Ask a friend why he bought his latest mutual fund and you will usually get one of three answers. It gave 40% last year. It has five stars on some app. Or the most honest one: someone in the office WhatsApp group said it’s good.
India’s mutual fund industry crossed ₹87 lakh crore in assets in August 2026, and a startling share of that money gets chosen exactly this way.
Now sit across the table from a trained financial planner choosing a fund for a client, and watch what happens. The last one-year return barely gets a glance. The star rating never comes up. Instead, out come questions that sound almost strange at first. How much does this fund overlap with what you already hold? What did it do in 2020’s fall, not just 2021’s rise? What is its Sharpe ratio against the category?
That gap between the two conversations is the subject of this piece. I want to walk you through five tools that CFP professionals and serious advisors use when they evaluate mutual funds, with enough worked arithmetic that you can actually try each one this week. If you are a student thinking about a career in personal finance, or an MFD or bank RM who wants to move beyond “sir, this fund is performing very well”, this is the level at which the profession actually operates.
None of this is secret knowledge. All of it is learnable. Let’s start with the check that surprises new investors the most.
Check 1: Portfolio overlap, or why owning five funds is not diversification
Here is a portfolio I see all the time. A large cap fund, a flexi cap fund, a “bluechip” fund from a second AMC, an index fund, and an ELSS for tax saving. Five funds, four AMCs. The owner is confident he is diversified.
Then you look inside.
SEBI’s scheme categorisation rules define the playing field very tightly. Large cap funds must put at least 80% of their equity in the top 100 companies by market capitalisation. Flexi cap and ELSS funds can go anywhere, but their managers know that straying too far from the big index names is career risk. So HDFC Bank, ICICI Bank, Reliance, Infosys and TCS appear at the top of almost every one of those five portfolios. Our confident investor does not own five funds. He owns the Nifty’s top ten stocks five times over, and pays five separate expense ratios for the privilege.
Planners put a number on this instead of guessing. The standard method: for every stock held by both funds, take the smaller of the two weights, then add those up. That sum is the overlap percentage.
A simplified example with just five common stocks:
| Stock | Weight in Fund A | Weight in Fund B | Smaller of the two |
|---|---|---|---|
| HDFC Bank | 9.5% | 8.0% | 8.0% |
| ICICI Bank | 8.0% | 9.0% | 8.0% |
| Reliance | 7.5% | 6.0% | 6.0% |
| Infosys | 6.0% | 7.0% | 6.0% |
| TCS | 4.0% | 5.0% | 4.0% |
| Overlap from these five alone | 32% |
Run this across the full portfolios of two large cap funds from different AMCs and overlaps of 50-65% are routine. The funds have different names, different fund managers, different marketing. They are substantially the same product.
What does a planner do with this number? There is no SEBI-mandated cutoff, so treat any threshold as judgment rather than law, but a common working rule is that two equity funds overlapping much beyond a third deserve a hard question: what is the second fund adding, other than a second expense ratio? Real diversification comes from combining funds that fish in different ponds: a large cap with a genuine mid cap or small cap fund (SEBI defines those ponds as the 101st-250th companies and 251st onwards respectively), or Indian equity with international equity, not from collecting lookalikes.
Try it yourself: pick any two equity funds you or your family hold, pull their latest portfolio disclosures, and compute the overlap for the top 15 holdings. It takes twenty minutes in a spreadsheet and it changes how you see “diversification” forever.
Check 2: Rolling returns, because your start date was luck
Every fund fact sheet shows point-to-point returns: 1-year, 3-year, 5-year, since inception. One start date, one end date, one number. The problem is that the start date is doing an enormous amount of hidden work.
Take a fund whose NAV fell hard in a crash and then recovered. Measure its 3-year return starting from the bottom of the crash and you get a spectacular number. Start six months earlier, at the pre-crash peak, and the same fund over almost the same period looks mediocre. Neither number is a lie. Both are useless in isolation, because no client of yours gets to choose the crash bottom as their entry date.
Rolling returns fix this by refusing to privilege any single start date. The method: compute the 3-year return starting from every single day (or month) in a long window, then look at the whole distribution of outcomes.
Suppose you compute all the 3-year rolling returns for two funds over the last ten years and summarise:
| Measure (3-year rolling, 10-year window) | Fund X | Fund Y |
|---|---|---|
| Average annualised return | 14.2% | 13.1% |
| Best 3-year stretch | 31% | 22% |
| Worst 3-year stretch | -4% | +3% |
| % of stretches beating 12% | 58% | 71% |
Fund X has the better average and the flashier best case. But a client who happened to invest at the wrong time in Fund X spent three full years going backwards. Fund Y never delivered a losing 3-year stretch and beat 12% more often. For a real household with real goals, a planner will very often prefer Y. Consistency compounds; brilliance that arrives only for lucky entry dates does not.
