Best Fund for Investment: The Framework We Use to Judge 8,000 Schemes

"Best fund for investment" is the most searched and least answerable phrase in Indian personal finance. Least answerable not because the data is missing - we recompute rankings across 8,000+ Direct-Growth schemes every morning - but because the question, as asked, is incomplete. A fund is not best in a vacuum. It is best for a horizon, a temperament and a job.

This is the pillar guide: the complete framework we use to judge every fund we track, the category-by-category evidence, and links to the deep dives for each specific job. By the end, "which fund is best?" should have become a question you can answer for yourself - which is worth more than any list, including ours. Descriptive research, not advice; all figures pre-tax from our live dataset.

Step 1: The horizon decides the asset class - before any fund name

The single most consequential number in investing is not a return. It is when you need the money back. Our rolling-window data - every possible entry date, not one lucky one - is blunt about what each horizon can bear:

Every mis-sold portfolio we see in the Portfolio Doctor breaks this table somewhere: three-year money in small caps, or thirty-year money in FDs. Get the row right and the remaining decisions are refinements. Get it wrong and no fund selection can save the outcome. (The perennial FD question has its own data treatment: mutual funds vs FD.)

Step 2: Inside the row, know what each category actually pays - and costs

The complete category evidence from the dataset, ten-year medians with the risk attached:

Three patterns worth internalising:

• Return and drawdown travel together - with one glaring exception. Mid caps pair the second-highest return with the best consistency in equity (92% of rolling years positive), which is why they anchor so many long-term structures.
• The conservative hybrid row is quietly remarkable: the median fund has never had a negative rolling year. That is the raw material for withdrawal plans - the entire SWP playbook is built on it.
• The spread within a category dwarfs the spread between adjacent categories. The gap between a top-quartile and bottom-quartile flexi cap fund is far wider than the gap between the flexi cap and large cap medians. Category picks the ceiling; fund selection decides how much of it you capture.

Step 3: Inside the category, judge funds on four numbers - not stars

The framework the terminal applies to every scheme, and the reason each number is there:

1. Consistency (rolling 1-year windows positive). The best single predictor of whether an investor actually stays invested. Above 85% is good; above 92% is elite.
2. Drawdown, priced in rupees. Percentages anesthetise. A -35% fall on the Rs 20 lakh you plan to accumulate is Rs 7 lakh on paper - decide now whether you can watch that. The fund you can hold through its worst month beats the fund with the prettier CAGR you sell at the bottom.
3. Quality rank within category (King Score, 0-100). Blends 3-year return percentile, consistency, drawdown and Sharpe - each ranked against true peers only, so a sector fund cannot fake greatness during its theme's rally. How it works.
4. Momentum rank within category (RS Rating, 1-99). Describes who is leading right now - useful for entries and for noticing decay early. High RS is a description of the recent past, never a forecast. The full explainer.

And one number to minimise rather than maximise: cost. The Direct-plan versions of the same funds run roughly 0.5-1% a year cheaper than Regular; index funds go as low as 0.06%. Over 20 years the fee gap alone compounds into lakhs - the arithmetic here.

Step 4: Match the fund to the job - the six jobs, with their deep dives

Most "best fund" searches are one of six jobs in disguise. Each has a dedicated data guide:

If the money has no job yet - "just make it grow" - the default evidence-backed structure is unglamorous: a core equity fund (flexi, large & mid, or index) taking most of the money, one mid or small cap satellite, a liquid fund for the emergency buffer. Two to four funds total. Our overlap data is unambiguous about what a fifth and sixth fund adds: mostly the same stocks, again.

Step 5: The mistakes that undo everything above

Ten years of fund data, and the same handful of errors explain most destroyed value:

1. Buying last year's #1. The top of the 1-year leaderboard is systematically where next year's mean reversion lives. If momentum is wanted, use a measured rank (RS) with an exit rule - not a brochure.
2. Owning ten funds. Diversification on paper, index-hugging in fact - at active-fund prices. Beyond ~8 funds the overlap is structural; our Portfolio Doctor's look-through makes it visible in seconds.
3. Confusing volatility with risk. For 15-year money, the -33% mid cap fall is an inconvenience the SIP exploits. For 3-year money it is a catastrophe. Same number, opposite meanings - the horizon table is the decoder.
4. Paying Regular-plan fees for Direct-plan funds. The most fixable leak in Indian portfolios.
5. The behaviour gap. Across funds we track, investors' money-weighted returns trail the funds' own returns by 2-4 points a year - the cumulative cost of buying high, stopping SIPs in falls, and switching on emotion. Every framework above is ultimately armour against this one enemy.

