Tom Sosnoff: Inside the Mind of a Trading Legend | In The Money by Zerodha Podcast 01

ELI5/TLDR

Tom Sosnoff spent 20 years as a pit trader in Chicago, then built two trading platforms (Think or Swim and Tasty Trade) that he sold for hundreds of millions and over a billion dollars. His core message: after watching markets every day for 43 years, he has no idea what happens next, and neither does anyone else — so stop predicting, and instead make lots of small, probability-based bets and let the math grind out a profit. He bets against trends, sells options for income, keeps every position tiny so a freak market move can’t sink him, and thinks human judgment still beats trading robots for ordinary people. He also has sharp, slightly bruising opinions on Indian regulators, Jane Street, and the idea that you can ever “think about it.”

The Full Story

The man and the résumé

Sosnoff is a self-described “boomer” who studied political science at SUNY Albany in the early ’70s — partly because finance and business degrees didn’t really exist yet. He fell into a Wall Street job out of college, discovered the Chicago trading floors, and was there “basically next day.” He traded options in the pits for two decades, was one of the original traders of S&P 100 index options, then founded Think or Swim (sold to TD Ameritrade for ~$700M) and Tasty Trade (sold to IG Group for $1.1B in 2021). At the time of this interview he had just done his last Tasty show and was starting a new venture, variously called Lost Dog and One Lucky Dog.

What floor trading actually was

The romantic image of the pit trader reading the market is, he says, wrong. The job was almost mechanical.

“Our job was to take the other side of what anybody else wanted to do because we would get a small amount of edge of theoretical edge to do the other side. So it had no… it didn’t matter what we thought like whether we were bullish or bearish or whatever.”

A few thousand people like him stood in the pit and took the opposite side of every customer order, collecting a tiny built-in profit margin (the “edge”) each time. Survival was about risk management, not prediction — only “five or 6% of the people survived.” The ones who blew up “trade too big, they have an opinion.”

The big admission: no edge

The most striking moment is Sosnoff insisting he has no forecasting ability at all.

“Nobody could watch more ticks in the S&Ps than I have over the last 40 years. So that should give me an edge. But it doesn’t… I don’t have any edge over anybody else in knowing what’s going to happen next.”

When the host pushes — surely you’ve developed intuition? — he holds the line. Markets are “incredibly difficult, irrational,” everyone is watching “the exact same movie,” and “there’s only one movie playing at a time.” His only edge is comfort: he’s seen everything once or twice, so nothing rattles him.

How he actually trades

If you can’t predict direction, you make money from structure and probability instead. A few mechanics worth unpacking:

Selling premium. Sosnoff is an option seller. When you sell an option you collect cash upfront (the premium) and take on the obligation to pay out if the market moves far enough against you. Most of the time it doesn’t, so you keep the cash. It’s like being the insurance company instead of the person buying insurance — you win small and often, and occasionally pay a big claim.

Law of large numbers. Because any single trade is a coin-flip-ish gamble, he makes many small trades so the underlying probabilities have room to play out. One trade is luck; a thousand trades is statistics. This is why his mantra is “trade small, trade often.”

Position sizing as the only real defense. Selling options exposes you to rare, violent moves (“tail events” — the August 2024 and April 2025 crashes come up). His answer is unglamorous: keep each position small enough that a blow-up is survivable.

“I don’t know that there’s another way you can get around outlier risk other than trade size.”

Implied volatility as the dial. Implied volatility is the market’s own estimate of how much a stock will jiggle around — and it’s baked into option prices. When volatility is “cheap,” he uses one set of strategies; when it’s “expensive,” he takes on more risk because richer option premiums mean more reward for selling them.

Two standard deviations. He’s trained himself to picture every trade within a two-standard-deviation band — roughly the range that covers about 95% of likely outcomes. Beyond that, he says, things get “pretty hard to quantify,” so that’s the edge of his mental world. Handily, US brokers set capital requirements to cover a two-standard-deviation move, so the margin number itself tells him where that boundary sits.

Contrarian, not trend follower

The host, a self-described trend follower, asks for an honest verdict on trend following. Sosnoff doesn’t soften it:

“I am not a trend follower. If anything, I fade trends… I don’t believe that there’s such a thing as a trend.”

To him a “trend” is just “a random case of multiple days in a row.” He’s skeptical of the whole managed-futures (CTA) industry, arguing the handful of winners are “exactly what’s supposed to happen statistically” — i.e. survivorship, not skill. His diversification comes not from many strategies but from uncorrelated products: gold, silver, bonds that don’t move with stocks.

Brains over bots

On automation — APIs, no-code trading bots, the stuff exciting Indian retail traders — he’s a skeptic. His logic is competitive, not nostalgic: your rented $200-a-month bot is racing against Citadel and Jane Street, whose billion-dollar systems “run in six milliseconds” with better data and pricing.

“If it doesn’t work for them, I don’t know how it could possibly work for a retail investor.”

At Tasty, API order flow was under 2% of total — a growing user count, tiny share of volume. AI, on the other hand, he likes — not as a trader replacement but as a tutor that accelerates learning and helps quantify risk you wouldn’t have spotted.

