The short version: Returns: CAGR and XIRR and Alpha, Beta and Sharpe
Most people judge funds by one number: last year's return. That is the worst way. Here is a better way.
Types of returns
- Absolute return: total gain over a period. Good for under a year.
- CAGR: yearly growth rate for a lumpsum held over more than a year.
- XIRR: yearly return for SIPs and irregular cash flows.
- Trailing returns: returns up to today over 1, 3, 5 or 10 years. These change every day.
- Calendar-year returns: returns for each year, such as 2022 or 2023. They show how the fund behaved in different market phases.
- Rolling returns: returns over every possible window of a fixed length, such as every 3-year period in the last 10 years.
Why rolling returns are better
Trailing returns depend on today's date. A fund that had a great last six months can show an excellent 1-year figure that has nothing to do with long-term skill.
Rolling returns ask: "If I had invested on any day, what would I have got after 3 years?" and repeat that for every day. You then see:
- The average return.
- The worst return.
- How often the fund beat its benchmark.
Example. Fund A beat its benchmark in 75 out of 100 rolling 3-year periods and its worst 3-year return was 4% a year. Fund B beat its benchmark in 45 of 100 periods and its worst was −3% a year. Fund A looks more consistent.
The right benchmark
A fund should be compared to an index that matches its style.
- A large cap fund against a large cap index.
- A mid cap fund against a mid cap index.
- A debt fund against a matching debt index.
SEBI requires funds to use the Total Return Index (TRI), which includes dividends. Comparing against a price-only index makes the fund look better than it is.
Compare with peers
Also compare against funds in the same category. A mid cap fund that returns 15% in a year when the average mid cap fund returned 22% is lagging, even if 15% sounds good.
Judging risk as well as return
| Measure | What it tells you |
|---|---|
| Standard deviation | Size of swings |
| Maximum drawdown | Worst fall from a peak |
| Recovery time | How long it took to get back to the old peak |
| Sharpe ratio | Return per unit of total risk |
| Sortino ratio | Return per unit of downside risk |
| Beta | Sensitivity to the market |
| Up capture / down capture | Share of market rises and falls that the fund took |
A good active fund often has down capture below 100% (falls less than the market) with up capture close to or above 100%.
Worked example
Two mid cap funds.
| 1-year | 3-year (yearly) | 5-year (yearly) | Worst fall | |
|---|---|---|---|---|
| Fund A | 42% | 15% | 13% | −38% |
| Fund B | 20% | 16% | 15% | −29% |
Fund A shines on the 1-year number, but over 3 and 5 years Fund B is ahead and fell less in the worst phase. Fund B looks stronger.
Traps to avoid
- Recency trap. Choosing a fund because of the last 1 year.
- Survivorship bias. Funds that did badly are often merged or closed, so the survivors look better.
- Star manager trap. The manager who delivered the returns may have left.
- Benchmark games. A fund compared to an easy benchmark.
- Start-date games. A chart that starts at a market low.
- Mixing lumpsum and SIP returns. CAGR and XIRR are not comparable.
Understanding where returns came from
Performance attribution splits a fund's return into:
- Stock selection: picking winners.
- Sector allocation: over or under-weighting sectors.
- Cash and timing.
If a fund's outperformance came from one big bet that worked once, it may not repeat.
A simple scorecard
For each shortlisted fund, note:
- Rolling 3-year and 5-year return vs benchmark and peers
- Worst fall and recovery time
- Down capture
- Manager tenure and process
- Expense ratio
- Fit with your portfolio
The best fund is the one that scores well on most, not the one that tops one chart.
Common mistakes
- Reviewing every month and switching on short-term results.
- Comparing a debt fund's 1-year return with an equity fund's.
- Ignoring costs when comparing returns.
Questions people ask
How long should I wait before judging a fund? At least 3 years, ideally through a full market cycle.
Is a fund that beats the index every year better? Rarely does any fund do that. Look at consistency over rolling periods.
Takeaways
- Rolling returns and risk measures beat a single return number.
- Use the right benchmark and compare with peers.
- Look for steady results, not a lucky year.
Next deep dive: Mutual Fund Tax in Depth
Mutual fund investments are subject to market risks. Read all scheme-related documents carefully. This page is for education and is not investment advice.