Tilt: What One Trade Costs

One day, two decisions

The arithmetic of a bad day splits cleanly, and the split is the lesson. A morning of four losses at a 1% risk unit costs about 4% of the account on a $50,000 balance, which is a bad day the plan already anticipated — at a 45% win rate, four consecutive losses happen, and the size was chosen so that they are survivable. That morning is the plan working. What makes it expensive is the trade that follows, and in the worked case a single position sized from the mood rather than the plan doubles the risk unit to $1,000 and loses 1.5R, costing $1,500 on one trade. The day then closes at $3,500, about 7% of capital, from a plan whose budget allowed 1% a trade. The last conversion is the one that changes behaviour, because it turns the day into a project. At the system’s own 0.2R expectancy and a restored 1% unit, each trade returns about $100 on average, so recovering $3,500 takes roughly thirty-five trades — six weeks of good work. The trader has to take those thirty-five trades while remembering the day, and the memory is of the kind that makes losses feel larger than they are, which is the state in which the next tilted trade is taken. That is why tilt is not a discipline problem in the ordinary sense: it does not just spend money, it converts a variance day into a recovery attempt, and it hands the attempt to the person least equipped to run it. There is a second cost that does not appear in any of those figures. The afternoon trade was sized from the mood, so it had no written setup, which means the day produced no information about the method — only a result. A planned loss is data: it counts toward the expectancy the size was computed from. An unplanned loss is noise with a price tag, and it is the reason a trading record can show a positive expectancy and a losing account at the same time. One day, two decisions — Morning: four losses at 1% risk (1R = $500): −4R = −$2,000 — within the budget ← · Afternoon: risk doubled to 2% (1R = $1,000): −1.5R = −$1,500 on a single trade · Total for the day: −$3,500, about 7% of the account ← · Recovery at 0.2R and a restored 1% unit: About 35 trades — six weeks of good work The morning is what a system costs; the afternoon is what a state costs. Both are measurable, and only one of them can be bounded by a rule written in advance.

The structural response

The response to tilt is a small set of rules that remove the decision rather than win it, and three of them do most of the work. A daily loss limit — a stated percentage at which the session ends, which is a number and not a feeling — bounds the worst day the way a stop bounds the worst trade. A rule that size never changes after a loss removes the single most expensive habit in the pattern, because doubling a unit after a loss is how a 4% day becomes a 7% day. And an unplanned-trade prohibition, meaning that an order without a written setup, entry, stop and size is not taken at all, converts the afternoon trade into a non-event. Two supporting practices make those rules survivable rather than merely strict. The first is that the streak has to be understood as scheduled: a system at a 45% win rate will produce runs of losses, the size was chosen so that they are an inconvenience rather than a decision, and the recognition that the run is the forecast is what keeps the limit from feeling like a punishment. The second is a record that separates planned from impulse trades, because the only way to know whether the limit is working is to have the days and the deviations in a form that can be counted rather than remembered. The honest limit of the whole approach is that it trades profit for survival, and it should be evaluated in those terms. A hard daily limit will occasionally end a session that would have recovered, and a size rule forbids the position that feels most deserved after a run of wins. What the rules buy is the ability to keep taking the trades when the method is working, which is the only thing that compounds — and the cost of not having them is not a bad day but a recovery project, which is the one activity in trading with no positive expectancy at all. • A daily loss limit: a stated percentage that ends the session. • A rule that size never changes after a loss. • No order without a written setup, entry, stop and size. • Read the streak as scheduled, because the size was chosen for it. • Keep the planned and impulse trades separable in the record. The most expensive version of tilt is not the angry trade; it is the reasonable-sounding one. Size rises because the setup “looks obvious”, the stop widens because the level is “just noise”, and both changes are made by someone who has already lost more than the day allowed.

