The problem with absolute thresholds
Imagine two funds. Fund A is a Small Cap fund that returned 22% CAGR over 3 years with a Sharpe of 0.85 and a max drawdown of −38%. Fund B is a Large Cap fund that returned 14% CAGR with a Sharpe of 1.10 and a max drawdown of −18%.
An absolute scorecard — one that awards points based on fixed thresholds like “Sharpe above 1.0 = good” or “drawdown below 25% = safe” — would score Fund B significantly higher than Fund A. But that comparison is almost entirely meaningless. Large Cap funds operate in a different risk-return regime than Small Cap funds.
❌ Absolute scorecard — what goes wrong
- Fund A (Small Cap): 22% CAGR, Sharpe 0.85, Drawdown −38% → Score: 52 / D
- Fund B (Large Cap): 14% CAGR, Sharpe 1.10, Drawdown −18% → Score: 81 / A
Conclusion: “Fund B is the better fund.” But Fund A beat its category median by 6.8 percentage points. Fund B lagged its category median by 1.2 percentage points. The absolute score rewarded the wrong fund.
✓ Peer-relative score — what actually works
- Fund A (Small Cap): ranks 82nd percentile in its category → Score: 82 / A
- Fund B (Large Cap): ranks 44th percentile in its category → Score: 44 / C
Conclusion: “Fund A is doing well in its category; Fund B is below-average in its.” You now know which fund is worth holding in each bucket — without confusing asset class with manager quality.
This is not hypothetical. In our dataset of 1,066 equity funds, absolute Sharpe-based rankings placed 24% of Large Cap funds in the “top quartile” even though they delivered negative alpha versus their benchmark. Meanwhile, 18% of Small Cap funds that were genuinely top-category performers were classified “risky” purely because their absolute drawdowns looked scary on a universal scale.
How the health score works
Every fund is scored within its SEBI category — the same category the fund is registered under with SEBI. “Flexi Cap Fund — Direct — Growth” is compared only against other Flexi Cap Direct Growth schemes. A score of 80 means the fund ranks in the top 20% of its category on the weighted metric mix for a given profile. That’s all it means — nothing more.
Three profiles because one number isn’t enough
A single composite health score hides more than it reveals. A fund can be exceptional at generating returns but take brutal drawdowns. Another fund might look “average” on returns but deliver them with extraordinary consistency — ideal for a monthly SIP investor who cannot stomach volatility. We separate these by investor type:
- Growth (Return-seekers). Weights toward rolling CAGR and alpha vs benchmark — actual return generation is the primary signal.
- Protect (Capital-preservation). Weights toward down capture, max drawdown and Sortino — downside protection dominates.
- Compound (Long-term SIP). Weights toward rolling consistency, cost efficiency (inverted TER) and manager tenure.
INSIGHT This is the same fund scoring A, C, and B across three profiles. A mid-cap fund, outstanding at returns, weaker on downside protection, solid on consistency. Which profile matters to you depends on where it sits in your portfolio — not on some industry-wide “good fund” definition.
What goes into each profile
Each metric is first converted to a percentile rank within the fund’s SEBI category — so a higher raw value is always better after direction adjustment (e.g., lower TER → higher percentile). The weighted average of these percentiles becomes the profile score.
| METRIC | GROWTH | PROTECT | COMPOUND |
|---|---|---|---|
| Rolling CAGR 3Y/5Y | 22 | 5 | 12 |
| Alpha vs benchmark | 18 | 8 | 10 |
| Sortino ratio | 12 | 16 | 12 |
| Down capture ratio | 8 | 22 | 5 |
| Max drawdown (abs) | 5 | 18 | 3 |
| Rolling consistency | 10 | 12 | 22 |
| TER (inverted) | 8 | 10 | 18 |
| Manager tenure | 7 | 5 | 15 |
| AUM | 4 | 2 | 3 |
CHART
Score distribution across 1,066 equity funds — Growth profile (3Y window)
By construction, scores follow an approximately uniform distribution within each category. The bell-curve shape emerges at the cross-category level because larger categories (Sectoral/Thematic, ~473 funds) dominate by count.
