Paul Tudor Jones: the loss that came before the legend
He is remembered as the man who was positioned short going into the worst day in stock market history. He would rather you remembered a cotton trade eight years earlier that took most of the account he was managing — because that is the trade his rules came from. The crash call is the famous part. The 1979 loss is the useful part.
KEY TAKEAWAYS
- The 1979 cotton trade came first. By his own account it cost 60–70% of the account he was running — because he had worked out what he could make, not what he could lose.
- Tudor's 1987 result is widely reported as +125.9% for the full year, after fees — not the "doubled in a day" version that circulates online.
- His stated edge is not prediction. It is defining the exit before the entry, so the worst case is known in advance.
- A 60% drawdown needs a 150% gain to undo. That asymmetry — not market timing — is why he calls trading a defensive game.
Who is Paul Tudor Jones?
Paul Tudor Jones II was born in Memphis, Tennessee in September 1954 and graduated in economics from the University of Virginia in 1976. He entered the business the unglamorous way: as a clerk on the floor of the New York Cotton Exchange, mentored by the cotton trader Eli Tullis. He worked as a commodities broker at E. F. Hutton, and in 1980, at 26, founded Tudor Investment Corporation.
He later served as chairman of the New York Cotton Exchange, and in 1988 co-founded the Robin Hood Foundation, an anti-poverty organisation in New York. In 1994 Tudor paid an $800,000 settlement to the SEC over alleged uptick-rule violations, neither admitting nor denying wrongdoing — a detail the admiring retellings tend to leave out, and one we include because this section prints the record.
None of that is why traders read him. They read him because of one week in October 1987 — and because of what he says about how he got there.
What happened in the 1979 cotton trade?
Before Tudor existed, Jones was trading cotton futures and took a large speculative position that moved against him. In interviews he has described losing roughly 60–70% of the account he was managing on that single trade, and being close enough to quitting the business that the loss became the dividing line in his career.
The diagnosis he gives is the part worth copying. He had reasoned forward from the price he expected and the number of contracts that would make: he had calculated the upside. He had not defined the point at which he would be wrong, or how much of the account that point would cost. There was no plan for being wrong, so being wrong had no floor.
That is not a nineteen-seventies problem. It is the single most common structure of a blown crypto account: a position sized by the profit it would produce if it worked.
| The loss | The gain that undoes it | What that means in practice |
|---|---|---|
| Down 25% | +33% needed | Recoverable inside one good run. |
| Down 50% | +100% needed | You now have to double what is left. |
| Down 60% | +150% needed | The band the 1979 cotton trade landed in. |
| Down 70% | +233% needed | Most accounts never come back from here. |
Read the bottom bar again. Down 70%, and you need to more than triple what is left just to be even. This is why he describes trading as a defensive game — not because defence is virtuous, but because the arithmetic of recovery is brutally one-sided. You can check the same maths against your own numbers in the drawdown recovery calculator.
How did he actually position for Black Monday?
On Monday 19 October 1987, the Dow Jones Industrial Average fell 508 points — about 22.6% in a single session, still the largest one-day percentage decline in its history. Jones and his research partner Peter Borish had been studying the market's structure against the period preceding the 1929 crash, and Tudor went into that week positioned short.
What happened next is where legend and record separate.
| Claim | The popular retelling | The documented record |
|---|---|---|
| The 1987 result | "Doubled his money in a day" | +125.9% for the year |
| The skill on show | "He called the exact date" | Short before the crash |
| How he got there | "One perfect call made him" | A 1979 loss came first |
| What he credits | "Reading the market" | His exit, defined first |
Tudor's 1987 return is widely reported at +125.9% for the full year, after fees. That is an extraordinary year. It is not the same claim as doubling money in a single session, which is the version that spreads on social media, and the difference matters: one describes a position held through a market regime, the other implies a single perfectly-timed click. Reporting it accurately is not pedantry — the inflated version teaches beginners to look for the click.
The 1987 documentary Trader, filmed in the run-up to the crash, shows a handwritten note taped near his desk. It reads: "Losers average losers." The most famous crash call in modern trading was made by someone whose desk decoration was a rule about not adding to losing positions.
What does he actually say about risk?
Risk comes first and prediction second: know where you exit before you enter, never add to a losing position, and cut size when you are trading badly — so that the maximum drawdown is a number known in advance. The quotes below come from his interview in Jack D. Schwager's Market Wizards (1989), the primary source for almost everything attributed to him.
"Don't focus on making money; focus on protecting what you have."Paul Tudor Jones — interviewed in Market Wizards, Jack D. Schwager (1989).