The body gets there first

Tilt is usually described as an emotion, and by the time it has a name it has already changed what you can do. The measurable part of it is physiological: a loss that matters raises arousal, and arousal narrows attention onto the threat, shortens the time horizon of decisions, and increases the appeal of action over inaction. Sleeping badly, skipping meals, being ill, or trading after a couple of drinks each shift the same dial in the same direction. This is why the tell for tilt is often physical before it is cognitive — a hot face, a tight chest, a hand moving to the keyboard faster than the thought arrives. The important consequence is that willpower is the wrong tool at that moment. A rule that has to be enforced by a decision made while aroused is not really a rule. What works is a rule that removes the decision: a daily loss limit that disables the platform for the rest of the day, a position-size cap that requires a written entry to override, a hard stop on the number of trades. The rule needs to be enforced by the environment, not by resolve, because the state that makes it necessary also makes resolve unreliable. There is a second loss worth naming because it is the expensive one. The first loss of the day is usually within the plan. The trades that do the damage are the ones taken to recover it, and those are systematically larger, faster and less planned than anything the strategy prescribed — the size goes up as the day goes wrong. That is why the arithmetic in this lesson ends with “trades needed to recover”, which grows non-linearly as size falls, and why the honest response to a limit breach is not a smaller version of the same bet but the end of the session. • Arousal arrives before the awareness of it, and it narrows attention and shortens the horizon. • Sleep, food, illness and alcohol move the same dial as a loss does. • Make the limit environmental — the rule must not require a decision made while tilted. • The expensive trade is the recovery trade, which is larger and less planned than anything the strategy prescribed.

Setting the limit, and ending the day

A daily loss limit is useful only if the number is derived rather than picked, and it has two inputs. The first is the strategy’s own worst *normal* day: at a given risk unit and hit rate, the number of consecutive losses that falls inside ordinary variation — the same calculation that made the morning’s four losses expected rather than alarming. The second is the account’s tolerance for a run of such days. A limit set at the worst normal day will end sessions that would have been fine, and a limit set far beyond it permits exactly the day this lesson priced. The workable band is a limit a normal bad day rarely reaches, and that a sequence of three or four cannot breach without the monthly budget being reconsidered. The second design choice matters as much as the number: the limit has to be enforced by the environment rather than by the trader. A stated figure that requires a decision to apply is not a limit, it is an intention, and the state that makes it necessary is the state in which it is least reliable. Practical versions remove the possibility instead of relying on the will: a platform setting that disables new orders for the day, a separate account with a fixed balance, an agreed hand-off to a colleague, or simply a rule that the terminal is closed and the day reviewed in writing. Whichever is chosen, the test is the same — could the rule survive being enforced by the tired, aroused version of you who has just breached it? Then the breach protocol, because limits will be hit and the response to a breach is where most of the damage actually occurs. Three steps, all written in advance. Stop: the session is over and the position is not replaced. Cool off: a fixed interval, at least overnight, before any decision about the next session is taken, because a review conducted five minutes after the loss is conducted by the tilted person. Review on paper: separate the planned losses from the impulse trades, and check which of the three rules — the limit, the size rule, the written-setup rule — was actually the one that failed. The answer is often that the limit was fine and the size rule was ignored earlier, which is a different fix. The final piece is the restart condition. A breach should carry a stated, checkable requirement before size returns: a written review of the losing trades, the plan restated, and a smaller unit for a defined number of sessions until the record shows the rules being followed. Whether that is a formality depends entirely on whether it has an observable trigger — the number of sessions, the size, the review completed — rather than a resolution to be more careful. The same claim this lesson makes about the recovery trade holds for the restart: a rule that depends on resolve will be tested on the day resolve is lowest. • Derive the limit from the strategy’s worst normal day and the account’s tolerance for a run of them. • Make it environmental, not wilful: a setting, a separate account, or a closed terminal rather than a resolution. • On a breach: stop, cool off for a fixed interval, then review on paper — never review in the moment. • Identify which rule failed, because it is often the size rule rather than the limit. • Give the restart a checkable condition: a completed review, a reduced unit, a stated number of sessions. A limit that is negotiated during the session is not a limit. The most common failure is not the trader who has no rule but the trader who has one, restates it as “just this once”, and thereby converts the rule into a judgement call — which is exactly the tool tilt needs to operate.

Sources

• Tilt, arousal and decision qualitySteenbarger, “The Daily Trading Coach”
• Loss chasing and escalation after a drawdownBehavioural finance literature on loss chasing
• Daily loss limits and trading policyElder, “Come Into My Trading Room”