This is also why planners are unimpressed by “top performing fund of 2025” lists. A single point-to-point number tells you what happened between two arbitrary dates. A rolling-return distribution tells you what an investor’s actual experience was likely to be. Those are different questions, and only the second one matters when someone’s daughter’s education fees are riding on the answer.
Check 3: Standard deviation and the Sharpe ratio, which price the ride along with the destination
Two funds both returned 14% annualised over five years. Same destination. But one strolled there and the other took you on a roller coaster that made you want to redeem everything in every correction. Are they equally good? Obviously not, and risk statistics exist to say so with numbers instead of adjectives.
Standard deviation measures how widely a fund’s returns swing around their own average. If a fund’s annual return averages 14% with a standard deviation of 18 percentage points, then in a typical year you should be unsurprised by anything from roughly -4% to +32%, and roughly one year in twenty will land even outside that. On a ₹10 lakh investment, that is the difference between ending a normal-ish bad year at ₹9.6 lakh and a normal-ish good year at ₹13.2 lakh. When a planner says “this fund is volatile”, this is the number behind the sentence.
The Sharpe ratio then asks the sharper question: how much return did the fund generate per unit of that turbulence, over and above what you could have earned taking no equity risk at all?
Sharpe ratio = (Fund return − Risk-free return) ÷ Standard deviation
The risk-free rate is typically proxied by short-term government paper such as the 364-day treasury bill. Work one example. Fund P returns 14% with an SD of 18; Fund Q returns 12.5% with an SD of 10. Take the risk-free rate as 6.5%.
- Fund P: (14 − 6.5) ÷ 18 = 0.42
- Fund Q: (12.5 − 6.5) ÷ 10 = 0.60
Fund P wins every “top returns” screener. Fund Q is the better fund per unit of risk taken, and by a wide margin. For a retiree, or for any goal closer than five years away, a planner will take Q’s 0.60 over P’s 0.42 almost every time.
One refinement worth knowing: standard deviation punishes upside surprises and downside surprises equally, which is a little unfair. Nobody complains when their fund goes up too fast. The Sortino ratio repairs this by counting only downside deviation in the denominator. Same idea, more honest denominator. When you see a fund whose Sharpe looks ordinary but whose Sortino looks strong, its volatility has mostly been the pleasant kind.
Check 4: Capture ratios, or how a fund behaves when the market falls
One piece of arithmetic every advisor should be able to do in their head: a portfolio that falls 50% needs to rise 100% just to get back to zero. Losses are not symmetrical with gains. This is why experienced planners obsess over how a fund behaves in bad markets, not just good ones.
Capture ratios make this obsession measurable. The upside capture ratio asks: in the months the benchmark rose, what fraction of that rise did the fund deliver? The downside capture ratio asks: in the months the benchmark fell, what fraction of the fall did the fund suffer? Both are expressed as percentages of the benchmark’s move.
A fund with 95% upside capture and 78% downside capture is telling you something lovely: it gets nearly all of the market’s gains while sitting out a fifth of its pain. A fund with 110% up-capture and 115% down-capture is a leveraged mood swing; it looks brilliant in bull years and destroys client relationships in bear years.
Compounding favours the first kind. Take a two-year sequence where the market rises 30% then falls 25% (ending at 0.975 of where it started, near enough flat):
- Fund A (105 up / 105 down): rises 31.5%, then falls 26.25% → ₹100 becomes ₹96.97
- Fund B (90 up / 70 down): rises 27%, then falls 17.5% → ₹100 becomes ₹104.78
The “boring” fund that captured less of the rally finishes nearly 8% ahead, in a market that went nowhere. Stretch that arithmetic over a 20-year investing life with several bear markets in it, and you understand why the planner’s favourite funds are so often ones that never topped a single annual leaderboard.
Check 5: Alpha and beta, and the habit of asking “compared to what?”
The final habit is less a formula than a discipline: a trained planner never evaluates a return in a vacuum. 18% sounds wonderful until you learn the fund’s benchmark did 21% in the same period.
Beta measures how much a fund tends to move when its benchmark moves. A beta of 1.1 means the fund typically amplifies the index’s moves by about 10% in both directions; a beta of 0.85 means it dampens them. Beta tells you how much of the fund’s behaviour is simply the market, borrowed.
Alpha is what remains: the return the fund added (or subtracted) beyond what its beta-adjusted benchmark exposure would have produced on its own. Alpha is the only part of performance you can genuinely credit to the manager, and you should insist on seeing it measured against the Total Return Index (TRI), which includes dividends, because comparing against a price-only index flatters every fund by a couple of percentage points a year.