Five ready structures, by life situation

The framework, pre-assembled. Illustrations built from the category data above - starting points for thinking, not prescriptions:

The starter (first salary, Rs 5-15k monthly). One fund: a flexi cap or Nifty 50 index fund, SIP with 10% annual step-up, plus one month of expenses in a liquid fund growing toward six. Nothing else until the emergency buffer is full. The temptation to own five funds at Rs 2,000 each is the starter's classic error - one fund, learned deeply, beats five funds owned shallowly.

The compounder (30s, surplus Rs 25-60k monthly). The two-SIP core-satellite: 60-70% into large & mid cap or flexi cap, 30-40% into mid cap (the 17.3% median / 92% consistency standout), both stepped up annually. ELSS instalments if the 80C bucket has room. Review once a year against the four numbers.

The family stage (40s, multiple goals). Same engine, plus explicit goal separation: the child's education corpus (10+ years out) runs its own equity SIP; the 3-5 year goals (car, home upgrade) live in balanced advantage or equity savings funds - never in the small cap sleeve, however tempting its median. Insurance fully separated from investment.

The pre-retiree (5-8 years out). New instalments shift toward hybrids; the accumulated corpus begins its tranche-by-tranche glide. The full mechanics - including why the glide starts five years out and moves annually - are in the retirement guide.

The income stage (post-60 or financially independent). The two-bucket SWP: a 24-month liquid runway plus a conservative-hybrid engine, withdrawing near 6%, reviewed annually. Complete walkthrough with named funds: best funds for SWP.

Notice what all five share: fewer than five funds, an automatic instruction doing the discipline, and one scheduled review a year. Complexity is not sophistication - in our Portfolio Doctor data, the correlation between fund count and outcome quality is negative past eight holdings.

How to read a fund page like an analyst

The framework's four numbers live on every fund page in the terminal. The sixty-second reading order:

1. The candle chart first - the full NAV history. You are looking for the personality: how deep were the falls, how long were the flat years, how steady is the staircase? A decade of price action tells you more about the holding experience than any single statistic.
2. Drawdown panel, in rupees. Find the worst fall and convert it mentally to your intended corpus. This is the pre-commitment test - the page shows the fall priced in money precisely because percentages fail to frighten properly.
3. Rolling returns, median and worst. The honest answer to "what does this fund pay?" is its median rolling window, not its trailing return - and the worst window is the price of admission.
4. Consistency percentage. Above 85%: patience has usually been rewarded. Below 80%: expect year-long deserts.
5. King Score and RS together. Quality (slow, structural) and momentum (fast, current) - a high-King, decaying-RS fund is aging; a low-King, spiking-RS fund is having a moment. The combination reads better than either alone.
6. Expense, last. Between two funds surviving the first five checks, the cheaper one wins by default.

Sixty seconds per fund, six checks, and the listicle dependency is broken permanently.

The vocabulary that unlocks everything else

Seven terms, defined the way the data uses them:

• CAGR - the constant annual rate connecting start to end value. Smooths a violent journey into one polite number; always ask for the drawdown beside it.
• XIRR - your personal return, weighting every deposit and withdrawal by its date. The only honest measure of your outcome, and the number our Portfolio Doctor computes from real transactions.
• Drawdown - peak-to-bottom fall. The maximum historical regret of buying at the worst moment. Full explainer.
• Rolling return - the return of every possible window of a given length, not one lucky window. The anti-cherry-picking statistic.
• Consistency - share of rolling 1-year windows that ended positive. The best predictor of whether investors actually stay.
• King Score - our 0-100 within-category quality blend (returns, consistency, drawdown, Sharpe). Slow-moving by design.
• RS Rating - our 1-99 within-category momentum rank, recomputed daily. Fast-moving by design; descriptive, never predictive.

