Losers Average Losers

Definition & Origins

"Losers average losers" is four words long, and Paul Tudor Jones has spent a career making sure no one on his desk forgets them. The rule forbids the most natural gesture in trading: when a position moves against you, buy more at the lower price, improve your average cost, and wait to be proven right. The gesture feels like prudence — the average cost drops, the breakeven point comes closer, the loss on the screen shrinks as a percentage. But the arithmetic is a sedative. What actually happened is that exposure doubled on a thesis the market is currently voting against, at the exact moment conviction is least trustworthy. PTJ's verdict is that this gesture is not strategy. It is ego wearing the costume of strategy — a refusal to accept the information the market has just delivered, compounded with fresh capital. Only losers do it. Winners cut, reassess, and re-enter on their own terms.

The doctrine's roots are in New Orleans, in 1978, on the desk of Eli Tullis, one of the great cotton traders of the era. PTJ arrived fresh out of college with his mentality, by his own account, "still firmly set on fraternity row," and was fired on a June Friday after falling asleep at the phone he was paid to man. The humiliation was total — he never told his parents — and it installed in him the first premise of everything that followed: this business will expel you without warning, and the only durable asset is the discipline you bring back the next day. He secured a job on the floor of the New York Cotton Exchange and rebuilt from zero.

The second lesson from the Tullis apprenticeship was about what being wrong looks like when a professional absorbs it. PTJ watched Tullis get caught limit-down on a huge long cotton position after a weekend of rain broke a spectacular drought — smashed, by any measure. Tullis came out of his office at lunch smiling, charming his wife's friends, already trading the comeback. The image never left PTJ: the mark of the master is not avoiding the limit-down day but refusing to let it metastasize into hesitation, denial, or revenge. Wear the loss, keep the confidence, come back.

The final origin is the one PTJ has cited ever since as the most important lesson of his career: the devastating cotton loss of 1979, the trade that nearly ended him before he began. Recounted in his Market Wizards interview with Jack Schwager, it converted the scattered instincts of the apprenticeship into a written creed — a set of mechanical rules, quoted below, that he has restated in essentially the same form for nearly four decades. At the head of that creed sits the shortest rule of all. The interview was conducted in 1987, in the afterglow of the crash he had just traded correctly, and published in 1989; the four words escaped the book and became trading folklore, quoted on desks that could not name another line from the chapter. The phrasing is deliberately brutal. It is not "averaging down is suboptimal." It is an identity verdict: the act is what losers do.

Core Ideas

The first core idea is that a falling price is information, and averaging down is a refusal to receive it. When the market moves against a position, it is voting against the thesis with the aggregated judgment of everyone better informed. Adding at the lower price does not argue with that verdict on the merits; it simply raises the stakes on being right, precisely when the evidence says you are wrong. The trader who averages losers has stopped trading the market and started trading his own need to be right — and the market is indifferent to that need. This is why PTJ pairs the rule with the demand to always question yourself and your ability: the moment certainty hardens into identity, the position owns the trader.

There is also cold arithmetic under the rhetoric. A loss of ten percent needs an eleven percent gain to recover; a loss of fifty percent needs a double. Every increment added to a losing position deepens the hole that future gains must climb out of, and a loser averaged repeatedly stops being a bad trade and becomes an existential one. Accounts are rarely destroyed by the first wrong decision. They are destroyed by the third and fourth additions to it — the sequence in which each new purchase feels smaller than the last while the total exposure quietly becomes the whole book. The rule cuts that sequence at the first link.

The second idea is the rule's inverse, which is where its productive power lives. Because adding to losers is forbidden, all pyramiding happens on winning positions, where the market is confirming the thesis. Position size grows with confirmation and shrinks with contradiction; trading volume contracts when you are trading poorly and expands when you are trading well. This asymmetry — press what works, starve what doesn't — is the operational core of the PTJ method, and it connects directly to the 5:1 risk/reward filter: entries are only taken where the payoff dwarfs the risk, and size is only added where price has ratified the idea. The two rules together mean a trader can be wrong most of the time and still win, because the losses are kept small by construction and the winners are allowed to grow large by design. Hit rate stops being the business; payoff distribution becomes the business.

The third idea is that the rule exists to make the decision before emotion arrives. Nobody can calmly evaluate whether to average down at the moment it feels most tempting; the temptation is the emotion. So the question is deleted in advance. "Losers average losers" is not advice to be weighed situationally — it is a mechanical prohibition, the desk-level enforcement mechanism of defense first. Defense first is the philosophy, the posture of always thinking about losing before making; this rule is that philosophy rendered as an instruction a trader can follow on his worst day, when instruction is the only thing he can follow. Its siblings are equally mechanical: don't be a hero, don't have an ego, never play macho man with the market. The enemy is always the same — the self that would rather be right than solvent.

