Report
The Skill Report
How much of Indian mutual fund performance is genuine stock-picking skill — and how much is just the market?
Inaugural edition · Data as of June 2026 · Based on AlphaPicker's factor decomposition of 606 active Indian schemes
The question nobody answers honestly
Every "best funds" list ranks by past returns. But a return is mostly the market rising, plus a few cheap, well-known style and sector tilts. Strip those away and you're left with the only thing that's genuinely the manager's doing — and the only thing worth paying active fees for: stock-selection skill.
We measured it. For every active Indian scheme with enough disclosed-holdings history, we decomposed returns against a multi-factor risk model, removed market, style and sector effects, and isolated what remains. Here's what the data says.
Finding 1 — Skill is real, but small and astonishingly uneven
Of the 280 funds with at least 12 months of disclosed holdings (enough to judge stock-picking), 67% showed positive stock-selection alpha and 32% negative. That sounds encouraging — until you look at the size and spread:
- Median stock-pick alpha: just +2.2% a year.
- The spread runs from −22% to +22% a year. The middle half of funds sit between −1.4% and +6.5%.
- Only 51 funds (about one in six of those rated) delivered +8%/yr or more of genuine selection alpha.
The takeaway: real skill exists, but it's modest for most and concentrated in a minority — and the noise is enormous. You cannot eyeball it from a returns chart. A fund returning 18% might have +6% of skill or −2%; the chart looks identical. Measuring it is the only way to tell.
Finding 2 — Skill lives in the corners of the market, not the mainstream
Where managers add stock-picking value is strikingly consistent with how efficient each part of the market is:
| Category | Median stock-pick alpha (%/yr) | % of funds positive |
|---|---|---|
| Small Cap | +8.4 | 93% |
| Mid Cap | +6.2 | 64% |
| Hybrid | +5.2 | 72% |
| Multi Cap | +4.4 | 100% |
| Value / Contrarian | +3.3 | 71% |
| Thematic / Sectoral | +2.0 | 67% |
| Flexi Cap | +1.7 | 61% |
| Focused | +0.4 | 60% |
| Tax Saver (ELSS) | −0.2 | 50% |
| Large Cap | −1.9 | 40% |
The least-efficient corners — small- and mid-cap, where information is scarce and mispricing is common — are where managers earn their keep. The most-efficient, most-researched corner — large-cap — is where active managers struggle most, with the median fund destroying value versus a passive alternative. This mirrors decades of global evidence, and it has a blunt practical implication: the case for paying active fees is strongest in small/mid-cap, weakest in large-cap.
Finding 3 — Managers can pick, but they give it back on timing
Decompose the average active manager's value-add and a pattern emerges: their stock-picking contributes positively (+3%/yr at the median), but their allocation and timing decisions subtract almost as much (−3%/yr). Style and sector tilts wash out close to zero on average.
In plain terms: the skill is in the picking; the leakage is in the timing. Managers who could quietly compound their selection edge often hand it back trying to time the market or rotate allocations. It's a reminder that "an active manager" is really several decisions bundled together — and only one of them (selection) shows durable skill.
Finding 4 — "Beating the benchmark" is not the same as skill
Here's the finding that should end the benchmark-beating obsession. The share of funds that beat their category benchmark swings wildly — from 0% (Mid Cap) and 3% (Large Cap) to 90% (Flexi Cap) and 100% (Multi Cap) — over the same period.
Yet mid-cap funds, where 0% beat the benchmark, showed some of the highest stock-selection alpha (median +6.2%/yr, 64% positive). How? Because beating a benchmark depends on the benchmark's own run, the prevailing style environment, and fees — forces that can completely swamp genuine skill in either direction. A manager can pick stocks well and still trail the index; another can beat it on a lucky style tilt with no skill at all. Benchmark-beating measures the environment as much as the manager. Decomposition is the only way to separate the two.
How we measured this (the method, in the open)
Transparency is the point, so here's the engine:
- Universe: active, regular-plan, growth-option schemes across equity, hybrid and solution-oriented categories. Index funds, ETFs, fund-of-funds, passive and arbitrage funds are excluded — they aren't trying to pick stocks.
- The score has three pillars: Skill (45%) — the statistical consistency of pure stock-picking, decomposing each fund's disclosed portfolio against a multi-factor model and testing whether what remains is skill or noise; Conviction (45%) — patience (low churn), concentration, and real weight behind top ideas; Cost (10%) — every rupee of fees is a rupee off your return.
- What we ignore: past returns, star ratings, brand and fund size — none predict future performance, and all are how funds get sold.
- Validation: signals must predict out-of-sample. At each historical month we rank funds using only data available then, and measure how that ranking predicted the next 12 months. The current formula was directionally correct in 76% of test windows; ranking by past returns alone predicted nothing.
What this means for you
- Stop ranking by past returns. They conflate market, factors and skill — the three things you can't tell apart by looking.
- Demand more from large-cap active fees. That's where managers add the least; a low-cost index alternative is a high bar to clear.
- Don't mistake a good year for a good manager. Skill is small and noisy; one year proves little. Persistence is what matters — and what we keep measuring.
- Beating the benchmark isn't the test. Genuine selection skill is.
About this report
The Skill Report is built on AlphaPicker's factor decomposition of Indian mutual funds, refreshed monthly. This inaugural edition uses our current production model; future editions will draw on a deeper, survivorship-free factor history for stronger persistence analysis. Methodology: how our score works.
Educational information only. This report provides non-personalised, aggregate analysis and does not constitute investment advice or a recommendation to buy, sell, or hold any security. AlphaPicker is not a distributor and earns no commissions. Mutual funds are subject to market risk; past performance is not indicative of future results. Consult a SEBI-registered investment adviser for advice specific to your situation.