The market rose. The leaders swapped.

The Nifty 500 gained +2.0% in July and closed the month at its high, 23,460.7, just 2.6% below the January peak. About 82% of the fall has now been retraced. Read the three columns below together: over one month large caps led; over three and six months, the small-cap order still holds.

Last month this table ran in one direction: the smaller, the better. In July’s column it runs in exactly the other direction. And the sector that broke the mood in June broke it again, the other way: technology was July’s best theme at +11.9%, after being the worst place in the market for a year.

The bounce trades stopped working

Last month’s issue read the recovery as a bottom-up rally: small and micro caps carried it, participation beat protection, and technology was the worst place to be. July kept the recovery going and reversed every one of those legs. The universe this month: 268 diversified direct-growth funds across nine categories.

Large caps beat small caps for the first time since the recovery began. By category, the share of funds beating the index in July against the rolling quarter:

The driver sat in one sector. The Nifty IT index rose about 16.7% in July, its best month in two years, while the global AI hardware trade sold off sharply. Global money rotated out of crowded chip positions and into markets and sectors that had lagged, and Indian software had lagged more than almost anything. The rally reached India as a positioning trade, not an earnings verdict.

The cleanest test of a reversal is what happened to last month’s named funds. We tracked both lists:

The pattern is precise. The sliders whose problem was style, a large-cap tilt like Parag Parikh’s, were rescued by the rotation and beat the index. The sliders whose problem was the cash whipsaw stayed broken: all three Samco funds lost money in a rising month. And among the rally leaders, most drifted back to the pack; the exception is the fund in Section 09.

So the reader’s question for August is simple: is this a one-month positioning snap, or the start of a leadership change? The rolling 3M and 6M columns still belong to small caps. July’s column does not. When the windows disagree, the honest answer is to watch, not to extrapolate.

The ranks turned over

Each fund’s percentile within its category on the rolling 3-month window, this issue against last issue. The climbers are large and mid-cap funds the bounce had left behind; the fallers include names that sat in June’s top-performer tables.

Bank of India Large Cap and SBI Large Cap both sat in June’s large-cap top five. One rolled month removed them. The notable climber is HDFC Mid Cap: one of the largest funds in the category, bottom of the table in June, now above the median. When giants move off the floor, flows follow.

The best funds of the quarter

Tables rank on the rolling 3-month return (May to July). The bracketed number is where the fund stood in last issue’s table on the April to June window. Small caps show almost pure persistence; the multicap mandate shows the reshuffle.

UTI Flexi Cap, 54th last issue and 2nd now, is the reversal in one line: a quality-growth book the bounce left behind, pulled up by July’s rotation.

The same five funds as last issue, reshuffled within themselves. On the rolling quarter the small-cap leaderboard has not changed; only July’s contribution has.

Last issue this table was wall-to-wall US and AI overseas funds. All of them are gone. Domestic technology funds entered instead: Digital India made +13.7% in July and is still negative over a year, which is the whole reversal in two numbers.

These tables are a record of what happened, not a list of what to buy.

IT rose faster than funds bought it

Sector weights of the diversified book, end-June against end-July. IT is the biggest riser; banks the biggest cut.

Read the IT row carefully. The weight rose 0.61 points, but IT prices rose about 16% in the month; the price move alone accounts for nearly all of it. The trades were mixed. Funds sold Infosys, cutting 4% of their shares, about ₹1,753 crore, while adding elsewhere in the sector: net buying of ₹897 crore across the 27 names traded, small change against a ₹1.3 lakh crore IT book. The book got more IT-heavy mostly because IT went up. Managers leaned in only slightly.

The real trades of the month, measured by change in shares held, valued at month-end prices:

Diversified direct-growth funds, month-end disclosures. Trade value approximated as the change in shares held, valued at the end-July price. The Torrent addition and the JB Chemicals exit in the next section are two sides of one merger, not two decisions.

New money went into IPOs

Funds opened positions in 34 companies absent from the June book, worth ₹6,576 crore, twelve times last month’s total. The character changed too: June’s new buys were small-cap fishing trips; July’s are institutional-size IPO anchor allocations, plus three demerger allotments that arrived without anyone buying anything.

Top 12 of 34 new names by position size. * listing-day close to 31 July. † move since listing, for companies that listed after the July book date. “Since 31 Jul” runs to 14/08/2026. Demerger allotments have no traded price yet.

