Rolling Returns in Mutual Funds: A Fairer Way to Judge Consistency

Open any fund factsheet and the first thing you see is a neat row of numbers: one-year, three-year, five-year returns. They look precise, and they are. They’re also a single snapshot taken on one particular day. Change that day by a few weeks and the same fund can look like a star or a dud. The rolling returns mutual fund investors look at to get past that problem are simply the same return, calculated again and again over a moving window, so you see the whole spread of outcomes instead of one lucky (or unlucky) frame.

This guide walks through what rolling returns are, why point-to-point numbers can mislead, how to calculate a rolling series yourself from NAV data, and how to read the result. There’s a full worked example with two hypothetical funds, and the numbers in it are made up so the arithmetic is easy to follow.

Rolling returns, mutual fund style: what they actually measure

A normal return figure has one start date and one end date. If a factsheet says a fund returned 12% a year over three years, that’s the annualised return (CAGR) from one specific day three years ago to the factsheet date. That’s called a point-to-point or trailing return.

A rolling return keeps the holding period fixed, say three years, and slides the start date forward at a regular interval: daily, weekly, monthly or yearly. Each step gives you another three-year CAGR. Do this across a fund’s history and you end up with dozens or hundreds of overlapping three-year results. Tata Mutual Fund’s explainer describes it exactly this way, as an annualised return calculated repeatedly by shifting the starting date, which is why it’s also called rolling CAGR.

Investopedia puts the purpose plainly: rolling returns show how often, and by how much, an investment had stronger and weaker stretches, without being skewed by whatever happened in the most recent month or quarter.

So the output isn’t one number. It’s a distribution. The best three-year window, the worst one, the middle, and how many windows cleared a level you care about.

Why a point-to-point return can fool you

The trouble with a trailing return is that it depends entirely on the two dates. A fund that had a bad week right at the start of the window gets flattered; one that had a great week right before the start looks worse than it really was.

ICICI Direct’s iLearn chapter on mutual fund return calculations has a tidy illustration. Two funds both start at a NAV of ₹100 and both show a one-year trailing return of 12% on one date. Move the measurement just one day later and one fund’s one-year return drops to 7.84% while the other’s jumps to 15.31%. Nothing about the funds’ quality changed overnight. Only the start and end points moved.

Investopedia makes a similar point over a longer horizon. A fund can show a 9% annualised return over ten years while swinging widely inside that decade, up 35% in one year and down 17% in another, in its illustration. The 9% average is true. It just hides the ride.

A trailing return answers the question “what if I’d bought on exactly that day and sold on exactly this one?” Very few investors actually did that.

That’s the core case for rolling returns. Real investors come in on different days, through lumpsums at random moments and SIPs spread over months. A measure that averages over many entry points is closer to what a typical investor in the fund actually experienced.

How to calculate rolling returns, step by step

You don’t need special software. A spreadsheet and the fund’s historical NAVs (AMCs publish them on their websites) are enough. The process, as both the Tata MF and ICICI Direct guides lay it out:

1. Pick the rolling period. This should match your own horizon. Three, five or ten years are common choices for equity funds.
2. Pick the interval. How far you slide the window each time. ICICI Direct’s rule of thumb: daily intervals for three-month rolling returns, weekly data for a one-year period, and monthly data for periods longer than three years.
3. Calculate the first CAGR. CAGR = (ending NAV ÷ starting NAV)^1/years − 1.
4. Slide and repeat. Move the start date forward one interval, keep the period length unchanged, and calculate again. Keep going until the window’s end date reaches the latest NAV.
5. Summarise the series. Look at the minimum, maximum, median and average, and count how often the return fell below a level that would have hurt you.

One detail on the summary. Some guides, including ICICI Direct’s, collapse the series into a single average figure. That’s useful, but we’d argue the spread matters more than the average. Two funds can have the same average rolling return and wildly different worst cases.

Worked example: two hypothetical funds, ten years of NAVs

Here are year-end NAVs for two made-up equity funds from 2015 to 2025. Both start at ₹100. These are illustrative numbers, not any real scheme.

The point-to-point view

Over the full ten years, Fund A went from ₹100 to ₹228, a CAGR of about 8.59%. Fund B went from ₹100 to ₹236, a CAGR of about 8.97%. On the factsheet, B wins. Not by much, but it wins.

Now watch how sensitive Fund B’s numbers are to dates. Its five-year return from end-2015 to end-2020 (₹100 to ₹135) works out to about 6.19% a year. Shift the window by one year, end-2016 to end-2021 (₹96 to ₹190), and the five-year CAGR becomes about 14.63%. Same fund, same five-year length, and the answer more than doubles because one window happens to start at a low and end at a high.

