Every fund page in India leads with trailing returns: 1Y, 3Y, 5Y. They are the industry standard, they are what this site shows in its returns grid, and they share one structural weakness.
A trailing return is measured from exactly one start date to exactly one end date. Move either, and the number moves. Sometimes a lot.
The start-date problem
A “5-year return” quoted today runs from this date five years ago. That start point was one particular day in one particular market mood, and it was chosen by the calendar rather than by anything meaningful.
Consider what that does to a fund’s headline figure. A five-year window beginning just before a crash and ending after the recovery produces a flattering number. A window beginning at a peak produces a poor one. Same fund, same manager, same decisions — the difference is entirely which day the window happened to open.
This also creates a slow, invisible drift in fund rankings. As a bad quarter ages out of the trailing window, funds re-rank without anything happening in the present. Investors reading a “top 5Y returns” list in two consecutive quarters can see a genuinely different list for reasons that have nothing to do with the funds.
What rolling returns do instead
A rolling return computes the same window over and over, sliding the start date forward through the fund’s whole history: every possible 3-year period, not one of them.
You stop getting a number and start getting a distribution. That is the point, because the distribution answers the questions the single number cannot:
- How consistent is it? A fund whose 3-year rolling returns cluster tightly behaved similarly for most investors, whenever they happened to arrive.
- What was the worst case? The minimum of the rolling series is the experience of the unluckiest investor who held for that full period — a far more honest risk measure than a smooth CAGR.
- How often did it beat its benchmark or its peers? Not “did it beat them over the one window I am looking at”, but in what proportion of all windows.
A fund with a strong 5-year trailing number and weak rolling consistency is usually telling you that one exceptional stretch is carrying the record.
Reading them honestly
Rolling returns are better, not magic. Three cautions:
- Overlapping windows are not independent observations. Consecutive 3-year windows share most of their days, so the “hundreds of data points” a rolling chart appears to offer are far fewer in substance. Do not read statistical precision into them.
- They still only cover the history that exists. A fund launched in 2019 has no window containing 2008, however many windows you slice.
- The manager may have changed inside the window. A ten-year rolling record can span three managers and two mandate revisions. See when to sell a fund for what that does and does not mean.
What this site shows, and why
The returns grid uses trailing figures — absolute under a year, CAGR at a year and beyond — because that is what makes two funds directly comparable, and it is the convention every other source uses.
Rolling returns sit alongside them on each fund page precisely to answer the question the grid cannot: whether the trailing number is representative or a happy accident of today’s date. The windows and conventions used are written out on the methodology page.
The practical habit
When a fund’s trailing return looks outstanding, ask one question before anything else: is this the fund, or is this the window?
Check the rolling distribution. Check whether the same fund looked good a year ago and two years ago. And remember that whatever either number says, your own result depends on when you actually invested — which is XIRR’s job, and usually a different number again.
Key takeaway
A trailing return is one start date and one end date, and moving either moves the number — which is why fund rankings shuffle when nothing has happened. Rolling returns replace a single figure with a distribution, answering how consistent the fund was and what the unluckiest holder experienced. When a trailing number looks outstanding, ask whether it is the fund or the window.
Terms used here
More in Module 5 — Advanced metrics, taxation and wealth architecting
Mutual fund taxation decoded: short-term vs long-term capital gains
Equity, debt, hybrid and ELSS are taxed under different rules, and the rules changed twice in three years. What applies now, and to which of your units.
Decoding alpha and beta: manager skill versus market risk
Beta is how much market you took; alpha is what you got beyond it; R² says whether either number means anything. Read in order, they catch a closet indexer.
Absolute, CAGR, XIRR: which return are you looking at?
Three numbers that all answer “how did it do?” and disagree. Which one your statement shows, which one this site shows, and when each is the honest one.
Sharpe and Sortino: measuring risk-adjusted returns
Volatility, Sharpe, Sortino and maximum drawdown measure four different things, and only one of them predicts whether you will still be holding in year three.
Treynor and information ratio: advanced tools for comparing funds
One prices market risk, the other prices the decision to differ from the index. For choosing between active funds in a category, the second matters most.
Tracking error and standard deviation in passive funds
One measures how much a fund moves, the other how much it moves differently from its index — and neither is the number that actually reaches your returns.
What is portfolio turnover ratio? Decoding a fund’s trading activity
How much trading it took to produce the returns, the invisible costs that come with it, and why the number is a consistency check rather than a verdict.
Credit risk and yield-to-maturity in debt funds
A high YTM describes the risk taken, not the return you will earn. How to read it beside the rating profile, and what a credit event permanently does.
Factor investing and smart beta: beyond market-cap weighting
A disclosed, rules-based tilt at a fraction of active cost — and active risk by another name, with long underperformance that is the reason the premium exists.
Estate planning for mutual fund investors: transmission and legalities
Nomination, a will and joint holding — what each does, what transmission involves without them, and why inheriting does not reset the capital-gains clock.
The psychology of a market crash: behavioural finance that survives contact
Loss aversion, herding and action bias are not character flaws — they are default settings. The pre-commitments that work when in-the-moment judgement fails.
From the blog
Active vs passive investing: which wins in the long run?
02 Oct 2026
To 1 October 2026, most active large-cap funds beat Nifty 50 index funds over 5 years, but the lead shrank over 10. What the data shows, and what it doesn't.
5 behavioural biases that cost investors money
02 Oct 2026
Recency, loss aversion, herding, overconfidence and anchoring, each measured in Indian fund and Nifty data, with a practical fix for every one of them.
Mutual fund factsheets: what normal looks like in 2026
02 Oct 2026
A factsheet's numbers only mean something next to the fund's peers. Here is what each section shows, with the normal range for flexi-cap funds in 2026.
