Laboratory and field studies show structured pre-mortem analysis improves judgment and outcomes.
Effect sizes vary by context. They often fall in the low double digits.
Those ranges sit around 10–25 percent for reduced overconfidence or decision error in experiments.
Report percentages as ranges. Link each estimate to the study context when possible.
Decision-makers often lose deals and options because of overconfidence, narrow exposure, and weak stop rules.
Process gaps, not fate, explain most missed opportunities.
Decision-Making Hacks for Luck: change decision processes to increase luck.
Use probabilistic thinking, pre-mortems, kill criteria, and exposure tactics to raise opportunity rate.
Measure outcomes with simple KPIs.
Evidence-based hacks increase expected value and reduce bias.
A 3–6 week experiment can validate effects.
Log every decision, run a pre-mortem before high-stakes choices, and set kill criteria.
A/B test checklists and track opportunity rate plus calibration error.
Small wins compound over weeks into clear improvements.
Key variables that change your luck outcomes: exposure
Measure exposure, expected value per attempt, and forecaster calibration weekly to see quick change.
A single process change that raises exposure by 20 percent can shift expected value more than chasing rare outcomes.
Measure each weekly to see change fast.
How to measure exposure as a KPI
Define exposure as countable events you control: outreach attempts, experiments run, or new connections made.
Count only events that involve an explicit attempt or contact longer than five minutes.
This yields an opportunity rate that you can trend weekly.
How to turn choices into expected value numbers
Estimate p(success) and payoff in dollars or utility, then multiply to get EV.
Compare EV per hour or per dollar spent to rank options.
This converts gut feelings into comparable metrics.
What calibration means and how to score it
Calibration measures how close probability estimates are to actual outcomes.
Use a scoring rule like the Brier score to quantify calibration.
Lower Brier scores mean better probabilistic judgment.
Daily or weekly forecasting drills let the Brier score show progress over 4–8 weeks.
Controlled lab and field work on structured decision techniques show measurable impacts that depend on context.
Pre-mortem analysis and formalized checklists reduce overconfidence and uncover failure modes.
Lab studies and field tests often report effect sizes in the low double digits.
Those ranges sit around 10–25 percent for reduced overconfidence or error.
Calibration training and forecasting drills produce concrete gains.
Short repeated drills with feedback commonly improve Brier scores by 10–30 percent over a month.
That improvement feeds into better p(success) inputs for EV math.
Small wins compound over weeks into clear improvements.
A/B testing and rapid pilots produce varying conversion uplifts depending on context.
Expect single-digit to low double-digit percent improvements from iterative optimization in many settings.
Frame these numbers as ranges to keep probabilistic thinking honest.
Translate improvements into dollars, time saved, or other utility units.
Concrete exposure metrics and dashboard KPIs you can track
Define opportunity_rate KPI as 'meaningful_attempts_per_week'.
This counts attempts longer than five minutes or full experiments.
Complement it with conversion_rate, average_EV_per_attempt, cumulative_EV, and EV_per_hour.
Track a 4-week rolling average for opportunity_rate and conversion_rate to smooth noise.
Add a volatility indicator to detect inconsistent effort.
Set explicit thresholds like a 20% lift in opportunity_rate or a minimum EV_per_hour.
Log whether each week meets the stop rules.
Present these numbers on a simple dashboard.
Include weekly attempts, conversion rate, average EV per attempt, EV per hour, and Brier score.
Small wins compound over weeks into clear improvements.
Apply simple decision rules to raise expected value now
Use a one-page EV scorecard, a pre-mortem checklist, and a kill criteria sheet to change which bets stay live.
These three rules reduce bias and increase long-run expected payoff.
Applying them moves choices from random luck to a controllable process.
What to put on an EV scorecard
Required fields include option name, estimated p(success), expected payoff, time cost, EV, and EV per hour.
Rank by EV per hour for time-limited contexts.
Rank by EV per dollar for budget-limited contexts.
The error most frequent in this step is underestimating time costs.
Include time explicitly.
How to run a pre-mortem in 10 minutes
Assume the decision failed and list three plausible causes.
Assign rough probabilities to each cause and add mitigations.
Adjust the original p(success) downward based on the sum of realistic failure causes.
This step surfaces hidden risks and reduces overconfidence.
Kill criteria
Pick at least one rule before you start.
Examples include stop if cost exceeds X dollars or after N unsuccessful trials within M days.
A useful default is stop after three failed experiments.
Alternatively stop once time spent exceeds 10 hours unless EV justifies continuation.
This prevents sunk-cost bias and frees resources for higher EV opportunities.
Set a baseline: track opportunity rate and EV per hour this week. Aim to improve either metric by 20% within four weeks; small lifts compound into larger expected-value gains.
