Every decision you make under market pressure is not equally consequential. Some can be undone cheaply if you turn out to be wrong. Others cannot — and the moment you step through them, the cost of being wrong multiplies. This article teaches a prior classification step: before asking how likely is this to work, ask how reversible is this if it doesn't. Once you can answer that question reliably, you can match the deliberation you spend to the reversibility of what you're deciding — not to how urgent it feels.

Two-Way Doors and One-Way Doors

In his 2015 Amazon shareholder letter (published April 2016), Jeff Bezos drew a distinction that maps directly onto trading decisions. He called consequential, irreversible choices Type 1 decisions — "one-way doors." Once you step through a one-way door, getting back is difficult or impossible at an acceptable cost. He called changeable, reversible choices Type 2 decisions — "two-way doors" — which can and should be made quickly, because you can always walk back through. The failure mode he described is treating a two-way door like a one-way door: moving too slowly, over-deliberating on something that barely matters if it goes wrong. The other failure mode — the one more dangerous in markets — is treating a one-way door like a two-way door: moving too quickly on something you cannot easily exit.

Applied to a trading session, the classification looks like this. Deciding which watch list to review this morning is a two-way door. You can change it in thirty seconds. Deciding to hold a concentrated, illiquid position through a scheduled risk event is a one-way door: if the event goes against you, your ability to exit at a reasonable cost shrinks dramatically at the precise moment you want to leave. The size of the potential move and the depth of the exit are not the same thing, and it is the exit depth that determines reversibility.

The Deliberation Budget

Attention and deliberation are finite. Most traders spend them approximately equally across decisions — roughly proportional to how emotionally charged the decision feels, or how large the dollar amount looks on screen. That is the wrong allocation.

A two-way door deserves a quick check: is there an obvious reason not to proceed? If not, proceed. Spending ten minutes deliberating over a small, liquid position you can exit in seconds is waste. Spending two minutes on a large, illiquid commitment to a volatile event is negligence dressed as decisiveness.

The disciplined allocation reverses the intuitive one. Recognize a two-way door, decide quickly, and preserve deliberation for the doors that cost more to exit than they cost to enter. That gap — entry cost versus exit cost under adverse conditions — is the operational definition of one-way.

Converting One-Way Doors Into Two-Way Doors

The most practically useful insight in the framework is that one-way doors are often structural, not fixed. You can redesign the decision to restore reversibility. Three specific mechanics do this.

Staged entry. Rather than committing full size at once, commit a fraction first. A 25% initial position in a less-liquid name or around a binary event gives you real skin in the game while preserving 75% of your decision. If the setup develops as expected, you add. If conditions deteriorate before you've added, you exit the small starter at a proportionally small cost. The decision stays two-way because the full commitment hasn't been made yet.

Smaller initial size. Distinct from staging: simply committing less than you intend to as a deliberate trial. Trial commitments acknowledge that your thesis might be correct but your timing uncertain. A position small enough that the maximum loss is genuinely acceptable converts a psychologically irreversible feeling into a practically reversible one. The goal is to keep the exit option emotionally available, not just theoretically available.

Explicit exit criteria set before entry. A decision becomes functionally one-way when you haven't defined what would make you leave before you enter — because by the time conditions deteriorate, the emotional cost of admitting the thesis is wrong makes exit feel like a loss, not a routine process step. Pre-defining the exit converts it from a judgment call under pressure into a rule you execute mechanically. This overlaps with pre-commitment (covered in Pre-Commitment and If-Then Rules), but the classification step here is prior: you are identifying that a decision warrants this treatment, not yet writing the rule.

A Historical Example: When Liquidity Assumptions Fail

Long-Term Capital Management nearly collapsed in 1998 and required a private recapitalization of approximately $3.6 billion, coordinated by the New York Federal Reserve and funded by a consortium of fourteen banks. One element worth examining from a classification standpoint: positions that looked, on entry, like two-way doors — large enough markets, seemingly ample exit options — became functionally one-way during the crisis. When correlations broke down and multiple markets moved against the fund simultaneously, the available exit prices degraded so severely that selling at a reasonable cost ceased to be an option.