The rolling view

Now calculate a three-year CAGR for every possible window, sliding the start forward one year at a time. That gives eight windows for each fund.

Summarise each column and the picture changes completely:

• Fund A: worst 7.42%, best 9.49%, median about 8.47%, average about 8.50%. Every single three-year window landed between 7% and 10%.
• Fund B: worst 2.60%, best 19.26%, median about 9.46%, average about 10.01%. Four of the eight windows beat 10%, but two came in below 6%.

So which is better? That depends on what you need. Fund B’s average three-year outcome was higher. But an investor who happened to enter at end-2017 and needed the money three years later earned about 2.6% a year. Same fund, same holding period, very different experience. Fund A never delivered anything spectacular, and never delivered anything awful either.

The ten-year factsheet number said B was slightly better. The rolling series said B was a coin toss with a fatter upside, and A was close to a sure thing.

Neither reading is wrong. They answer different questions. The rolling series is the one that tells you what range of experiences you’re signing up for.

Reading a rolling return series without fooling yourself

A few habits help when you look at rolling data on a fund research site or build it yourself.

Match the window to your real horizon

If you’re saving for something four years out, a one-year rolling series tells you about noise you don’t care about, and a ten-year series covers a period you won’t be invested for. Pick the window closest to how long the money will actually stay in.

Look at the worst case first

Average returns are what fund marketing leads with. The minimum rolling return is the number that tells you whether you could have lived with the fund in its worst stretch. If the worst three-year window for an equity fund was negative, ask yourself honestly whether you’d have stayed put.

Compare against the benchmark, window by window

A rolling series means more when you run the same windows for the fund’s benchmark index. If a fund beat its benchmark in most rolling windows, that’s stronger evidence of skill than beating it on one trailing date. Pair this with a fund’s up market capture ratio to see how that outperformance splits between rising and falling markets.

Watch the length of history

Ten years of data gives you only eight three-year windows at yearly steps, and they overlap heavily. Monthly steps give you many more windows, but they’re still drawn from the same underlying decade. A fund with five years of history simply can’t tell you much about five-year rolling returns.

Remember it’s still the past

Rolling returns are a better summary of history, not a forecast. A sector fund that had a fantastic run shows strong rolling numbers right up until the cycle turns. Treat a narrow range as evidence of consistency so far, nothing more.

Rolling returns and SIP investors

If you invest through a SIP, you’re buying units every month, so your own experience is already a blend of many entry points. ICICI Direct’s guide makes the point that SIP investors in particular should focus on rolling returns, because what they’re really counting on is consistent performance rather than one good stretch.

ICICI Direct also offers a simple test. If a fund’s rolling return and its trailing return aren’t far apart, its performance has been fairly consistent. A big gap suggests the trailing figure owes a lot to the dates chosen.

Where rolling returns fit with other fund metrics

Rolling returns tell you about the range of outcomes. They don’t tell you directly how deep the falls were inside each window, or how much risk the manager took to get there. For a fuller picture, read them alongside:

• Drawdown measures, such as the Calmar ratio, which compares returns against the fund’s worst peak-to-trough fall.
• Volatility and risk-adjusted ratios, which show how bumpy the ride was in between the start and end of each window.
• Costs and portfolio changes, because a fund’s past rolling record was earned under a particular manager, mandate and expense ratio.

And whichever metric you lean on, the discipline is the same one we apply to trading: decide in advance how much downside you can sit through. Our guide to risk management in trading covers that thinking in detail.

A quick checklist before you pick a fund on returns

• Ignore the single trailing number as a verdict. Treat it as one data point.
• Pull rolling returns for a window that matches your horizon.
• Note the minimum, the median and the share of windows below the return you need.
• Run the same windows for the benchmark and a couple of peers.
• Check how many years of history sit behind the series.
• Decide whether you could have held through the worst window, not just the average one.

Bottom line

Point-to-point returns are easy to read and easy to game by timing, even unintentionally. Rolling returns take more work, but they show what a fund delivered to investors who came in at many different moments, which is much closer to real life. In our two-fund example, the factsheet winner turned out to be the fund with the wider, scarier range. That’s the kind of thing a single number will never tell you.

Disclaimer: This article is for education only and is not investment advice. All NAVs, returns and funds in the worked example are hypothetical and do not represent any real scheme. Mutual fund investments are subject to market risks; past performance, including rolling returns, does not guarantee future results. Read all scheme-related documents carefully and consult a SEBI-registered adviser before making investment decisions.