Behavioral Finance & Microstructure
Thinking, Fast and Slow
Kahneman's map of the two systems that produce every trading decision: a fast one that answers instantly with a biased guess, and a slow one that could check the guess but usually doesn't. The edge is in forcing the check.
The big picture
The book distills four decades of research into one architecture: System 1 produces impressions instantly, effortlessly and with systematic biases; System 2 can audit them but is lazy and gets invoked mainly to rationalize what System 1 already decided. The core bet for an investor: your errors are not random noise to diversify away — they are predictable, directional and exploitable, by you (through process) or against you (by the market).
Why it matters now: a narrative-heavy AI tape is a System-1 amplifier. Recency, anchoring on round numbers and highs, and story-driven confidence all get stronger exactly when dispersion is widest and mistakes are most expensive.
The 3 strategic pillars
-
Two systems, one output
Every buy/sell impulse arrives pre-formed from System 1; deliberation mostly decorates it afterwards.
The practical consequence: interventions work at the process level (checklists, cooling-off rules), not at the willpower level.
-
WYSIATI — what you see is all there is
Confidence tracks the coherence of the story at hand, not the completeness of the evidence.
The corrective is a forced base-rate question: of all situations that looked like this, how many resolved the way the story predicts?
-
Loss aversion and narrow framing
Losses weigh roughly twice as much as gains, and each position gets judged alone rather than as one draw from a portfolio-sized series.
Produces the disposition pattern — selling winners early, nursing losers — and evaporates when outcomes are framed as a batch of repeated bets.
What a Closelook reader does with it
The working use is a pre-trade gate that interrogates the impulse before the order: what is the base rate, what would change my mind, am I anchored on a price that means nothing, is this position sized as one of fifty bets or as a verdict on my judgment? The mistake it prevents is the compound one — a System-1 entry defended by System-2 rationalization and then held by loss aversion. Ten honest questions before the order cost two minutes; the biases they catch cost drawdowns.
The bridge to the Closelooknet approach
Closelooknet's data layer is, in Kahneman's terms, an externalized System 2. Money Temperature measures the crowd's System-1 state directly — when the temperature runs hot, this book explains exactly which errors are being manufactured at scale. The structure read on the asset pages exists to replace anchor prices with computed levels, and the analog backtest answers the base-rate question ("how often did this setup hold?") that WYSIATI never asks on its own. The pack's checklist is the manual counterpart: the questions our tooling can't answer for you, asked before the order instead of after the loss.
Action-Kit — from theory to practice
Tooling & data
| What you need | Where to get it | Cost |
|---|---|---|
| A trade journal with entry reasons written BEFORE the fill The only reliable bias detector — hindsight rewrites everything written after | Any note system; the pack's decision log adds the structure | Free |
| Base rates for your setups The WYSIATI antidote — replace story-confidence with frequency | Your journal statistics; the asset-page analog backtest publishes held-rates per structure setup | Free |
The formulas
-
Bias-gate score
Gate = passed checks / 10 — below 8, size down or stand down- 10 pre-trade questions across anchoring, recency, confirmation, overconfidence, loss framing, sunk cost, base rates
The score is deliberately blunt; its job is to interrupt, not to measure.
-
Batch framing
Judge outcome distributions per 20 trades, never per trade- Journal outcomes in blocks of 20
Kahneman's broad-framing prescription operationalized — single-trade P&L is noise to the process.
Applied Pack · free members
Kahneman Applied Pack
The pre-trade bias gate: ten questions that interrupt System 1 before the order, plus a decision log that scores your last twenty trades as a batch.
- Bias_Gate.xlsx — the 10-question pre-trade checklist with scoring, a per-bias explanation column, and a batch view that aggregates your last 20 logged decisions
- decision_log.py — stdlib-only CLI: logs each trade decision with its gate score to a CSV and prints the batch statistics (gate score vs. outcome) so the checklist proves or disproves itself on your own data
- README.txt — the ten biases in one page each, sourcing guidance and the educational-use disclaimer
Pack security
Macro-free Excel · plain-text Python you can read before you run it · no installers, no network access — the code works only on files you provide. Served only from closelook.net; we never distribute through download portals or email attachments. How to verify in 30 seconds →
SHA-256 9c2eccdb8e6782a9fbbe629a6e67a1459fb7694829c2311cdd12df2e94687661
The pack is a free-membership download — no card, free forever. Membership also delivers the Daily Pulse and Weekly Signal to your inbox.
Get free access → Already a member? Sign in →
Educational templates — a research diary companion, not investment advice.
Closelook publishes a market diary, not investment advice. This condensed read restates the book's ideas in our own words for education — for the author's full argument, go to the source.