Said by the man having one of the best years in hedge fund history at the time. Defence is not what you do instead of offence — it is what makes a long offence possible.
"I'm always thinking about losing money as opposed to making money."Paul Tudor Jones — Market Wizards (1989).
"The most important rule of trading is to play great defense, not great offense. I know where my stop risk points are going to be. I do that so I can define my maximum possible drawdown."Paul Tudor Jones — Market Wizards (1989).
The second sentence is the mechanism, and it is the one people skip. The stop is not there to be cautious — it is there so the worst case is a number he knows in advance rather than a discovery.
"Don't be a hero. Don't have an ego. Always question yourself and your ability."Paul Tudor Jones — Market Wizards (1989).
What does this mean on a $2,000 crypto account?
On a $2,000 account the 1% risk rule means accepting a loss of about $20 on any single trade — and one position entered without a defined exit that costs 60% of the account ($1,200) does the damage of 91 consecutive disciplined losses. Jones's numbers are unreachable. His arithmetic is not.
Now suppose one position is entered without a defined exit and ends up costing 60% of the account, the band his 1979 trade landed in. That is $1,200.
How many disciplined losses would it take to lose the same amount? At a flat $20 per trade, sixty of them in a row. Sized properly as 1% of a shrinking balance, the answer is 91 consecutive losing trades — because each loss is 1% of a smaller number. Ninety-one losses in a row is not a realistic run for any strategy with an edge. One undefined position gets there in an afternoon.
| How the $1,200 is lost on a $2,000 account | Loss per trade | Trades needed to lose $1,200 |
|---|---|---|
| One position with no defined exit (the 1979 band, 60%) | $1,200 | 1 |
| Disciplined losses at a flat $20 each | $20 | 60 |
| Disciplined losses at 1% of a shrinking balance | 1% of what is left | 91 in a row |
That is the whole argument for position sizing in one comparison. Risk rules do not exist to make small losses. They exist to make the catastrophic loss structurally unreachable.
PRACTICE CORNER
Jones's rule is that the exit exists before the entry. Every major exchange lets you attach a stop when you open the position, on the same screen — so the worst case is set while you are still calm rather than while you are watching it happen.
Referral links — they never change our assessment. Education only; most retail traders lose money.
Where does this advice stop working?
In thin markets, on very small accounts, and whenever the aggression is copied without the infrastructure behind it. A stop is not a guarantee of price. Jones traded liquid futures markets. In thin crypto markets a stop is an instruction to exit, not a promise of the level — in a fast move it fills lower, sometimes much lower. Read slippage and liquidity and spread before assuming a stop caps your loss precisely.
"Cut size when trading badly" assumes you have a size to cut. On a small account, halving an already-minimum position may put you below the exchange minimum. The equivalent action is to stop trading for a defined period, not to shrink toward zero.
His macro method needs conditions you probably do not have. A research partner, decades of tape, and capital that survives being early. Copying the aggression without the infrastructure is the single most common way traders misread this profile.
And the record is one man's. Tudor's historical returns are the results of a specific fund in specific decades. They are not a forecast, an expectation, or an argument that anyone reading this will do anything similar.
What are the common mistakes when learning from Jones?
Four, and all of them come from reading the 1987 outcome without the 1979 cause. Quoting the crash, skipping the cotton. The 1987 position is the outcome; the 1979 loss is the cause. A profile that only tells the first half teaches nothing you can act on. Hearing "defence" as "trade small and hope". He is not describing timidity — he is describing a known maximum loss, which is what lets you take a position at all. Treating "never average losers" as a preference. On leveraged positions, adding to a loser also drags your liquidation price toward the market while you do it. Assuming the stop was the hard part. Deciding the exit is easy. Honouring it when the position is open and moving is the entire skill.
FAQ
Did Paul Tudor Jones predict the exact date of Black Monday?
He was positioned short in advance, based on work he and Peter Borish did comparing the market's structure to the period before 1929. Being positioned ahead of a move is not the same as naming a date, and the distinction matters: one is risk placement, the other is fortune telling.
How much did Tudor actually make in 1987?
The figure widely reported is +125.9% for the full year, after fees. Claims of doubling in a single session are an inflation of that number. We use the reported annual figure and say where it comes from.
Is "losers average losers" really his?
The note is visible in the 1987 documentary Trader, taped near his desk. It is one of the better-evidenced pieces of trading folklore — unlike the "cutting losses" line, which belongs to Ed Seykota.
Which of his rules matters most for a beginner in crypto?
Define the exit before the entry, and size the position from that exit rather than from the profit you are hoping for. It is the exact reversal of the mistake he says caused his 1979 loss, and it is the one rule that makes the others possible.