Elon Musk and Hilarious Market Stories | Samir Arora and Dinshaw Irani of Helios Capital

ELI5 / TLDR

Two veteran Indian fund managers swap war stories instead of stock tips. The lessons hide inside the anecdotes: never get close enough to a promoter to lose the ability to sell, always go visit the factory before you believe the story, and accept that the rules of running a fund will force you to trim your biggest winners on the way up — which is fine. They’ve owned 100-baggers and they’ve watched companies go to zero, sometimes within the same career.

The Full Story

Krish Kothari deliberately steers away from “what do you think of the market today” and asks Samir Arora and Dinshaw Irani of Helios Capital for the stories nobody else gets. They oblige.

Drawdowns and the slow death of caring about net worth

Arora starts by deflating the premise. The recent fall isn’t erosion — his long-short fund was up that month. Real erosion was 2008, and not because the market fell, but because they fought it.

“Not only the market fell a lot but we believed that it should not fall and we fought the market for a long time… saying India is not affected, we don’t have exposure to any company which has exposure to Lehman Brothers or subprime.”

The deeper lesson of 2008: being fundamentally sound is not enough. “If the world comes down it’ll pull you down because of association. You might be Indian market, they are that market, but in the end they’re all financial markets and flows are there.” Over the years, Arora says, the numbers stop meaning anything — you start with thousands of rupees, then millions, then a hundred million, and eventually only the percentage registers, not the level. As a small ritual, he collects bull and bear figurines. He used to own only bulls. Now he keeps seven or eight bears too.

The Trent story and “long term is a series of short terms”

In Alliance days (2001–03), the fund owned nearly 10% of Trent and 5% of Pantaloon. They met a young Noel Tata — pens in his shirt pocket — who explained West Side’s whole strategy in one line his daughter had given him: for 500 rupees, you should get two items. The business turned out far better than they imagined. But Arora refuses to romanticise foresight. His philosophy is that “long term is a series of short terms.” Even an insider guessing about three or four years out is guessing; an outsider has no business anointing anyone a “star for life.”

“It’s a liquid market. We are not selling. If you are doing well we are very happy. But why should we say today that you are a star for life? What will I do extra?”

Why he keeps promoters at arm’s length

The single most useful idea in the conversation. Get close to a promoter and you lose the ability to sell — you start phoning him for explanations instead of acting on what you already know.

“In 20 years I would not have had a meal with a promoter alone, ever, and consciously done that… wherever we have really lost big money is because we knew the promoter.”

The rule he was taught on day one at Alliance: if a promoter hugs you and invites you home, avoid the stock. Not because he’s lying — he may genuinely believe his story will be fine — but because friendship corrupts judgement. The clean version: “If I’m buying on the screen, then I will sell on the screen.” You only owe management loyalty if they gave you an IPO anchor allocation.

The mechanics of holding a 100-bagger

In the ’90s, Arora owned ~10% of a company that went up 100x while he held it. Mutual-fund rules cap ownership at 10%, which it hit almost immediately — so he was selling nearly every single day on the way up, and still held 8–10% when it crashed 30–40% in March 2000. Bajaj Finance later did 43x, yet its weight never exceeded 4–5% because they start positions at 2–2.5% and won’t let a single name balloon. His honest accounting note: the returns he quotes are first-purchase-to-last-sale, not what the fund actually captured across its whole position.

Selling: when price outruns earnings

Helios’s selling discipline is mechanical. When fundamentals stop keeping pace with price, one of the two has to give.

“Suppose the earnings growth is in mid single digits and the price growth every year is like 25–30%. You can hold it for a year or two, but beyond that… there’s no way you can justify those expanded multiples.”

HDFC Bank and Kotak were easy holds for 20-plus years because earnings and price both compounded around 20–25%. The hard ones are consumer names growing earnings 10–15% while the stock does 25–30% for years — eventually you have to take money off the table, trimming rather than selling in one clinical day. Do this ten times and you’ll be right eight; the other two keep running, and that’s fine.

Sizzle versus steak, and the businesses they refuse

When a confident, believable founder pitches 50% growth, Helios doesn’t argue — it goes back to the worksheet and checks whether the numbers are even possible. Zomato had no earnings when they bought it at 52, but the logic held: 60 rupees to skip parking, pollution, traffic and stay home watching Netflix is cheap. The quick-commerce thesis was equally concrete — a kirana is open 10 hours, a dark store 20, so half the day there’s no competition; break-evens are low; private-label optionality sits on top. They’ve been surprised by the scale of every winner, Varun Beverages included (10x for them, 25x from IPO).

But they won’t buy a business model that has never worked anywhere in the world — Indian timeshares, gym chains — where the hard sell is so aggressive the buyer regrets it within five days. If the world has done it and the logic holds, fine. If you’d be the first, pass.