FAQ

So which fund should I actually buy?
Run the steps: horizon → category → four numbers → job. On this site the work is pre-done daily: the rankings hub lists every category's current field, ranked by the quality framework above, refreshed each morning. A list without the framework is a horoscope; the framework without a list is homework - use both.

Is there one fund that is best for everything?
No, and distrust anything claiming otherwise. The closest single-fund compromises are a flexi cap fund (whole-market mandate) for growth money or a balanced advantage fund for calmer money - each giving up something the specialised structure captures.

Active funds or index funds?
The dataset holds both without ideology. Nifty 50 index funds delivered ~12% over the decade at 0.06-0.26% cost with zero selection risk; the top quartile of active funds beat that meaningfully, the bottom quartile trailed it while charging 10x the fee. Index-as-core, selective-active-as-satellite is the structure the evidence keeps ratifying. Current index field.

How often should I review?
Annually, twenty minutes, against the same four numbers - consistency trend, King Score, mandate drift, your own horizon. More often is not diligence; it is the behaviour gap warming up.

Where does gold or silver fit?
As a 5-15% diversifier, sized like the volatility it carries - the complete silver data and gold rankings cover that world.

Everything in this guide - every median, ranking and drawdown - regenerates daily in the terminal from AMFI data, free. The framework is the product; the tables are just today's snapshot of it. Tomorrow's snapshot will differ. The framework will not.

Are New Fund Offers (NFOs) worth buying?
Almost never on the evidence: an NFO has no track record to judge by the four numbers, launches when its theme is marketable (usually late in the theme's cycle), and starts at Rs 10 for marketing reasons that have zero bearing on value. The framework's answer is structural - a fund becomes evaluable at 3 years old and comparable at 5. Until then it is a story, and stories are what the framework exists to replace.

Should market conditions change which fund I pick?
The category mix, mildly; the framework, never. When our regime data shows a bear phase, new lumpsums lean toward the hybrid cushion and the safest-funds list; SIPs simply continue. What should never respond to conditions is quality standards - a bear market makes cheap funds cheaper, not bad funds better.

How is all of this taxed?
By fund type and holding period, under rules that have changed repeatedly - equity and non-equity funds carry different rates and thresholds, every SIP instalment has its own clock, and each switch between funds is a taxable redemption. Every figure on this site is pre-tax by policy; the framework survives any tax regime, but the execution details deserve a current-year check before large moves.

Where do international funds fit?
As a satellite for genuine diversification - different markets, different currency - sized like a satellite (10-20% at most). Their records in our dataset span from excellent to dismal, and several have faced regulatory subscription limits. Apply the same four numbers, plus one more question: does the portfolio actually need exposure the Indian market cannot give it?

What is the minimum amount to start investing in mutual funds?
Most Direct plans accept Rs 500 SIPs and Rs 1,000-5,000 lumpsums - the entry barrier is administrative, not financial. The framework does not change with the amount: a Rs 500 instalment deserves the same four-number scrutiny as a Rs 5 lakh one, because the habit being built is the actual asset. What small amounts buy that large ones cannot is cheap experience: a full market fall lived through with Rs 20,000 at stake is the training that makes Rs 20 lakh survivable later.

Should I invest through a distributor, a platform, or directly with the AMC?
The plan type matters far more than the pipe: Direct plans through any route (AMC site, registrar apps, direct-plan platforms) carry no embedded commission; Regular plans through any route carry roughly 0.5-1% a year of it, forever. Choose whatever interface you will actually use consistently - then verify the word "Direct" appears in every scheme name it sells you.

How do I know if my existing portfolio follows this framework?
Upload it to the Portfolio Doctor - a CAS PDF, an Excel, or even a screenshot of your holdings screen. It scores the whole book on exactly these numbers: weighted quality, momentum, cost, concentration, overlap between funds, and your actual money-weighted XIRR against what the funds themselves delivered. The gap between those last two numbers is your personal behaviour gap, measured - and closing it is worth more than any new fund you could add.