Two practical uses. First, alpha connects back to overlap: an “actively managed” fund whose portfolio overlaps 70% with its index, charging 1.8% while an index fund charges 0.2%, needs to generate real alpha on the remaining 30% just to justify its fee. Most don’t, most years. Spotting this pattern, called closet indexing, is bread-and-butter work for a planner. Second, beta lets you set expectations honestly: a client holding a 1.15-beta fund must be told, before the fall and not after, that a 20% market correction will likely feel like 23% to him.
The five checks side by side
| Check | The question it answers | Red flag it catches |
|---|---|---|
| Portfolio overlap | Are my funds actually different? | Five funds, one portfolio, five fees |
| Rolling returns | Was performance consistent or lucky timing? | Great since-inception number built on one lucky stretch |
| Std deviation & Sharpe/Sortino | Was the return worth the turbulence? | High returns earned with reckless volatility |
| Capture ratios | How does it behave when markets fall? | Bull-market hero, bear-market disaster |
| Alpha & beta (vs TRI) | Is the manager adding anything beyond the market? | Closet indexing at active-fund fees |
Notice what is missing from this table: last year’s return and star ratings. Both are results of a process rather than the process itself, which is why they never make a planner’s checklist.
Where do people actually learn this?
Everything in this article is learnable from public materials, and I would genuinely encourage you to compute one overlap and one Sharpe ratio by hand this week. Nothing builds conviction like doing the arithmetic yourself.
But there is a difference between knowing five tools and having a complete framework: knowing which tool answers which client question, how these metrics fit into asset allocation and goal planning, when a “worse” fund is the right recommendation, and how to explain all of it to a nervous client in plain words. That framework is what a structured professional education gives you. In the CFP certification pathway, the FPSB Investment Planning Specialist module treats everything above as early-chapter material and builds from there into portfolio construction, behavioural finance and goal-based advice; it is one of the specialist modules on the way to the full CFP credential awarded through FPSB. If reading this piece felt less like homework and more like finally seeing behind the curtain, that is usually a sign the profession would suit you. A programme like the one we teach at House of Financial Planners is the organised way in.
Either way, the next time someone in the office group says a fund “is good”, you now have five better questions to ask.
Frequently asked questions
What is a good portfolio overlap percentage between two mutual funds?
There is no official cutoff, but many practitioners get uncomfortable when two equity funds overlap much beyond about a third of their portfolios. At 50-60% overlap the second fund is adding fees, not diversification. Overlap between a fund and its own benchmark index is a separate check, used to detect closet indexing.
Are rolling returns better than CAGR?
They answer different questions. CAGR (point-to-point) tells you what happened between two specific dates; rolling returns show the full range of outcomes across every possible entry date, which is closer to what a real investor experiences. Planners use rolling returns to judge consistency and CAGR for simple communication.
What is a good Sharpe ratio for a mutual fund in India?
Sharpe ratios move with market conditions, so compare a fund’s Sharpe against its category peers over the same period rather than against an absolute number. Within a category, a meaningfully higher Sharpe means the fund earned its returns with less turbulence per unit.
Where do I find standard deviation, Sharpe ratio and capture ratios for Indian funds?
Fund factsheets publish standard deviation, Sharpe and beta monthly. Research portals and AMC sites publish capture ratios and rolling-return tools. The raw portfolio disclosures for overlap analysis are on each AMC’s website, updated monthly under SEBI’s disclosure norms.
Do I need the CFP certification to use these tools?
No. Everything here is public knowledge and free to practise. The certification matters when you want the complete framework these tools sit inside, the credential to advise clients professionally, and the training to connect fund analysis to real financial plans.
Sources
- SEBI, “Categorization and Rationalization of Mutual Fund Schemes”, circular SEBI/HO/IMD/DF3/CIR/P/2017/114, 6 October 2017 (large, mid and small cap definitions and the one-scheme-per-category rule): https://www.sebi.gov.in/legal/circulars/oct-2017/categorization-and-rationalization-of-mutual-fund-schemes_36199.html
- AMFI: Indian mutual fund industry AUM of ₹87.08 lakh crore as on 31 August 2026: https://www.amfiindia.com/articles/indian-mutual
- FPSB India: FPSB Investment Planning Specialist, the first specialist module on the CFP certification pathway: https://india.fpsb.org/students/fpsb-investment-planning-specialist/
All fund figures in the worked examples (Funds A/B, P/Q, X/Y) are illustrative, constructed to show the arithmetic; they are not real schemes. The 6.5% risk-free rate in the Sharpe example is an assumption chosen for round numbers; check the current 364-day T-bill yield when you compute your own.
