You named the enemies: affect paints probability before math; confirmation defends the story; anchoring glues the first price; overconfidence inflates edge and size; loss aversion distorts exits; endowment worships what you hold; recency chases the last headline; hindsight rewrites the past after resolve.
13.9 is engineering—how to build systems that counter your biases so you do not rely on mood, caffeine, or a motivational thread when prices spike. Willpower depreciates; friction, defaults, and logs compound like edge.
This closes Psychology & Behavioral Finance. Everything below assumes you already have expected value, bankroll caps, journaling, and structured forecast habits—and makes them default-on.
System design principles
Default-on discipline means checklists required to trade, not “I’ll remember.” Friction for sins means hard steps for market orders after headlines. Measurement beats memory with locked forecasts, tags, and Brier—not post-hoc stories. Separation of roles splits scout, trader, and scorer even on a solo account. Fail closed means no trade if a gate fails. Small reversible bets mean limits and fractional Kelly, not illiquid all-in heroics.
Systems do not remove bias—they tax biased actions. A gate that takes ninety seconds is cheaper than a tilt trade that costs ninety dollars and an afternoon of rumination.
Willpower is a depreciating asset
Sleep debt, losing streaks, election weeks, and personal stress drain willpower the same week you need it most. Defaults win when you are tired: limit-only mode, alerts off, scout-only mornings. Design for Thursday-you at eleven p.m., not Monday-you after coffee.
Bias-to-system map (Module 13 complete)
Affect: emotion score four or higher blocks market orders; twenty-four-hour sleep on brand-new thesis. Confirmation: steel-man paragraph before entry; tier-one-only week if tags pile up. Anchoring: hide entry; blind re-buy test; fresh probability line in journal. Overconfidence: consensus check; calibration shave map; half-Kelly cap. Loss aversion: thesis stops; bankroll freeze after drawdown. Endowment: exit sketch at entry; weekly re-buy audit. Recency: cooling room plus event calendar; shadow forecasts. Hindsight: immutable timestamp log; Brier on locked forecast.
If a trade violates two systems, pass is often highest expected value.
Three roles, one person
Scout reads news, updates dossier, drafts probability—no orders. Trader sizes, limits, exits per rules—no probability changes without scout log. Scorer runs weekly Brier, tags, rule changes—not same day after huge P&L.
Rotation trick: scout note in morning, trader after lunch, scorer Sunday only. Same-day scorer after a big win is hindsight bait; same-day trader after a scary headline is recency bait.
One-page pre-trade gate (worked example)
Watchlist contract: “SEC approves ETF by DATE.”
Scout quotes resolution PDF and edge cases. Blind 62% written before opening the app. Trader sees 58¢ market near 57–60¢ consensus, positive expected value near four cents pre-fee, steel-man NO case read aloud, size 1.2% bankroll half-Kelly, invalidation text if withdrawal denied headline, emotion 2/5, tags macro and regulatory.
Decision: pass if edge within fees—set 56¢ limit, seven-day time-to-live.
Gate failures: emotion five, no steel-man, probability moved only after seeing price—no trade.
Cooling room as scheduled system
CPI tier-zero triggers thirty-minute alert mute—scout log only. Five-cent move inside an hour without news locks trader. Viral tweet tier-three only—no probability change. Wednesday evening review unlocks trader actions.
Outcome: fewer trades, higher average edge; weekly audit shows fewer recency tags.
Dashboards that matter
Monthly Brier by bin triggers shave on hot bins. Weekly overconfidence tag rate forces consensus check. Percent of holds failing re-buy triggers forced exit review. Market-order share triggers extra confirmation friction. Shadow versus traded gap promotes shadows with stable Brier. Rule violations drive one new habit install—skip P&L-only dashboards that reward hindsight.
Environment and portfolio
User experience defaults: limit orders, dispersion alerts, separate play-money wallets. Portfolio: correlation caps, theme limits, joint expected value on trees.
Thirty-day rollout (worked example)
Week one: one-page gate on every entry. Week two: cooling room on macro calendar—zero T+0 market orders. Week three: blind probability before price—timestamps prove order. Week four: Sunday scorer hour—Brier plus one rule change. Keystone habit: one shadow forecast daily feeding calibration without bankroll risk.
Minimum viable bias operating system
One pre-trade gate. One cooling rule after tier-zero news. One weekly scorer hour. One bankroll cap without exceptions. One shadow forecast per day. Add friction before adding indicators.
Buddy systems (solo-adaptable)
Blind probability submit before debates; weekly steel-man swap; immutable log share per event; monthly single process patch. Past-self letter Monday read Friday works when no partner exists. Accountability does not require a desk neighbor—it requires a record another day can contradict.
Compliance as edge
Track gate compliance percentage, not just P&L. A week at one hundred percent gates with flat P&L beats a hero month that violated every cooling rule—because the hero month teaches habits that blow up later. Expand systems only when compliance is stable; adding indicators before gates work is decorative suffering.
Module arc tie-in
Crowds need your discipline, not ego. Microstructure and cascades imply cooling and limits. Math inputs lock before story. Complex trees need dossiers per leg. Venue choice is Friday data, not identity. Execution chapters become policy. Signals feed scout; trader waits. Calibration closes the loop. Builders shape user bias through defaults. Professional use needs audit trails.
Passing and shadowing are professional features—not cowardice.
When systems fail loudly
If you override a gate, log override with reason. Occasional overrides are human; unlogged overrides teach you gates are decorative. Monthly review: count overrides; if rising, shrink scope or fix the gate design, not your willpower.
Builders and readers
If you run a portal or teach this curriculum, defaults matter: stale-data banners, limit-first UX, links to resolution PDFs, and copy that praises passes reduce user-level recency and overconfidence. Product shape is a bias system for strangers you will never meet.
Start small, stay small
Install one gate this week, not twelve. Compliance on a single habit beats a poster of twenty rules you ignore. Module 13 is complete when the habits run on bad nights—not when you memorized the vocabulary.
Handoff to case studies
Systems are insurance. Module 14 Case Studies & Post-Mortems applies them to real elections, ETF approvals, disputes, portfolios, manipulation attempts, and repeated mistakes on actual price paths and resolutions.
Next: Module 14 Case Studies & Post-Mortems — Case Study: 2020 US Presidential Election Markets