Deep dives on algorithmic markets, generative models, engineering decisions, and the mechanics of an institutional edge.
PB Fintech closed 36% lower on 24 September after IRDAI's consultation paper on insurance commissions, its worst day since listing. The business is intact: FY26 revenue up 37%, profit up 115%, no real competitor online. But the company lost money every year until FY24 and was trading at 130 times a two-year-old profit. The regulator supplied the excuse; the multiple supplied the fall.
Mukul Pandit
24 Sept 2026
Revenue doubled two years running, 19x earnings, zero OFS, and not a rupee of operating cash flow in three years. Where lab-grown diamonds are heading, where Nityas actually sells, what the empty QIB book means, and why this is an application for the long term and not for listing day.
A railway sleeper maker with real cash flows, a ₹582 Cr order book and 93% of revenue from ten customers, asking 23x earnings when its peers trade at 14x. What the RHP says, what the CBI matter actually is, where the subscription stands, and the price at which this becomes worth a look.
The Adani Group closed a six-year SEBI case on 28 September for about ₹1.48 crore ($154,100), covering four listed companies and 14 directors. One camp says SEBI went easy; the other says the small amount proves nothing was wrong. Both miss how the process works. A SEBI settlement amount is set by a formula, carries no admission of guilt, and closes only one specific case. It says almost nothing about how serious the allegation was.
₹2.08 lakh crore of foreign portfolio money has net left India since April 2024. Nifty is the cheapest it has been in five years and India pays the highest sovereign yield among major markets. Neither is bringing the money back. Six questions, answered with NSDL, exchange and central bank data.
A ₹218 Cr fresh issue from a 26-year-old government IT contractor: 4x subscribed, 30% return on net worth, a ₹1,205 Cr order book, and negative operating cash flow in its best profit year. What the RHP says, who the anchors are, where FII money in IT is going, and where it lands.
The Sanctioning Russia and Iran Act, signed on 18 September 2026, lets Washington put a tariff of up to 100% on the five largest buyers of Russian crude. India is second on that list. What the law does, why India is on it, what sits on the other side of the ledger, and why the first determination on 18 October matters for the rupee, refiners and exporters.
The ₹303.9 crore IPO closes at ₹139 with real, cash-backed growth and 95% of the money going into capacity. Against that: a tax demand equal to 26.6% of net worth, 97.7% of revenue from one state, and a pre-IPO round twelve months ago at ₹270. The bull case, the risk case, and what the RHP says.
Snapdeal's parent closes its ₹420 crore IPO at ₹32 a share. SoftBank and Foxconn exit at 92% and 96% losses, tangible net worth is negative, the marketplace has never made money, and the one asset with a visible value is already listed. What the RHP actually says.
Moneyview opens its ₹1,092 crore IPO on 24 September at ₹32 to ₹34, a ₹5,985 crore valuation for an app-based personal-loan platform with 14 crore registered users and ₹21,380 crore of loan AUM. Total income rose 43% in FY26 to ₹3,404 crore; profit rose from ₹240 crore to ₹243 crore. The gap between those two numbers is credit cost, and ₹325 crore of the fresh issue is earmarked to absorb more of it.
Foreign investors bought ₹1,617 crore on 23 September, their first net buy after Monday's ₹3,810 crore of selling. September is still ₹9,810 crore of net FII selling on the exchanges (₹15,237 crore on NSDL's count, which includes IPO subscriptions), against more than ₹45,000 crore of DII buying. That has been the pattern since October 2024, and the Nifty at 23,447 is still about 7% below where it was a year ago. Why the DII bid sets the floor but not the direction, what valuations say now, and what would make one buy day into a turn.
Elevate Campuses opened its ₹2,100 crore IPO on 23 September: entirely fresh issue, no offer for sale, price band ₹343 to ₹362. No promoter is selling shares. But ₹1,100 crore of the proceeds buys K-12 school assets from promoter group entities and ₹750 crore repays debt, so most of the money is not building anything new. What the structure means, what the FY26 profit jump hides, and why the QIB bucket by 25 September is the only number to watch.
Optiemus Infracom hit its 20% upper circuit at ₹708.90 on Tuesday after a 7:14am filing: a binding term sheet with Nothing Electronics for a JV to sell CMF phones in India, with Optiemus taking 51.1%. Volume ran 20.9x normal on ₹541 crore of turnover, but only 17.82% of it was delivered against a 47.1% norm. What the term sheet actually commits to, what the tape says, and why the Q1 margin line matters more than the headline.
