Mid Cap Wins. It Has Never Won a Single Year.

Every month, NSE Indices publishes a two-page PDF called the Index Dashboard. It is a scorecard. Every index, one row each, returns over one month, one year, three years, five years, all on a total-return basis. It is dull, it is official, and almost nobody reads it.

I read nine of them — every 31 December from 2018 to 2025, plus the latest one, 31 July 2026 — and lined them up in order.

Here is what came out.

Over the twelve years from 1 January 2014 to 31 December 2025, the mid cap slice of the Indian market returned 20.74% a year. Large caps returned 14.27%. ₹10 lakh became ₹96 lakh in one and ₹49.6 lakh in the other.

And in that entire stretch, mid caps did not finish first in a single calendar year. Not one.

That is not a rounding artefact. It is the whole answer to the question, and it is the reason most people pick the wrong category.

First: what a "category" actually is

In October 2017, SEBI issued circular SEBI/HO/IMD/DF3/CIR/P/2017/114 and drew hard lines that had never existed before. A fund can no longer call itself whatever it likes. By that circular:

• Large cap = the top 100 companies by full market capitalisation.
• Mid cap = companies ranked 101st to 250th.
• Small cap = 251st onwards.

Those are not opinions. They are rank numbers.

Which is convenient, because NSE builds its broad-market indices on exactly the same rank numbers. From the July 2026 factsheets, in NSE's own words: Nifty 100 "represents top 100 companies based on full market capitalisation." Nifty Midcap 150 "represents the next 150 companies (companies ranked 101-250)." Nifty Smallcap 250 "represents the balance 250 companies (companies ranked 251-500)."

So each SEBI category has an index that is, by construction, the exact slice of the market that category is required to buy. SEBI also requires every scheme to be benchmarked against a Total Return Index — one that counts dividends, not just price. That is what the Index Dashboard reports.

Five categories, five indices, and one honest note on the sixth: Flexi Cap and ELSS funds both roam the whole listed market, so both sit against the Nifty 500.

Exhibit 1 — The leaderboard, as it was actually published

Below is the five-year table from nine real Index Dashboards. Not recalculated, not smoothed — the numbers NSE printed on that date.

Move through the dates. Watch the order change.

Nine dates. Three different answers.

At the end of 2019, five years of your life in Indian equity had paid between 4.1% and 9.9% a year, whichever category you picked. Small caps were dead last. Large caps had just returned 11.8% for the year while mid caps returned 0.6%, and every sensible person concluded that large caps were the answer.

At the end of 2024, small caps sat on top of the five-year table at 30.7% and large caps were last. Every sensible person concluded the opposite.

Both groups of sensible people were reading a real number off a real document. That is the trap: the five-year table is not a fact about categories, it is a fact about the last five years.

Exhibit 2 — Eight years, ranked

So let us stop reading five-year averages and look at who actually won each year.

Each column is one calendar year. First place at the top. The number on each chip is that category's total return for that year.

Read the top row.

In eight completed calendar years, exactly two categories ever finished first: large caps, four times, and small caps, four times. In the two dullest years on the board — 2018 and 2022 — the winner was large caps with +2.6% and +4.9%. In the three wildest — 2020, 2021, 2023 — it was small caps with +26%, +63%, +49%.

Mid caps finished first zero times. Second four times, fourth four times, first never — their average finishing position across the eight years is exactly 3.0 out of 5. Dead centre, every single year.

And yet:

That last box is the one worth sitting with. Small caps won four of the eight calendar years — more first-place finishes than anyone — and over the same eight years compounded at 11.96%, the lowest of the five categories. Large caps, which won the other four years, compounded at 13.04%.

Winning years and winning money are different sports.

Why the loser of every year keeps ending up ahead

The mechanism is not mysterious. It is arithmetic about losses.

A category that falls 26% needs to gain 35% just to be level. Small caps did that in 2018, and again — smaller, but four times in eight years they finished a calendar year underwater. Large caps did not have a single down year in that stretch; their worst was +2.56%.

Mid caps sit between the two, and the position is not a compromise. It is the point. They capture most of what small caps capture when the market runs — 25.6% against 26.5% in 2020, 44.6% against 49.1% in 2023 — while their worst year was −12.6% instead of −26.2%.

Give up a little of the best year. Give up a lot of the worst one. Compound for twelve years. That is the entire trick, and it never shows up on a one-year leaderboard, which is the only leaderboard most people ever look at.

Exhibit 3 — Your money, and the price of admission

Here is the part that is actually yours to decide.

The panel below runs your number through all five categories. Two things you control matter more than anything else on this page.

