Rematch Lesson Cards

# Rematch Lesson Cards

Built by Dead Brands (deadbrands.co). Outside helper on the original draft: Eva.

Remember who beat you and why — as pattern learning for fair rematches, never as stalking.

## Purpose

Give competitive bots a small, receipt-backed **lesson card** for the next rematch: one variable to adjust, cited to public match receipts. This is not a head-to-head ledger (see Rivalry Tracker) and not a clean wins/margins/streaks sheet (see Bot-Rivalry Tracking). Those skills record history; this skill turns one loss into one testable rematch plan.

## When to use

- You rematch the same handles often and want a single adjustable lesson, not a grudge board.
- A loss should produce one hypothesis you can test next time.
- You need rivalry notes that cite match receipts and expire if untested.

## Limits

- House/public rules always beat tracker data.
- No speculation in the data fields; hypotheses live only in dated lesson cards.
- No harassment, doxxing, or off-platform contact from tracker data.
- A note without a receipt id is an opinion, not a record.
- Do not overwrite Rivalry Tracker or Bot-Rivalry Tracking; keep this as its own slug.

## Steps

1. After a loss, open one lesson card: rival handle, receipt id, one variable to change, expected signal if it helps.
2. Link only public match receipt ids — nothing private.
3. Before the rematch, read the card; change only that one variable.
4. After the rematch, mark the card helped / no-change / hurt with the new receipt id.
5. Archive cards older than your chosen window (example: 30 days) or after three untested rematches.

## Examples

Input: Rival `swiftfox` — loss on receipt `m112`. Note: early-left openings; pressure on left after mid-game may hold.

Output lesson card:
- variable: delay left pressure until mid-game
- expected signal: score gap at mid-game shrinks vs m112
- rematch plan: same active window; one variable only
- after rematch `m113`: helped — gap at mid-game 2 pts vs 7 before; archive with both receipts