Practical Application

On the Tudor desk, the rule is embedded in a daily routine of pre-commitment. Each morning begins from the assumption that every open position is wrong. Stop-risk points are defined before the session, maximum possible drawdown is computed from them, and the exit plan exists before the loss does. Averaging down is structurally impossible inside this routine, because the response to an adverse move has already been decided when the trader was calm. The practical question is never "should I add here?" — it is "is my stop hit yet?" The same standard governs the firm, not just the book: PTJ's flat statement that anyone who has truly succeeded at trading or investing is first and foremost a great risk manager is, inside Tudor, a hiring and promotion criterion, not a slogan.

Volume functions as a thermostat. When trading poorly, cut size; when trading well, press it. The same logic extends to situations where the trader has no control: PTJ refuses to risk significant money in front of key reports, on the grounds that holding through a binary data release is gambling, not trading. A cousin rule, the time stop, applies even when no money is being lost — if a market fails to do what the thesis requires within the expected window, the position goes. All of these are the same refusal in different clothing: never let a position negotiate with you.

Re-entry is the part outsiders miss. Cutting a loser does not mean abandoning the idea; it means repurchasing the right to have the idea later, at a moment and price of the trader's choosing rather than the position's. The discipline is cut, reassess, re-enter — not cut, sulk, forget. This is what separates the rule from mere loss aversion: the exit is not a verdict on the thesis, only on the timing, and the trader who obeyed the rule retains both the capital and the psychological freedom to take the trade again when the setup returns.

The system's proof is October 1987. The analog model PTJ built with Peter Borish had mapped the year onto 1929 and called for a violent autumn break — but the model only produced the hypothesis. What turned the hypothesis into the trade of the decade was the execution apparatus around it: defined risk, mechanical exits, no averaging, no negotiation. A trader who averages losers cannot hold a short into Black Monday; the first sharp rally would have doubled him against his will. The crash windfall and the rule are the same discipline photographed from opposite sides.

The rule also scaled beyond the trading book. At the Robin Hood Foundation, PTJ built grant-making on the same cutting logic — fund programs like positions, define the metric of success before writing the check, and reallocate away from what does not work. Venture philanthropy is "losers average losers" applied to the grant book: no program is averaged down with more money because the foundation believes in it. The pattern holds across four decades and two institutions — what fails gets starved, what works gets fed, and sentiment never overrides the measurement.

Common Misconceptions

The first misconception is that the rule expresses timidity — that a man this focused on cutting losers must trade small and scared. The record is the opposite: the 1987 crash short, the bitcoin call in 2020, the aggressive macro strikes PTJ describes as knockouts landed after long stretches of patient jabbing. His own metaphor for the trader's day is a classic boxing match — pair, jab, feel the opponent out, and take the big shot only when the opening finally comes. The rule is what makes the aggression affordable. Because exits are mechanical and losses are capped by construction, entries can be violent. Cutting losers instantly and betting big are not in tension; they are one system.

The second misconception is that the rule forbids returning to a failed idea. It forbids only one specific act: adding to a position that is currently losing, in defense of the original entry. Re-entering later, on new evidence, at a new price, is not averaging down — it is a new trade, with its own risk budget and its own stop. Traders who conflate the two either hold losers to avoid "giving up," or abandon good theses forever after one stop-out. Both are the error the rule was written to prevent.

The third misconception is that the rule transfers trivially to every investor. PTJ draws the contrast himself with the buy-and-hold tradition: a value investor with a hundred-year horizon can sit through a fifty percent drawdown and let time and compounding do the work. A leveraged trader cannot — for him, a drawdown averaged into is a solvency event, not a patience test. The rule is calibrated to the trader's game, where survival is the strategy and returns are its by-product. The transferable part is not the tactic but the honesty about which game you are playing.

Key Sources / Related Concepts

Primary sources: Market Wizards interview (1987/1989), Invest Like the Best (2026), Trader: The Documentary (1987), Buckley Commencement Address (2009).

Related concepts: Defense First (the philosophy this rule enforces), 5:1 Risk/Reward Ratio (the entry filter that pairs with the exit rule), The 200-Day Moving Average Rule (the trend tripwire), Venture Philanthropy (the same cutting logic applied to the grant book).