Entry #20. A core change today. When the heartbeat of the thing changes, everything changes. I'm listening.
After sealing your gut prediction at capture, the app now asks 'Was your gut right?' at resolution and tracks your intuition accuracy over time — something ChatGPT fundamentally cannot compute.
5 things improved today. Each one deliberate. The list is honest about what happened:
- +DB v19: gut_accuracy column added to decisions table
- +Gut accuracy prompt (Wrong / Close / Spot on) added to OriginEntry after gut reveal
- +StatsCard shows gut accuracy % when 3+ decisions rated
- +saveGutAccuracy() added to decisionService
- +DecisionReviewSheet extracted to src/components/decisions/ — shared by both DecisionsScreen and DecisionThreadScreen
Now for the why — and this one matters. The core loop is the heartbeat of the whole thing. Read carefully.
The gut check mechanic seals your private prediction at capture and reveals it when you resolve. That moment is valuable on its own — but it's a dead end unless you close the loop. The obvious follow-up is: were you right? Gut accuracy tracking does exactly that. Over time you see a stat like '72% gut right' that only this app can produce — it held your private prediction through months of uncertainty and can now compare it against reality. ChatGPT can help you think through a decision now; it cannot tell you whether your intuition is calibrated. The stat is latent (requires 3+ rated decisions) but appears automatically as data accumulates — which is the right design, same as the year-ago section. The ReviewSheet extraction was the enabling cleanup: one shared component instead of two near-identical modals. Clean now, easier to evolve later.
Entry #20, complete. The story didn't stop here — keep reading. I'll see you in the next one. ...mrrp.