Reckon vs Writ

Reckon ranks hypotheses by what hasn't been knocked down — not by what has the most support. Most decision tools let you write a long pros-and-cons list and call it analysis. That doesn't help — you can rationalize anything in prose. Reckon makes you compare every piece of evidence against every hypothesis and ranks them by Heuer's ACH technique (the method CIA analysts use): fewest refutations first, with a tiebreak on most support. WHAT YOU GET • A starter library: pick from six pre-built example boards (shortseller, founder, researcher, investigator, life decision, doctor) or describe your situation in a paragraph and let on-device AI seed the matrix for you. • Inline bias guardrails: when one hypothesis collects evidence all in the same direction, Reckon flags it. "Bears out" without "Cuts against" is the structural signature of confirmation bias. • Falsifier pre-commitment: state up front what evidence would prove a hypothesis wrong. The app reminds you whether that evidence ever showed up. • Steel-manning before ruling out: before you dismiss a hypothesis, Reckon asks for the strongest case for it. Stops you from throwing out the right answer too early. • Pre-mortem: after you reach a conclusion, the app asks you to imagine being wrong about it in a year. Catches wishful thinking that even refutation-first scoring misses. • Calibration tracking: when you finalize, set your confidence (0–100%) and a check-in date (1 week to 1 year). Come back when the date arrives, log the actual outcome, and Reckon shows your calibration curve over time. Are you actually right when you say you're 80% sure? • Strict Bayes mode (optional): set explicit priors per hypothesis and the matrix shows real Bayesian posterior probabilities updating with each rating, using likelihood ratios per Tetlock/Heuer convention. Real probability, not a heuristic. • Plain English mode: turns off the celestial vocabulary if you'd rather see "Hypotheses / Evidence / Matrix" than "Stars / Observations / Star chart." Same depth, less poetry. • Templates: save your hypothesis structure (e.g. forensic-accounting suspects, differential-diagnosis options) for reuse. Faster on your 12th board than your 1st. PRIVACY Local-only. Your boards, evidence, conclusions, and calibration history stay on your device. No server, no analytics, no ads. AI seed runs entirely on-device via Apple Intelligence — your situation never leaves your phone. WHY ACH Confirming evidence is cheap. Refuting evidence distinguishes. Most decision tools amplify confirmation bias by rewarding biggest positive sum. Reckon does the opposite. iOS 17+. AI seed and Strict Bayes posterior require iOS 26+ on Apple Intelligence-capable devices.

Write the call down before you know how it ends. Writ checks back on the date you set, and shows you whether you were right. Most decisions vanish the moment you make them. The reasoning fades, the outcome gets rewritten in memory, and your gut never has to answer for anything. Writ keeps the record straight. WHAT MAKES WRIT DIFFERENT • Odds on every option, not just a mood. Give each option the probability you'd bet on, and your confidence comes from your top pick. • A tripwire. Name the warning sign that would tell you the call is going wrong, and Writ asks at review time whether it fired. • An honest score. Forecast sharpness shows whether your calls beat a coin flip, one review at a time. HOW IT WORKS 1. Log the decision: a plain statement of the call, the options you see, and how sure you are. 2. Set a review date. Writ reminds you when it arrives. 3. Answer honestly. What actually happened sits next to what you expected. 4. Watch your calibration sharpen, decision by decision. WHAT YOU CAN LOG • The situation and the options you considered, each with the odds you'd give it • Your confidence, the stakes, and the mood you decided in • A tripwire for every call • Tags, so Insights can compare your judgment across areas of life WHAT WRIT SHOWS YOU • Calibration: every dot is a call, your confidence against reality • Forecast sharpness: whether you beat a coin flip • The Forecaster Compass: bias and gut signal as one position • When you say 80%, does it happen 80% of the time? • Your top pick: how often your favorite option wins • By stakes and state of mind: where your judgment runs hot FREE TO START Your first five decisions are free, with full reviews, your running totals, the calibration chart, and the tag breakdown. Writ Pro removes the cap and unlocks the full insight set, PDF reports, and CSV export. PRIVATE BY DESIGN No accounts. No ads. No tracking. Your journal lives on your iPhone, with optional iCloud sync for backup. Writ is an indie app, built by a single developer. Terms of Use (EULA): https://www.apple.com/legal/internet-services/itunes/dev/stdeula/ Privacy Policy: https://writ.ku-software.com/privacy