Two exit types, both pre-named at initiation. (1) THESIS-COMPLETION exits — the price target is reached, the catalyst materializes, the variant perception has been priced in. Scale out, take profits, redeploy. (2) THESIS-BROKEN exits — a pre-named operational signal fails, a falsification trigger fires, a new bear case emerges that the original thesis did not price. Exit decisively at any size you would size a new position with the new evidence; usually that is zero. Both exit conditions are written down at initiation; both are evaluated mechanically when the trigger fires.
| Exit type | Trigger | Discipline |
|---|---|---|
| Thesis-completion | Price reaches the base-case target; catalyst plays out; variant perception fully priced | Scale out into strength; never let a winner round-trip to break-even because of attachment |
| Thesis-broken (operational) | A pre-named operational signal moves against you (margin contracts beyond a named threshold, churn accelerates beyond a named rate) | Investigate the explanation; exit if the trigger holds; never re-define the trigger after it fires |
| Thesis-broken (structural) | A new bear case emerges that the original thesis did not contemplate (regulatory shift, technological substitute, balance-sheet deterioration) | Reset to first principles — would you initiate this position today given the new evidence? Size to that answer. |
| Stop-loss (mechanical) | A drawdown percentage hits a pre-named level | Useful for portfolio-level discipline; weaker than thesis-broken exits because it confuses price action with thesis status |
The pre-mortem discipline. Before initiation, write down the answer to: "If this trade loses 30%, what is the most likely reason?" The pre-mortem forces you to name the bear case in concrete operational terms BEFORE you have any attachment to the position. The thesis-broken exit conditions then follow naturally — they are the named bears made into observable signals. Most exits that go wrong fail because the pre-mortem was never written, and so the analyst is improvising the exit decision under pressure with capital at stake.
Worked example — exit plan at Brimwood Lumber initiation. Thesis-completion: scale 30% of position out at $54 (base-case target), 30% more at $58, hold remaining 40% pending fresh thesis if price exceeds $58 with the original catalyst intact. Thesis-broken: exit fully if next-quarter receivables-vs-revenue gap exceeds 4 percentage points without a contracting-mix explanation; exit fully if pass-through speed slows to under 60% of historical (verified by gross-margin walk); exit fully if a regulatory shift on lumber-grading rules surfaces (this is the contemplated structural risk). Stop-loss: portfolio-level rule, exit at 18% drawdown regardless of thesis status — that is a Brimwood-specific risk-budget call, not a thesis-status call.
## See also: deeper references - **The memo discipline and the explicit exit section:** `memo-3` and `memo-4` in `client-practice-201` — for the memo template that institutionalizes the pre-written exit plan. - **Anchoring on cost basis and the disposition effect:** `bf-3` in `behavioral-finance-201` — for the behavioral mechanism that breaks improvised exits. - **The investment-thesis structure and the bear case:** `ptk-1` in `practitioner-toolkit-201` — for the anatomy of an investment thesis that includes the falsification trigger. - **Thesis-drift checks before triggering an exit:** `ptk-12` in this path — for the diagnostic that distinguishes a broken thesis from a disagreeing market. - **Portfolio-level stop and drawdown discipline:** `risk-2` and `risk-5` in `risk-management-201` — for the portfolio-level constraint that overlays the single-name exit plan.
Sit with the ideas.
You wrote at initiation: "Exit if next-quarter receivables-vs-revenue gap exceeds 4 percentage points without a contracting-mix explanation." The quarter prints and the gap is 5.2 percentage points; management's MD&A attributes it to a one-time integration of a recently acquired distributor's billing system, with documented rollback in the following quarter. Which exit discipline applies?