund X | Fund Y | |
|---|---|---|
| Trailing 5-year return today | 17% | 13% |
| Median 5-year rolling return | 11% | 12% |
| Worst 5-year period | −1% | 5% |
| Best 5-year period | 22% | 18% |
| Periods above 10% | 55% | 70% |
| Periods beating benchmark | 50% | 75% |
On the trailing table, Fund X looks clearly better. These overlapping-period figures tell a different story: Fund Y has had a higher median, a far better worst case, and has beaten its benchmark much more consistently. Fund X's recent figure reflects a favourable period; its history shows wider swings in outcomes.
Neither fund is necessarily the right choice, but rolling-period returns change the question from "who is top today?" to "who has been dependable?"
Reading rolling returns sensibly
Match the period to your horizon. If you invest for five years or more, five-year rolling-period returns are more relevant than one-year ones, which mostly reflect market noise.
Look at the worst case. The minimum rolling return shows the worst outcome an investor in that fund would have experienced over that holding period. Ask whether you could have lived with it.
Compare with the benchmark and peers over the same dates, not just in isolation.
Enough history. Rolling figures are only informative with a long record. A fund with five years of history has just one five-year period.
Manager and strategy changes. A long rolling-return history may cover several managers or a change in mandate; the past may describe a different fund.
Overlap. Rolling periods overlap heavily, so they are not independent observations; a single great year affects many of them.
SIP rolling returns and calendar-year returns
For SIP investors. Lump-sum rolling figures assume all money went in on one date. Someone investing monthly experiences something different, because each instalment has its own start date. Some research tools calculate rolling SIP returns: for every possible start month, they simulate a monthly SIP over the chosen period and compute its XIRR. The resulting range is usually narrower than for lump sums, because regular investing spreads entry points across good and bad markets.
Calendar-year returns are another simple check. Listing a fund's return for each calendar year alongside its benchmark shows how it behaved in strong years, weak years and crashes. A fund that beat its benchmark in most years, including the bad ones, has a more convincing record than one whose long-term figure rests on a single exceptional year.
None of these measures predicts the future. Together, they show how a fund has behaved across many different conditions, which is the best evidence history can offer.
Calculating rolling returns yourself
With a spreadsheet and the fund's historical NAVs, available from fund house websites and AMFI:
Step 1. List NAVs by date in two columns.
Step 2. For each date that has a NAV five years later, calculate the annualised return: (later NAV ÷ earlier NAV) raised to the power of 1/5, minus 1.
Step 3. Summarise the column: median, minimum, maximum, and the share of results above your chosen thresholds.
Step 4. Repeat for the benchmark index over the same dates, and count how often the fund beat it.
Many fund research websites also offer rolling-return charts, sometimes for SIPs as well as lump sums. Check the period and frequency they use.
Point-to-point returns are not useless; they are simply one sample. These overlapping-period figures show the whole distribution, and for long-term investors the distribution is what matters.
An account for direct mutual funds
Compare funds carefully, then invest in direct plans through SIPs or lump sums. Free to open
Figures are illustrations, not data for any real fund. Past returns do not indicate future returns. This is educational material, not advice, and no fund is recommended. We run no tips group and no signal service, we manage nobody's money, and no return is promised.
Disclosure: the account-opening link on this page is under Atul Shrivastava's Zerodha Authorised Person registration (NSE AP Reg: AP2516003481; Zerodha Broking Ltd. SEBI Reg: INZ000031633) and earns a revenue share. TheFinBaba is not a SEBI-registered Investment Adviser — this content is educational, not investment advice.
Frequently Asked Questions
What are rolling returns?
Returns calculated over a fixed length, such as five years, starting on every day, week or month across a long history, then summarised by median, worst, best and consistency.
What is a point-to-point return?
The return between two specific dates, such as the trailing returns ending today shown on factsheets. It depends heavily on the chosen dates.
Why are rolling returns better for comparing funds?
They show consistency across many periods rather than one, reducing the effect of a lucky or unlucky start date.
Which rolling period should I look at?
One that matches your investment horizon, typically three or five years or longer for equity funds.
How can I calculate rolling returns?
Using historical NAVs in a spreadsheet, calculate the annualised return for each start date with a matching end date, then summarise the results.
Related Reading
- Sharpe, beta and other fund ratios
- Annualising returns on your own investments
- How entry timing affects outcomes
- Checking what your funds hold
- Why new funds lack a track record
- Inflation's bite on fund returns
Disclaimer: TheFinBaba provides educational content only - this is not investment advice. Trading involves risk of loss.