Websol Energy System went from ₹26 crore of revenue and a ₹121 crore loss in FY24 to ₹1,049 crore of revenue and ₹303 crore of profit in FY26. The FY24 year was a deliberate rebuild: ₹223 crore of capex with the plant mostly idle. The balance sheet is now net cash, the sector has policy money behind it, and the stock is down about a third in a year. What the numbers say, what the holders are doing, and the one figure that still needs watching.
Varmora Granito's ₹708 crore IPO opened on 22 September at ₹140 to ₹148. Revenue grew 1.4% and then 4.7% over the last two years, ROCE is 9.9%, and 55% of the issue is a seller cashing out. At the top of the band the stock is priced at 60.7 times FY26 earnings, above Kajaria (40.3x in the RHP's own peer table, 34.5x on today's price) and Somany (28.6x). A walk through the numbers using the method I use for every IPO.
On Monday, 21 September, KPI Green Energy rose 12.9% on ₹581 crore of turnover and 22 times its normal volume. Delivery was 10.06% against the stock's norm of 45.4%: roughly nine-tenths of the day's trading never left the trading desk. The only filing was a notice of an analyst meet three days later. A short lesson in the one number that separates a real move from trader noise.
On 18 September Avalon Technologies jumped 13.8% on eight times its normal volume with no filing and no news. Two Vanguard index funds bought ₹177 crore of it in the bulk-deal file, on the day the FTSE Global Equity Index review took effect. That is not a leak and not a breakout. It is an index fund doing its paperwork, and it is one of the most useful patterns a retail investor can learn to spot.
On 15 September the Nifty fell 1.2%, foreign investors sold ₹2,978 crore, and the rupee closed at a record low near 96 to the dollar. Nifty IT rose 2.2%, and seven of the ten biggest F&O gainers were IT companies. Two things explain it: the rupee, which is the boring one, and a call from the leaders of the AI industry to slow AI down, which is the interesting one. Here is what each is worth, and what it would take for the rally to stick.
Brent is near $107, the rupee is at 95, and foreign investors sold ₹2,251 crore of Indian oil and gas stocks in the last NSDL fortnight, more than any other sector. At the same time, every one of those companies is about to report higher revenue. Both things are true. Here is the simple mechanics of why, company type by company type.
On 11 September an early investor sold ₹155 crore of BlackBuck in a single block deal, and the stock closed up 7.2%. That is not a contradiction. The seller was only half the story; the buyer, the price, and what the tape did afterwards were the other half. A short lesson in how to read a block deal, using one real day.
After a decade of waiting, India's largest stock exchange is coming to market. NSE is a near-monopoly cash machine: about 93% of cash-equity turnover, the world's largest derivatives exchange, roughly 75% EBITDA margins and 33% return on equity. The surprise is the price. At ₹1,700 to ₹1,785 it is being offered below its own unlisted market price and at a lower P/E than smaller, listed BSE. The catches: it is 100% an offer for sale, its biggest revenue line (F&O) is exactly what SEBI is curbing, and it carries a scarred governance past. Here is the full read, with a clear NSE versus BSE comparison and the red and amber flags.
ARCIL is the opposite problem from most recent IPOs. It is not expensive. At about 11x earnings and 1.47x book, it looks reasonable for India's oldest and largest private asset reconstruction company. The catches are a 100% offer for sale and a fourfold jump in debt used to buy more stressed assets, even as recoveries dipped. But the timing could actually work in its favour: with bank write-offs rising and lenders increasingly selling bad loans to ARCs at a discount for quick relief rather than a long fight, ARCIL's pipeline may be about to widen. My lean: worth a smaller subscription, if you accept it is a cyclical, lumpy recovery business.
Karamtara is a good business in a good industry. It is the largest integrated Indian maker of solar mounting structures and trackers, growing revenue at about 33% and profit at about 49% a year, with a healthy 23% ROCE. The best part of the IPO is that most of the fresh money goes to cutting debt, not to promoters cashing out. My problem is the price. At around 35.7x earnings, well above the peer median near 19.6x, and with heavy debt and working capital, the valuation does not leave enough room for things to go wrong. I would watch it rather than subscribe at 254.