The regime switch. The same five categories, measured over two real windows. Twelve years from 2014 puts you through one full mid-and-small boom. Eight years from 2018 starts you at the top of the previous one, just before small caps fell 26%. Same categories. Very different answers. Nobody knows which of the two the next decade will rhyme with, and any article that pretends otherwise is selling something.

The flinch slider. Set the worst single-year fall you could sit through without selling. Categories you could not have held go grey — because a return you did not stay invested for is not a return you got.

Drag the flinch slider from left to right and the categories unlock in a very specific order, at very specific prices. Large caps cost nothing — their worst year in this window was a gain. Flexi cap costs 2.1%. Large & mid costs 5.2%. Mid cap costs 12.6%. Small cap costs 26.2%.

That ladder is the honest version of "which category gives better returns". Every step up the return table has a published admission fee, and the fee is paid in one bad year, in your own account, while your neighbour tells you he moved to FDs.

Exhibit 4 — Where the money actually went

Now the uncomfortable part, and this one is from AMFI's own monthly report rather than NSE's.

Between July 2021 and July 2026, Indian investors opened new folios in every equity category. But not evenly.

Small cap folios went from 59.2 lakh to 2.92 crore — up nearly five times, the fastest growth of any equity category, and now the largest single equity category in India by number of accounts. Large cap folios grew 1.5 times.

Over exactly that stretch, small caps were the worst-compounding of the five categories.

And it is still happening. In July 2026 alone, small cap funds took in a net ₹7,768 crore while large cap funds saw a net outflow of ₹1,322 crore. India's money is moving down the market-cap ladder, five years after the move that made that ladder look attractive.

I am not going to pretend that is irrational. Small caps returned 27.2% in 2024 and 49.1% in 2023. Anybody watching those numbers would want in. That is precisely the problem — the years that pull money in are the years after which the admission fee comes due.

So what is the answer?

If you want the single number, here it is, and it is the honest one:

Over the longest common window I can build from official data — twelve years, 2014 to 2025 — mid caps compounded fastest, at 20.74% a year. Over the shorter eight-year window they also came first, at 15.84%. On the latest published dashboard, 31 July 2026, they lead the five-year table again at 17.93%. It is the only category that is on top of all three.

But the number is not the lesson. These three are:

One. The category sets your ceiling and your floor together. You cannot buy the 20.74% without buying the −12.6% year that comes with it. There is no version of this where you get the mid cap return with the large cap ride.

Two. The best category on a one-year view is almost never the best category to own. Small caps won half the calendar years on this board and finished last over eight. If the reason you are looking at a category is that it topped last year's chart, that is a reason to be careful, not a reason to buy.

Three. Your holding period decides more than your category does. Look again at Exhibit 1 — at the end of 2019, every category on the board had paid between 4% and 10% over five years. Five years was not long enough to separate them. Twelve years was. If your money is going to be needed in three years, the interesting question is not which equity category — it is whether equity at all.

What I do in practice: a category mix is set from the horizon and the flinch number, in that order, and the fund is chosen after. Someone with a twenty-year goal and a genuine tolerance for a bad year has no reason to sit entirely in large caps. Someone who has never watched a portfolio fall 25% should not find out with small caps whether they can.

If you want to go through your own mix with the same data in front of you, that is a conversation I am happy to have.

Method and sources

Every number in this article comes from a published document. Nothing is estimated.

• Calendar-year and rolling returns — NSE Indices, Index Dashboard, the 31 December editions of 2018 through 2025 and the 31 July 2026 edition. All figures are total-return-index based, as stated on the document. Dashboards before 2018 report price returns and are not used.
• The 2014–2018 block — the five-year CAGR column of the 31 December 2018 dashboard, which covers exactly those five calendar years. The twelve-year figure is that block compounded with the individual calendar years 2019 through 2025. Cross-check: compounding my 2019–2023 yearly figures reproduces the five-year CAGR printed on the December 2023 dashboard to the second decimal for every index.
• Year-to-date 2026, and index construction — NSE Indices factsheets dated 31 July 2026 for Nifty 100, Nifty 500, Nifty LargeMidcap 250, Nifty Midcap 150 and Nifty Smallcap 250.
• Folios, flows and AUM — AMFI, Monthly Report for July 2026 and July 2021.
• Category definitions — SEBI circular SEBI/HO/IMD/DF3/CIR/P/2017/114, 6 October 2017.

Index returns are not fund returns. Funds carry an expense ratio and active decisions; they will not match these figures. Past performance is not indicative of future results. This article is general information, not personalised investment advice.