The recommendation works only if the decision log is kept honestly and reviewed weekly. In practice, many people stop logging after the first week.
Expect bookkeeping friction and plan for 10–20 minutes weekly review time.
An anonymous case: a mid-level product manager tracked 12 decisions for four weeks.
They discovered 40 percent of time went to low EV work.
A reallocation then raised weekly EV by 45 percent.
Small wins compound over weeks into clear improvements.
Measure and grow your opportunity rate
Track opportunity rate as a single weekly number.
Count meaningful outreach attempts, experiments launched, or new introductions made.
Increasing that number increases chances for positive outcomes.
Each attempt carries nonzero EV.
How to define a meaningful opportunity
Count only actions that could plausibly lead to a positive outcome.
Examples: a five-minute conversation, a logged cold email, or a finished experiment.
Exclude low-signal items like passive follows or brief social likes.
This avoids diluting the metric with noise and makes growth measurable.
A/B test your exposure tactics
Run two outreach strategies in parallel for four weeks and compare EV per hour and conversion rates.
Randomize contacts to avoid selection bias.
Compute uplift with a simple difference-in-means.
The data then show which tactic scales and which wastes time.
Expected baselines and growth targets
Baseline opportunity counts vary widely by role and industry.
Some contributors run fewer than five meaningful attempts weekly.
Some sales roles may exceed twenty meaningful attempts weekly.
Target a 20 percent increase in month one.
Then aim for a 10 percent increase per month thereafter for sustainable growth.
If the baseline is below five, focus first on low-cost, high-exposure moves like two outreach templates per day.
Small wins compound over weeks into clear improvements.
Count
Log each meaningful outreach or experiment.
Estimate
Record p(success) and payoff for each attempt.
Compute
EV = p × payoff and EV per hour.
Review
Weekly review: drop low EV items, double high EV ones.
Calibration drills to improve probabilistic intuition and decision-making
Practice forecasting with daily binary or interval questions and score with the Brier score or hit rate.
Calibration drills reduce overconfidence.
They improve p(success) estimates that feed EV math.
Evidence shows measurable progress after consistent practice over 4–8 weeks.
Binary forecast drill and scoring method
Pick ten binary questions per week you can verify within days to weeks.
Assign a probability to each question, such as 70 percent.
Score with the Brier score and track the weekly trend.
Lower scores are better.
A typical beginner improves Brier score by 10–30 percent after four weeks of practice.
Interval estimates and hit-rate drill
Give 90 percent confidence intervals for numeric questions.
Track the percent of true values inside those intervals.
The target is close to the nominal level, ninety percent.
If too many answers fall outside, widen intervals and review the hypotheses causing the misses.
Practical 4-week calibration plan
Week 1: baseline quizzes and simple Bayesian reminders.
Weeks 2–3: daily drills with weekly scoring and one feedback session.
Week 4: double exposure while continuing drills.
Expect measurable improvement if scoring and feedback stay consistent.
Kill criteria and the decision log templates you must use
A rigid template reduces the chance of keeping poor ideas for emotional reasons.
The required columns force explicit stop rules and improve long-term expected value.
Use the templates below verbatim for reliable tracking.
Decision log template
| Date |
Decision |
Hypothesis |
p(success) |
Payoff ($ or utility) |
Time est (hrs) |
EV ($) |
Kill criteria |
Action taken |
Outcome date |
Result |
| 2026-06-01 |
Launch outreach A |
15% will respond |
0.15 |
10000 |
8 |
1500 |
Stop after 3 fails or 20 hrs |
Sent 50 emails |
2026-06-22 |
TBD |
Kill-criteria sheet
- Stop if cost > $X relative to expected payoff.
- Stop after N failed experiments where failure = no measurable progress toward the stated hypothesis.
- Stop if EV per hour falls below a preselected threshold for three consecutive reviews.
Small wins compound over weeks into clear improvements.
Pasteable EV calculator and an A/B test checklist are ready to copy.
EV per hour calculator (one-line): EV = p(success) × payoff
EV per hour = EV ÷ time_hours.
Example: p=0.10, payoff=$20,000 → EV=$2,000.
Time=8 hrs → EV per hour=$250.
Decision log snippet (compact): Date | Decision | p | Payoff | Time_h | EV | Kill_rule | Status.
Simple A/B test checklist (use these bullets as a reproducible script):
- Randomly assign contacts to A or B.
- Run for a prespecified exposure window, for example 4 weeks.
- Predefine a primary metric such as EV per hour or conversion rate and stop rules.
- Record sample sizes and compute difference-in-means and simple confidence intervals for uplift.
- Operationalize the winner by doubling down on the higher EV tactic.