On India: regulators, Jane Street, and zero-DTE

This is the section Indian listeners will replay. Asked about SEBI’s case against Jane Street for alleged market manipulation, Sosnoff is blunt: the Indian regulatory and exchange technology is “a little bit antiquated” and “very protective.” His read is that Jane Street probably “didn’t do anything that would be even remotely considered illegal in the US” — they just exploited weak exchange technology and routing rules, “like shooting fish in a barrel.” His surprising conclusion: the episode is a blessing that will force India to modernize, attract real high-frequency market makers, and end up with fairer, deeper, more efficient markets.

He’s equally dismissive of SEBI sunsetting weekly options expiries to curb retail speculation:

“The last thing you want is the regulators quote trying to protect you… they make things worse not better.”

On zero-DTE options (contracts that expire the same day), the US fear was systemic risk. His verdict: it never materialized, markets are at record highs, and the flood of volume is precisely what keeps the US the deepest liquidity pool in the world.

The philosophy: decide fast, never “think about it”

A recurring theme is decisiveness. The phrase that drives him “out of my mind” is “I need to think about it.”

“You already know what you’re going to do. Everybody already knows what they’re going to do. Nobody that’s ever a successful decision maker says, ‘Let me think about it.’”

He frames active trading less as a money-making scheme than as a decision-making gym — the value is learning to process probability and act fast, a skill that transfers to all of life. (He’s careful to say he has nothing against passive investing; he just calls it “brain dead activity” you learn nothing from.)

He also defends university degrees hard — the maturation, networking, and foundation matter more than ever in the ChatGPT era, because raw information “does not make it actionable.” And his next act, Lost Dog, aims to use option-style math to attack worker pay inequity through education — going after overpaid CEOs and corporations that “don’t treat their workers fairly.”

Key Takeaways

Floor traders weren’t forecasters. Pit traders profited by taking the other side of customer orders for a small theoretical edge, regardless of their own market view. Only ~5-6% survived; the rest traded too big or traded on opinion.
More screen time does not create predictive edge. Sosnoff watched four decades of S&P ticks and insists it gives him zero advantage in knowing what happens next — only the comfort of having seen most scenarios before.
Selling premium = being the insurance company. You collect cash upfront and usually keep it; you occasionally pay a large claim when the market moves far against you. Profit comes from doing it repeatedly, not from being right on any single trade.
Law of large numbers is the engine. Many small, probability-favorable trades let statistics override luck. Hence “trade small, trade often.”
Position size is the only real defense against tail events. Outlier crashes (Aug 2024, Apr 2025) are unavoidable; keeping each position small is, in his view, the only durable protection.
Implied volatility is the strategy selector. Cheap IV and expensive IV call for different structures; he takes on more risk when IV is rich because premiums (his income) are fatter.
He trades inside a two-standard-deviation band — the ~95% range of outcomes. US broker margin requirements are set to cover a two-SD move, so the buying-power reduction number itself reveals that boundary.
He fades trends and doubts trend following exists — to him a trend is “a random case of multiple days in a row,” and CTA winners are statistical survivors, not skilled.
Diversification comes from uncorrelated products, not multiple strategies — gold, silver, bonds that don’t track equities.
Open interest carries no informational edge for liquid underlyings (indexes, large caps); he only glances at it for illiquid names to gauge participation.
He never looks at charts — only raw price, which he memorizes. Forty-plus years without a chart.
Retail trading bots are structurally disadvantaged — they compete against Citadel/Jane Street systems running in ~6 milliseconds. Even elite prop firms find automating opening trades nearly impossible. API order flow at Tasty was under 2% of volume.
AI helps as a learning and risk-quantifying tool, not as a trader replacement — “brains over bots.”
Decisiveness over deliberation — “I need to think about it” is, to him, a red flag; good decision-makers already know.
Tasty’s moat was creative, credible, contrarian content — commissions and tech are commoditized; differentiated content (made by an actual trader) is not.

Claude’s Take

This is a good interview let down slightly by a soft interviewer. The host is knowledgeable and clearly a genuine fan, but he repeatedly fishes for the “you must have a secret intuition” answer and Sosnoff keeps swatting it away — which is actually the most valuable thing in the conversation. A 43-year veteran telling you he can’t predict the market is worth more than a hundred gurus claiming they can.

The honest stuff is genuinely honest: position sizing as the only tail-risk defense, edge living in mechanics and repetition rather than foresight, the brutal floor-trading survival rate. These aren’t sales pitches; they’re hard-won and slightly deflating, which is the tell that they’re real.

Where to keep a hand on your wallet: Sosnoff sells an active-trading worldview, and “everybody should try active trading” is conveniently aligned with having built two brokerages. The claim that passive investing is “brain dead” and teaches you nothing is rhetorical overreach — for most people, the boring index fund quietly beats their active trading, a fact the data is fairly settled on. His framing of trading as a “decision-making gym” is appealing but also the kind of thing that keeps people churning trades (and paying spreads). Take the philosophy, be wary of the implied permission slip.

The India section is the most quotable and the most one-sided. His “regulators make it worse” line is a libertarian reflex; the Jane Street “shooting fish in a barrel” read is plausible but he openly admits he doesn’t trade Indian markets and is guessing. Still, his structural point — that weak exchange tech, not villainy, was the real story, and that the episode will force overdue modernization — is a sharper frame than most of the moralizing coverage.

Seven out of ten: substantive, refreshingly candid, occasionally self-serving, and worth it mainly for the rare sight of an expert insisting on the limits of his own expertise.

Tom Sosnoff: Inside the Mind of a Trading Legend | In The Money by Zerodha Podcast 01 | Full Episode