Rentomojo is a genuinely good business - a profitable, cash-generating furniture-rental platform growing 45% with 41.5% EBITDA margins and 25% ROCE. I actually like it more after studying Furlenco, not less. But the IPO is 88% an offer for sale, the FY26 "142% profit jump" is heavily flattered by a one-time ₹36.6 cr tax credit (real PBT was ₹67.7 cr), and at ~₹4,246 cr you are paying ~41x reported - closer to ~80x once you normalise. The real question: if the smart early money is selling ₹1,106 cr at this price, what growth am I paying for?
This is that rare IPO where I like the business, the growth and the use of proceeds together - even though the valuation is not cheap. A 100% fresh ₹720 cr issue funding real data-centre capex; revenue ₹287→₹472 cr and PAT ₹14→₹121 cr in FY24-26; EBITDA margin expanding 35.6%→49.6%; debt falling to 0.08x D/E. The catch: ~42x FY26 P/E at the top band (~66x at the current GMP), customer concentration, and big-tech competition. Good business, expensive valuation, worth watching.
Rs 22,006 crore of admitted claims. Rs 6.25 crore of proposed repayment. A 99.97% haircut - and it very nearly went through. The Subhash Chandra personal-insolvency saga is not just one promoter's story; it is a stress test of what a personal guarantee is actually worth in India, and the answer, until a five-member NCLT bench stepped in, was close to nothing. My honest read on the wrong-way risk it exposes.
India's August 2026 GST came in at nearly Rs 2 lakh crore and the headlines cheered. Look beneath the number and the story flips: domestic transactions grew a soft 9.3%, the real muscle came from a 29% jump in GST on imports, and a 68% surge in refunds quietly drained the pool - so net collections grew just 8.3%. Are we celebrating a thriving domestic economy, or watching consumers foot a rising import bill while the gross headline does the cheerleading?
Symbiotec is a genuine global No.1 in steroid and hormone APIs - 38.2% world share in corticosteroids, US-FDA and EU-GMP plants, 26.6% EBITDA margins. The grey-market premium is running 40%+. And I am still sitting this one out. Here is the uncomfortable part the loud GMP hides: 91% of the issue is an OFS (a promoter/PE cash-out, not growth capital), it is priced at ~58x on modest growth, its core markets grow just 1-3% a year, and the RHP carries a heavy promoter pledge, tax litigation and a Rs 100 cr civil claim.
ITC is one of India's most profitable companies and a dividend machine - yet it has trailed almost every index it belongs to for the better part of a decade. A cigarette business throws off ~80% of profit and funds everything else; the FMCG, paperboard, agri and (now demerged) hotels businesses add revenue at far thinner margins. This is a full study of what you are actually buying, why the market discounts it, and the right reasons to own it - or not.
India's No.1 air freight forwarder by AWBs (ranked #1 by World ACD four years running), in a >$250bn logistics market, priced at ~32-38x when listed peers sit at 54x-1,548x. It looks like a steal. Then you find the other side: wafer-thin margins, borrowings up 75% in two years, and an open Economic Offences Wing FIR alleging ~Rs 480 cr of fraud with no provision made. My read inside.
On a friend's insistence I started reading up on the sugar sector, and some of what I found was eye-opening even for me. A sugar company is really three businesses. Four levers drive its earnings. Ethanol is the growth engine with a catch. And there is one efficiency number - the recovery rate - that quietly decides margins. Sharing all of it, plus a downloadable ready reckoner.
India's first standalone listed PE firm. Income ₹104cr to ₹158cr (FY24-26), FY26 PAT ₹82cr, priced at ~22x. Looks reasonable. But a smooth P&L hides two risks: carried interest makes earnings lumpy, and you are really buying an AMC whose fortunes ride on its own funds - with ~82% of fresh proceeds flowing straight back into them. Investable yes, low-risk no.
A ~Rs 2,949 crore block deal in Paytm's parent set off "the founder is cashing out" panic. He isn't. The seller on paper is Vijay Shekhar Sharma's entity Resilient, but the money flows to China's Antfin - the tail end of a 2023 arrangement built to clear India's FDI rules. Explained simply.
SEBI changed how the closing price is set for F&O stocks - from a 30-minute average to a proper auction. Explained like a school bell: why it matters, who it helps, who it makes life harder for, and the cautionary tale of Hong Kong's HSBC closing-auction shock.