IPOs, liquidity, and the China rotation

Arora dismantles the “can the market absorb $40 billion of IPOs?” worry. Supply isn’t an independent overhang — IPOs only come because the market is strong. If it fell 20%, the pipeline would shrink with it. On the foreign selling of that month (driven by a China rotation and the Trump-tariff trade), his bet was that it reverses: the money comes back, maybe not at much higher prices, but it comes back.

The zeros: ABCL, Educomp, and the Disney animation that wasn’t

The dark comedy. Arora once put a million dollars into the unlisted Amitabh Bachchan Corporation — 18 movies started at once, the rights to Miss World in Bangalore — and watched it go to zero. Irani recalls a Kolkata company whose management casually explained that a “330-million-pound” acquisition actually cost 30 million; and Educomp, which booked the margin on the software and the hardware and the projector. (“You’re adding the two margins.”)

The crown jewel is Irani’s TMT-boom field trip. A foreign broker tipped him on a Hyderabad company sitting on a “$200 million” Disney animation order; the stock was already up 20–30%. Irani insisted on visiting anyway.

“He took me to this building… newly constructed, just a shell. People lining up outside dressed in lungis and slippers, carrying folders. The promoter says: these are all the animators I’m going to hire for the Disney order. He doesn’t even have a laptop in front of him.”

The stock went to zero. Kothari calls it the Theranos move — the one analyst who bothered to physically check. Irani’s read: the buy side had grown complacent, assuming any company with a foreign contract would do well. The whole edge was being willing to drive to Hyderabad and look.

Why still bullish on India

Irani closes with the structural case, not a stock tip. Nifty 500 has compounded 15–16% in rupee terms over 25 years, riding nominal GDP growth of 10.5–11%. He sees no reason that breaks. What’s new is scale: India is now a ~$4 trillion economy growing at those rates, and no economy that large grows that fast.

“You can hate us, you can love us, but you can’t ignore us.”

His kicker: the wealth effect hasn’t kicked in yet. As Indians get wealthier they’ll take more risk, and risk means equity.

Key Takeaways

Long term is a series of short terms. Even insiders guess past 2–3 years. Don’t anoint a company a “star for life” — re-underwrite it every quarter, because the market is liquid and you can always change your mind.
Distance from promoters is a feature, not a flaw. Friendship destroys your ability to sell. Buy on the screen, sell on the screen. Big losses cluster where you knew the promoter personally.
Sell when price outruns earnings. A name compounding earnings at 10–15% but priced at 25–30% can be held a year or two, never indefinitely. Returns from PE expansion are borrowed; returns from earnings are owned.
Position-size discipline caps the winners too. Starting at 2–2.5% and never letting a single name run to 10–12% of the book means you trim multibaggers all the way up — and that’s the right trade, because the last 30–40% can vanish in a month.
Sizzle vs steak: go back to the worksheet. A good salesman’s story might be true; test whether the numbers are even arithmetically possible before buying the narrative.
Won’t buy a model that has never worked anywhere in the world. If the world has proven it and the logic holds (food delivery), fine. If you’d be the first (Indian timeshares, gyms), the high-pressure sell is the tell.
Go and physically check. The edge in the Disney-animation fraud was simply driving to the factory. Public data plus one site visit beats public data alone.
IPO supply is not an independent overhang. Issuance exists because the market is strong; if the market falls, the pipeline shrinks with it. Liquidity is plumbing — money changing hands, not money being consumed.
“150 companies do well” is permission to walk away. You never have to own a fraud, however well its stock is doing, because there are always ~149 other names that will perform.
On rejection, you’re rarely contrarian on the fact. The buyer of a name you reject usually agrees the management is dodgy or the stock expensive — he’s just weighing it against a counter-factor (a turnaround, a discount). You’re disagreeing on the trade-off, not the facts.

Claude’s Take

This is the good kind of investing podcast — no price targets, no “where’s the Nifty going,” just two people who’ve been doing this for thirty years telling stories that happen to carry the lessons. The signal-to-noise is high precisely because nobody is trying to sound smart.

The promoter-distance point is the keeper, and it’s the opposite of what most retail investors believe — they think access is edge. Arora’s argument is that access is a liability that quietly removes your sell button. The position-sizing honesty is also rare: most managers quote their Bajaj Finance 43x without admitting the position was capped at 4–5%, so the fund-level return was a fraction of the headline. He says it out loud.

The bull case for India at the end is the weakest part — “15% forever because GDP grows 11%” is a clean extrapolation that assumes margins, multiples and the rupee all cooperate, which over 25 years they may not. It’s a sales pitch wrapped in a structural argument, and he half-admits it. But it’s a footnote to an otherwise grounded conversation.

Docking nothing for the mid-roll sponsor read; the content earns its keep. The transcript is rough (auto-captioned, names mangled — “Shin” is Shinsegae, “gig boas” is the old BSE) but the ideas survive. An 8: durable mental models delivered through genuinely funny war stories, light on fluff, honest about its own accounting.