The problem
You're about to make a real decision — the job, the investment, the confrontation, the trade. You don't need help wanting to decide well. You need help with the three specific places good intentions fail: you commit without checking whether the bet is actually worth it, something rattles you mid-decision and you don't notice, and afterward you grade the whole thing by how it turned out instead of whether the process was sound.
Each of those failure points already has its own page on this wiki, built from the same decision discipline professional poker players use to separate a winning career from a losing one. What's been missing is the map showing they're one system, not five unrelated tricks. This page is that map.
The mechanism
Every high-stakes decision has three moments, and each one fails a different way.
Before you commit, the failure is skipping the math. Intuition grabs the most vivid possible outcome — usually the scariest one — and reacts to it as if it were the only branch. The correction is the 5-Question Pre-Bet Audit, which forces the slow, deliberate brain online for 30 seconds before you commit, and folds in expected-value thinking — weighing each outcome by how likely and how valuable it is, instead of fixating on one.
While pressure is live, the failure is not noticing your own state. Something just went badly — a loss, a rejection, a mistake — and the next stretch of decisions gets made from a different, worse-informed state than the one before it. The Tendler 7-Tilt taxonomy diagnoses which of seven distinct emotional states is actually firing, because the generic advice to "calm down" fails when the correction doesn't match the specific belief underneath the tilt.
After it's over, the failure is grading the result instead of the decision. A sound choice can still lose; a reckless one can still win — that's resulting, and it quietly trains you toward worse decisions while feeling like learning. The Mental Hand History is the structured five-step debrief that catches what free-form reflection skips, precisely so the lesson you take away is about the process, not the scoreboard.
The operating system
Before: run the pre-bet audit
Thirty seconds, five questions, before you commit to anything with real stakes: your range of possibilities, the expected value, whether you're tilted, your best alternative, and the worst case you can live with. It's the checkpoint that catches a decision made on pattern-match instead of math.
Before: weigh it by expected value, not the scariest outcome
Inside the audit, question two is doing the real work: list the realistic outcomes, estimate how likely and how costly each one is, and pick the best weighted average — not the option that avoids the worst case, and not the one with the flashiest upside.
During: name the tilt, don't just try to calm down
If something has already gone sideways and you're mid-decision, generic calm-down advice will fail if it doesn't match the specific belief driving the state. Diagnose which of the seven tilts is firing — Running-Bad, Injustice, Hate-Losing, Entitlement, Revenge, Desperation, or Mistake — and apply its specific correction, not a generic one.
After: write the mental hand history
Ten to fifteen minutes, five fixed steps: the facts, the emotion, the tilt type, the flawed belief underneath, and the corrected belief in your own words. The fixed format is what makes the hard step — naming the flawed belief honestly — mandatory instead of optional, which free-form journaling always skips.
After: judge the process, not the result
Once the debrief is written, grade it honestly: given what you knew at the time, was the call sound? That's a different question from whether it worked out, and conflating the two is resulting — the trap that teaches you to abandon good decisions because a good decision can still lose.
The five tools, in order
Each one is a full free page with its own printable card. This page is the index; they're where the actual protocol lives.
The State-Matched Decision Tree
Run the audit and the tilt check as one guided tool instead of reading both pages separately. Answer honestly and it tells you: correct the state, or make the play.
Run the tool → Wiki · Free · Before5-Question Pre-Bet Audit
The 30-second checklist to run before any high-stakes decision. Range, expected value, tilt check, BATNA, worst case.
Read the page → Wiki · Free · BeforeExpected-Value Thinking
Weigh each outcome by how likely and how valuable it is. The logic behind question two of the audit.
Read the page → Wiki · Free · DuringTendler 7-Tilt OS
Tilt isn't one thing. Seven distinct types, each with a different trigger and a different correction.
Read the page → Wiki · Free · AfterMental Hand History OS
The five-step structured debrief that turns an emotional spiral into a learnable pattern.
Read the page → Wiki · Free · AfterProcess vs Outcome
The right call can still lose. Judging decisions by results trains you to decide worse over time.
Read the page →Two more pages apply this same discipline to a specific high-stakes decision rather than the general case: the Medication Decision Framework for a prescriber conversation, and salary negotiation for a comp conversation. Same OS, narrower input.
The printable: the decision OS card
Print it. One card, three moments, five tools.