This illustrates how liquidity assumptions can turn a two-way door into a one-way one. The reversibility of a decision is not fixed at entry; it is partly a function of market conditions at the moment you want to exit. Decisions made assuming normal liquidity carry a hidden one-way component that only becomes visible under stress. The classification step should therefore ask not just "can I exit this?" but "can I exit this under the conditions that would make me want to exit?"

One-Way Blindness: The Failure Mode

One-way blindness is the consistent tendency to classify decisions as two-way doors when they are not. It produces a recognizable pattern: confident, quick entries into positions that later turn out to be difficult or very costly to exit. The behavioral driver is usually one of three things.

First, liquidity overestimation. Normal-market bid-ask spreads feel like the real exit cost, but they understate exit cost when you are holding a large position relative to average volume, or when the market conditions that would make you want to exit are also the conditions that would widen spreads or reduce available buyers.

Second, concentration under confirmation. A trader who has done extensive research on a thesis is psychologically primed to see that thesis as reversible ("I'm right, so I can always get out near my entry") even as their position size makes exit expensive. Conviction about the direction of a trade is not the same as reversibility of the decision.

Third, urgency masquerading as speed. A decision that feels time-pressured — the price is moving, the setup is forming right now — gets classified as needing immediate full commitment. But urgency is a feeling, not a fact about reversibility. In many cases, the right response to time pressure is to confirm that you are looking at a two-way door before proceeding quickly. If the answer is no, the urgency is a signal to be more deliberate, not less.

When these three combine — thin actual liquidity, high conviction, felt time pressure — one-way blindness is at its most dangerous. The result is a class of decisions that are made as if they were cheap to undo and turn out to be expensive to undo at the worst moments. This is a separable cost from the decision being right or wrong — you can be right directionally and still lose on a one-way door by being forced to exit at a cost you didn't price in at entry. The concept is covered more fully in the context of pre-entry analysis in Running a Pre-Mortem Before a Trade.

The Discipline: Classify Before You Deliberate

The process fix is a short classification step before any significant decision. It does not need to be elaborate. Two questions are sufficient.

How expensive is the exit under adverse conditions? Not how expensive is the exit now, when you are calm and the market is normal — how expensive is it if the position moves against you by a meaningful amount in a less liquid environment. If the honest answer is "significantly more expensive than entry," you are looking at a one-way component.

What is the cost of being wrong? Not just the mark-to-market loss, but the cost in terms of reduced future optionality — cash tied up, emotional capital consumed, time spent managing a deteriorating position instead of identifying new ones. One-way doors tend to consume optionality in ways that are not fully captured by the dollar loss.

If both answers point toward low reversibility, the decision warrants staged entry, smaller initial size, or explicit pre-defined exit criteria — and it warrants more deliberation time before the first dollar is committed. If both point toward high reversibility, proceed quickly and preserve your deliberation budget for decisions that actually need it. The matching of deliberation to reversibility, not to emotional intensity, is the behavioral skill this framework is developing.

Speed Run Drill: The Door Check

In Abu Terminal's Speed Run, before you confirm any entry, Abu surfaces a brief door check prompt: classify this decision as two-way (you can exit this cheaply if wrong) or one-way (exit is costly or slow under adverse conditions), based on the stated liquidity context of the event. Then choose a deliberation level: quick, standard, or slow-with-pre-mortem.

The drill flags any one-way classification taken at "quick." That flag is not a block — it is feedback. After the run, review every flagged moment and ask: did you know it was a one-way door, or did you classify it without stopping to check? Did the flag change how you felt about the decision in retrospect?

The purpose is not to always choose "slow" — that would be its own failure mode, applying maximum deliberation to every two-way decision and paralyzing your process. The purpose is to feel the difference between quick decisions that were appropriate given high reversibility, and quick decisions that were quick because the situation felt urgent, not because the door was two-way. That distinction — reversibility versus urgency — is the behavioral skill this drill is building. It will not be stabilizing after one session; it develops across many runs as the classification becomes faster and more accurate.

Reversible vs Irreversible Decisions: Spend Your Caution Budget on the One-Way Doors