These micro-tools convert recommendations into immediately actionable calculations and checklists the reader can copy into a spreadsheet or a lightweight app.
Compare affordable coaching and self-study options for skill development
Choose a format that fits budget and target EV improvement per hour.
Self-study scales cheaply but demands discipline.
Group courses add structure.
One-on-one coaching shortens the learning curve but costs more.
Estimate expected skill gain as EV uplift over baseline.
Divide that by cost and hours.
Choose the cheapest option that yields positive EV per hour relative to your time value.
If time is scarce and EV per hour target is high, one-on-one coaching often wins despite the price.
| Option |
Cost (est) |
Time |
Expected skill gain (4 wks) |
| Self-study (books, drills) |
$0–$200 |
3–6 hrs/wk |
Small improvement, requires discipline |
| Group course (online) |
$200–$1,000 |
2–4 hrs/wk |
Moderate improvement, structured feedback |
| One‑on‑one coaching |
$500–$3,000 |
1–3 hrs/wk |
Rapid improvement, tailored guidance |
How to pick based on ROI
Estimate expected skill gain as EV uplift over baseline and divide by cost and hours.
Choose the option that gives the best EV per hour relative to your time value.
If time is scarce, favor higher EV per hour even if cost is higher.
Small wins compound over weeks into clear improvements.
Numeric before/after cases you can copy this week
Short concrete cases show how the method changes measurable outcomes.
Each case gives baseline numbers, the change applied, and the new expected-value result.
Replicate these steps with your own numbers to validate change.
Case 1. sales outreach: exposure lift
Baseline: 10 contacts per week, p=0.08, payoff=$20,000, weekly EV=$160.
Change: increase to 13 contacts per week (+30%) and use calibrated p=0.10.
Result: weekly EV rises to $260, a 62 percent increase in expected value.
Case 2. side project selection: EV filter
Baseline: three side projects running, none tracked, time scattered, no clear ROI.
Change: apply EV scorecard and kill criteria, keep only projects with EV per hour above your threshold.
Result: time focused on one project with positive projected EV; completion rate rose and expected payoff became measurable.
Case 3. forecast training improves decision
Baseline: untrained forecasts, Brier score X.
Change: run daily binary drills for four weeks with feedback and adjust priors.
Result: Brier score improves by 15–30 percent depending on practice intensity.
Richard Wiseman observed that people who act on more opportunities end up luckier.
Behavioral science from Kahneman and Gigerenzer explains why process matters.
For the original research and summaries see Wiseman’s work on luck.
Commit to a 4-week experiment now. Copy the decision log template above and pick three weekly metrics.
Choose opportunity rate, EV per hour, and Brier score as those metrics.
Schedule a single 30-minute weekly review block.
Frequently asked questions
How fast will these hacks change outcomes?
Expect measurable change in exposure and EV within four weeks.
Calibration gains usually take four to eight weeks with consistent practice.
Track weekly and compare baseline to week four to see progress.
How to estimate payoff for nonfinancial personal outcomes?
Translate outcomes into utility units like hours saved, skills gained, or networking value.
Assign a dollar equivalent when possible and keep the scale consistent across options.
If assigning dollars is impossible, rank options by expected impact and time cost for a rough EV per hour.
What does a reliable opportunity look like in practice?
A reliable opportunity is an attempt that could plausibly deliver value.
Examples: a 30-minute discovery call, a prototype test, or a scheduled mentor meeting.
Avoid counting passive interactions and set rules before you start to keep the metric clean.
Can these methods harm creativity or exploration?
If applied rigidly, EV filters can cut off useful exploration.
Use a two-track approach: one track for exploration with looser kill windows and one for exploitation with strict EV rules.
Balance both to preserve discovery while improving overall payoff.
How does one validate p estimates objectively?
Use pre-mortems, outside views, and small pilot tests to update priors.
Solicit independent estimates from peers or mentors and compare results to spot bias.
Record and compare predicted p to actual outcomes to improve calibration over time.
The plan to run your 4-week test
Week 0: copy the decision log template and record baseline opportunity rate, EV per hour, and Brier score.
Weeks 1–3: run the exposure growth plan, apply EV scorecard to new decisions, and do daily calibration drills.
Week 4: review the three KPIs, compute the change, and decide which processes to keep, refine, or kill.
⚠️ This plan does not apply to emergency decisions, legal constraints, or strictly random events like lotteries. If time or oversight prevents honest logging, skip the test until accountability is restored.
Which scoring method should be used for calibration?
Use the Brier score for binary forecasts and hit rate for confidence intervals.
Brier score gives a continuous penalty for distance between predicted probability and outcome.
It is simple to compute and shows steady improvement when practice is consistent.
