Short answer: A point-to-point return measures growth between one fixed start date and one fixed end date, so it describes only that particular pair of dates. Rolling returns repeat the same holding period across many overlapping start dates, producing a range of outcomes instead of a single figure.
Last reviewed: 17 November 2026
At a glance
| Term | What it means | Key limitation |
|---|---|---|
| Point-to-point return | Growth between one fixed start date and one fixed end date | Describes only that pair of dates |
| Rolling return | Return for a fixed holding period, calculated across many start dates | Needs a long history of data |
| Rolling return series | Choose a holding period, calculate, shift the start date, repeat | More demanding to calculate and understand |
| Minimum, maximum and average | The range of outcomes for a holding period | Depends on entry and exit dates |
| Annualised return | Compound annual growth rate (CAGR) for the chosen holding period | Expresses the result per year, not over the full span |
Why does a single start and end date mislead?
A point-to-point return is simple to compute. You take the value on one date, take the value on a later date, and measure the change. The result is accurate for those two dates.
The catch is that the answer depends heavily on which two dates you picked. If the start date moved a month earlier or later, the figure would usually change. A single number therefore tells you about one particular window, not about the holding period in general.
How is a rolling return calculated?
You begin by choosing a holding period, such as three years. You calculate the return for that period starting from one date. Then you shift the start date forward by a day, a week or a month, and calculate the return again for the same length of time.
Repeating this across many start dates produces a continuous series of returns. Each figure covers the same holding period, but the windows overlap, so neighbouring points in the series look very similar.
Why are rolling returns shown as annualised figures?
Rolling returns are typically expressed as annualised returns, such as a compound annual growth rate for the chosen holding period. Compound annual growth rate (CAGR) is the steady yearly rate that would take a starting value to its ending value over the period.
Here is a hypothetical example with invented numbers. Suppose a unit’s value rises from Rs 50 to Rs 60 over two years. The total gain is 20 per cent, but the annual rate is the square root of 1.2, minus 1, which is about 9.54 per cent a year. Annualised figures let you compare windows of different lengths on the same scale.
What do the minimum, maximum and average show?
For a rolling return series, the minimum, maximum and average describe the spread of outcomes for that holding period. They show the range of results an investor could likely have achieved, depending on when they entered and exited.
This is more informative than one figure, but it is still a description of past data. A high maximum and a low minimum tell you the outcomes varied widely across the start dates tested. They do not tell you which outcome any future period will produce.
Why do rolling returns reduce the effect of one date?
Because rolling returns span many start dates, they reduce the bias that comes from the particular start or end date you happened to choose. A single point-to-point figure has no such averaging.
Point-to-point or trailing returns also give limited insight into how consistently a fund performed across different market phases. A rolling return series shows whether results stayed within a similar range across many windows, which is a better guide to consistency over time.
What are the catches with rolling returns?
Rolling returns are more demanding to calculate and to understand. A reader needs to know the holding period, the annualisation and how to interpret a range rather than a single number. Spreadsheets or data services can do the arithmetic, but the reasoning still takes effort.
They also need a long history of data. A newer fund may not have enough history to produce a meaningful rolling series for a long holding period. In that case, a rolling return chart may simply not exist yet, and a point-to-point figure may be the only number available, with its limitations.
Frequently asked questions
Is a point-to-point return wrong?
No. It correctly measures growth between two specific dates. The issue is that it describes only that pair of dates, so it can look different if the dates change.
Why are rolling returns usually shown as annualised figures?
Rolling returns are typically expressed as annualised returns, such as a compound annual growth rate for the chosen holding period. This lets windows of the same length be compared on a common yearly scale.
Can rolling returns be calculated for a newer fund?
Not always. Rolling returns need a long history of data, so a newer fund may not have enough history to analyse.
Sources and last check
Facts in this article were checked against the sources below on 17 November 2026. Rules and figures can change; check the primary source before acting.
- zerodha.com (broker education)
- groww.in (broker education)
- kotakneo.com (broker education)
This article is general information, not investment advice. MyFinanceBandhu is not a SEBI-registered investment adviser.