Technocrats Plasma's ₹61 Cr IPO looks a steal - cheapest among peers at ~15.5x, ROCE ~49%, all fresh money, no founder selling. But just 10 clients drove ~84% of last year's sales. My honest read on why I would wait rather than chase listing day.
Credent Connect's ₹94 Cr IPO closed ~80x subscribed. It runs no labs - it is the picks-and-shovels behind them. Low margins and an acquisition-driven profit jump, but a tiny ~₹344 Cr valuation standing in front of a ₹1.54 lakh crore market. My honest long-term read.
For years the playbook was simple: FIIs sell, market falls. Not anymore. A demat boom from ~4 crore to ~23 crore accounts, relentless SIPs and rising DIIs mean domestic money now absorbs foreign selling - and in March 2025, DIIs overtook FIIs for the first time in 22 years.
Shiprocket's ₹1,617 Cr IPO closed ~99x subscribed. My honest retail read: nearly half is old investors cashing out (some at a loss), it still loses money so no dividends or near-term re-rating, and it priced as a down-round.
Nvidia lined up a $500 billion pool to fund AI data centres, and guaranteed the chips inside won't lose too much value. It is genuinely clever, and it quietly ties Nvidia's balance sheet to the AI boom staying strong. Here it is in plain English.
Milky Mist's ₹1,553 cr IPO: a genuinely differentiated, high-margin dairy business - but one still more leveraged than Dodla and Parag, and priced at ~85x earnings. My honest read on quality versus price.
Over squash, a friend raised a great question: won't forced ETF buying make BSE jump when it joins the Nifty 50? It sounds right. So we looked at the last 6 additions - five of six actually lagged the market on the way in, then recovered after.
Taj GVK just posted a ~4x jump in profit - but ₹283 crore of it is a one-time, non-cash accounting gain. Strip it out and the "5x P/E deep-value" story mostly disappears, leaving a genuinely good hotel business at a fair-ish price.
A ~₹1,370 crore pharma company quietly remaking itself - revenue and profit compounding at ~40% a year, debt paid down to almost zero, and a valuation ~40% below its peer floor. The earnings sprinted; the price only jogged.
For the third year running, Sakthi Sugars' auditors couldn't give a clean opinion - ₹252 crore of interest receivable sits on the books with no provision against it. Add loan defaults, overdue statutory dues, and fundamentals that have collapsed, and the picture gets hard to ignore.
CG Power trades near 115x earnings while EPS has grown just ~7% a year. So what is the market actually paying for? A data-first look at the premium, the semiconductor bet behind it, and what the ownership shifts are quietly saying.
An estimate turns a reaction into a decision. Part 3 of the series is about what a valuation actually does once the market moves – the margin of safety, why price-first trading fails most people, and what fundamental analysis really is.
Valuation is not a formula - it is a question: what is this business worth to its owners? Part 2 of the series reframes valuation as disciplined thinking, not spreadsheet math.
Ask an investor why they bought a stock and answers come easily. Ask what the business is worth, and the confidence disappears. Why so few retail investors ever value a company - and why it isn't entirely their fault.
A step-by-step walkthrough of Discounted Cash Flow analysis calibrated for the Indian equity market - from estimating free cash flows to choosing the right discount rate.
FII outflows can trigger 5% corrections in a week. DII buying absorbs them. Understanding these two opposing forces is essential for timing entries and exits in Indian equities.
A systematic multi-factor screening approach to filter the NSE universe down to high-probability investment candidates - without reading a single annual report first.
How to apply the father of value investing's simple but powerful intrinsic value formula to Indian equities - and where it works, and where it breaks down.
Different sectors lead the market at different stages of India's economic cycle. Knowing which sectors to overweight when can add 3–5% annually to your equity returns.
Net profit is an opinion. Free cash flow is a fact. Here's why the gap between these two metrics reveals the most important truth about an Indian company's financial health.
Price momentum - buying winners and avoiding losers - is one of the most well-documented return anomalies in global markets. Here is what the Indian data says, and how to implement it.
Promoter shareholding is one of the most India-specific signals in equity analysis. High stake is not always good. Low stake is not always bad. Here is how to read it correctly.
P/E is the most quoted valuation metric. EV/EBITDA is what institutional analysts actually use. Here is why, and how to choose the right multiple for each Indian sector.
SEBI's algorithmic trading regulations have evolved significantly. Retail investors using any automated or semi-automated strategy now operate under a specific regulatory framework. Here is the complete picture.