Two positions deserve the reader’s attention. INDO-MIM entered 42 of 268 fund portfolios in its first month, the broadest instant adoption in this data, and is up 9.4% since month-end. And the largest AMC listing of the year is underwater: the SBI Funds Management anchor position closed July 3.4% below its listing price and has slipped another 5.1% since. Anchor books are locked in; the funds that took them are down on day one and cannot leave.

Exits were a non-event. The one large disappearance, JB Chemicals at ₹4,700 crore, is the Torrent amalgamation: shareholders received Torrent stock, so the position moved rather than sold. Genuine full exits totalled under ₹30 crore.

Pharma held. Technology woke up.

Last month this section said pharma led and technology was the worst place to be. Half of that held: pharma remains the only theme still beating small-cap funds over six months. The other half inverted in four weeks: technology funds returned +11.9% in July, the best of any theme, and are still down 7.8% over six months. Both facts belong in the same sentence.

Median fund return per theme, ranked on 6M (marked ▼). 1M is context, never a ranking basis. The standing benchmark: small-cap funds’ 6M median is +16.4%. Only pharma clears it.

Infrastructure, June’s 3-month leader, is the quiet casualty of the rotation: barely positive on the quarter and negative in July. The rotation did not lift everything that was cheap; it lifted the one cheap sector with a global story attached.

The reversal was global

Last month’s line was that “tech is dead” was a domestic-IT story, not a global one: US and AI overseas funds were the best corner of the market while Indian IT bled. July inverted the pair. US and AI funds fell a median 6.2% as the AI hardware trade sold off, while domestic technology funds rose 11.9%. The money that left crowded global chip positions is part of what landed on Indian software.

Median fund return per region, overseas fund-of-funds, gold and commodity funds excluded.

The 12-month columns are a reminder of what a year of the AI trade built: even after July’s fall, US and AI funds are up 30% over a year and emerging markets 47%. One bad month has not undone that. But the direction of travel flipped, and for the first time in this publication’s short life, the domestic technology fund beat the overseas one.

Motilal Oswal Multi Cap: it won both months

The outlier of the month’s data. In June’s rally it went from the bottom of the multicap mandate to the top, gaining 23.0% against the index’s 12.0%. Then July arrived, the bounce trade died, and every other rally leader drifted back to the pack. This one made +6.8%, near the top of the whole universe again, in a month that rewarded the opposite style. A ₹4,492 crore fund holding 38 stocks, and the only name that led both regimes.

Windows measured month-end to month-end (the January 2026 point uses the 02/01 peak). Edge is the fund’s return minus the index’s, in percentage points.

The record is not a smooth one, and that is the point. This fund defended the 2024-25 falls better than almost anyone, then missed the late-2025 rally entirely, the stretch that left it with a negative 1-year number in our June issue. It fell harder than the index in the January crash, hitting the bottom of its category. Then it changed character completely: top of the category in the recovery, top again in July’s rotation. Five wins in seven episodes, but the two losses were the market’s most crowded moments.

Character: a concentrated book that swings between the bottom and the top of its category within a year. Its 12-month return, +5.1%, tells you nothing about the ride: category-worst and category-best sit inside that single number. The July test is the interesting one, because it won without its style tailwind. August tells us whether that was skill or a hot streak extending.

Union Small Cap, featured last month, returned +1.7% in July against a small-cap median of +1.3%: in character, modestly ahead, nothing to revise.

How to read this issue

Universe. All diversified direct-growth equity funds across nine categories (Flexi, Large, Large & Mid, Mid, Small, Multi, Contra, Value, Focused): 268 funds this month against 282 last month. The difference is mostly new funds entering the universe as they accumulate the return history these tables need, plus schemes without a full one-month record on the anchor date.

Windows. All rankings use the rolling 3-month return (May to July this issue). One-month figures appear only as context and are never a sort key. Rolling 6M spans end-January to end-July, almost exactly the fall-and-recovery episode. Percentile ranks are within category, direct growth.

Holdings. Portfolio figures use month-end disclosures. Trade values are approximated as the change in shares held, valued at month-end prices; corporate actions (the Torrent and JB Chemicals amalgamation, the GSPL and Triveni demerger allotments) are flagged where they occur rather than counted as decisions.

All figures as on 31/07/2026 unless stated. Post-month-end prices run to 14/08/2026.