Investing during a market crash

Survival rules keep you solvent in a drawdown. They do not keep you sane, and an investor who is insolvent in nerve sells the bottom just as surely as a margin call does. The behavior gap (M6.05) is temperament denominated in money, and it opens widest precisely in drawdowns, because losses hurt about twice as much as equivalent gains feel good (prospect theory). A 50% paper loss does not feel like giving back gains. It feels like a five-alarm emergency demanding the one action (sell) that is usually exactly wrong.

M6.05 built the temperament and the tools. What follows applies them inside the loop as an explicit protocol, because in a real drawdown you will not improvise your way to good behavior. You will run a pre-written procedure or you will panic.

What should you do when a stock you own falls sharply?

The position-level drawdown protocol. A position you own is down sharply. Before you are permitted to do anything, in this order:

1. Do not act on the down-day. Enforce the cooling-off rule (M6.05): no sell on a down-day unless a kill criterion has objectively, unambiguously triggered. The first job is to insert friction between the amygdala and the order button.
2. Open the original journal entry, the one your calm, unpositioned past self wrote (see the decision journal). Read field 7 (kill criteria) and field 8 (monitoring metrics) as written then, not as you would rewrite them now.
3. Ask the only question that matters: has anything structural actually broken, or is this price/sentiment? Check the monitoring metrics against reality. Did a load-bearing assumption (field 2) fail? Did a kill criterion (field 7) trigger? Or did the business perform while the price fell?
4. Branch on the answer. If a kill criterion has genuinely triggered, the thesis is broken and you sell, not because the price fell but because the reason you owned it is gone; the recovery math ("but I need +100% to get back to even") is now irrelevant and is itself the sunk-cost trap (M9.03). If nothing structural broke, it is sentiment, the drawdown is a lower price for the same business, and the disciplined action is to hold, or, if your sizing policy and conviction warrant and it does not breach a limit, to add. Either way, log the decision and its reasoning as an appended, dated journal update. You are making a decision, and it belongs in the record.

What should you do when the whole market falls?

The portfolio-level (market-crash) playbook. When everything is down 30–50% at once (2008, 2020, a seeded simulator cycle), the same logic scales up, plus three additions. (a) Confirm the survival floor is holding: no margin call, no forced redemption, cash reserve intact. If you built the survival floor correctly it is holding, and that fact alone converts the crash from emergency to opportunity. (b) Return to the bench (M9.02). Crashes are when the best businesses go on sale, so this is when your pre-built watchlist and pre-written theses pay off, because you can act from prepared analysis rather than trying to think clearly in a panic. (c) Rely on the environment design (M6.05): the scheduled (not compulsive) price-checking, the muted tip channels, the pre-committed rules. A crash is exactly the moment willpower fails and only structure holds.

The investors who compounded through history's crashes were not the ones who felt no fear. They were the ones whose process felt the fear for them, a calm past self written into kill criteria and a bench and a survival floor, making the decisions a panicked present self merely executed. Temperament, in the end, is downstream of process: build the process, and equanimity has somewhere to stand.