What the grades look like in practice
Six representative funds across categories. Same fund, three profile scores — each telling a different story.
| FUND (ILLUSTRATIVE) | CATEGORY | 3Y CAGR % | GROWTH | PROTECT | COMPOUND |
|---|---|---|---|---|---|
| Mirae Asset Mid Cap (D-G) | Mid Cap | 26.4 | 91 / A | 58 / C | 74 / B |
| SBI Bluechip (D-G) | Large Cap | 15.1 | 72 / B | 84 / A | 88 / A |
| Nippon Small Cap (D-G) | Small Cap | 28.8 | 88 / A | 42 / D | 61 / C |
| HDFC Flexi Cap (D-G) | Flexi Cap | 18.2 | 68 / B | 75 / B | 82 / A |
| Axis Small Cap (D-G) | Small Cap | 19.4 | 54 / C | 70 / B | 80 / A |
| Franklin India Contra (D-G) | Value/Contra | 21.6 | 79 / B | 66 / B | 55 / C |
Scores and returns are illustrative and representative of the scoring methodology. Live scores available in the Punji app.
INSIGHT The key insight from this table: Nippon Small Cap scores D on Protect despite being a top-3 fund in its category on Growth. That D is not a criticism — it’s a statement that small-cap investing as a category involves heavy drawdowns, and this fund is below average within small-cap on downside protection. If you’re adding a small-cap position for growth, an A on Growth is what you want. If you’re near retirement, the Protect score tells you whether this is the right vehicle within the small-cap category at all.
What the score doesn’t tell you
The health score is deliberately scoped to be useful and honest about its limits. Three things it explicitly does not do:
- It doesn’t predict future performance. Historical percentile rankings are a measure of what happened, not what will. No fund score — ours included — has predictive power over a 1-year horizon that exceeds noise.
- It doesn’t compare across SEBI categories. A score of 85 in Mid Cap and 85 in Large Cap are not the same fund quality — one means top 15% of mid-cap, the other means top 15% of large-cap.
- It doesn’t account for your portfolio context. A Protect-A large-cap fund might be the wrong addition to a portfolio that already has three conservative large-cap funds.
WARNING One common misreading to avoid: A low Protect score does not mean the fund is “bad” or “dangerous.” It means the fund prioritizes return generation over downside protection within its category. For aggressive investors with a 10+ year horizon, a Protect-C Growth-A fund in small cap may be exactly the right exposure. The grade describes the fund’s behavior — not whether you should own it.
The methodology debate: what we chose not to do
We considered a single composite score. Every major rating agency and most robo-advisors publish one number. The problem is that a single composite requires a decision about what an “average investor” is — and that investor doesn’t exist. A 28-year-old adding Small Cap through SIPs and a 58-year-old rebalancing near retirement have diametrically opposed requirements.
We considered star ratings. Stars (1–5) are more intuitive than 0–100 scores. But stars compress too much information — a 3-star fund might be a 2.8 or a 3.4, and that difference matters when comparing two 3-star funds side by side. We kept the 0–100 scale with grade boundaries (A/B/C/D).
We did not normalize across time windows. A fund’s 3Y score and 5Y score are computed independently. An investor who wants to know “is this fund still performing under the current manager?” looks at 3Y. An investor who wants “did this fund survive a full cycle?” looks at 5Y. They should see different numbers because they’re asking different questions.
METHODOLOGY Technical notes. Input metrics: rolling returns, Sortino ratio, Sharpe ratio, alpha, beta, max drawdown, down capture ratio, tracking error — all sourced from Punji’s risk & returns analytics engine. Category percentiles recomputed monthly. Manager tenure sourced from AMFI scheme metadata. TER sourced from AMFI’s latest available monthly disclosure. Profile weights are version-controlled — any weight change triggers a version bump so historical scores remain auditable. Scores available in the Punji app for 3Y and 5Y windows.
Data: AMFI public disclosures. Analysis: Punji Research. Not investment advice.