How to Use This Book

The three reading modes

This book is optimized for three different reading sessions. Pick the one that matches what you’re doing right now.

Mode 1 — Read Before Your Next Game (15–30 min)

You have time before play. You know your spirit and adversary. Do this:

1. Open the spirit’s chapter in Part III.
2. Read At a Glance + Spirit Overview + Key Strategic Principles. That’s ~600 words.
3. Scan the Adversary Matchup Matrix row for your adversary. If the grade is B or worse, read the matching row carefully and note the 1–2 reasons.
4. Read the Opening Strategy subsection matching your draft choice.
5. Skim the Common Mistakes list. That’s the payoff per unit of reading time.

If your adversary is one you haven’t played much, add the adversary’s At a Glance + Spirits That Shine/Struggle from Part IV. Another ~300 words.

Mode 2 — Build a Session Plan (45–90 min)

You’re new to the spirit and new to the adversary. Or you’ve been losing and want to diagnose why. Do this:

1. Read the spirit chapter fully, including Card Priority Ratings and Tempo Profile.
2. Read the adversary chapter’s level-by-level breakdown for your chosen level.
3. Pick an Opening Strategy variant; write down its round-by-round goals on paper (literally — pen on paper, not a screen).
4. Play.
5. Debrief with the si-post-game Claude skill and compare your turns against the chapter’s Common Mistakes.

This is the highest-leverage mode. Most readers should default to this for any new spirit/adversary combination.

Mode 3 — At-the-Table Lookup (under 30 sec)

Mid-game. You have a decision and need signal, not a dissertation. Do this:

• Use the si-at-the-table Claude skill. You describe board state; it returns options with tradeoffs, not a solve.
• Or, if you have the book open on a second screen: search for the specific card name, or jump to the spirit’s Tempo Profile for the current round.

Never try to read a whole chapter mid-game. It takes too long and breaks the table’s flow.

The active-learning loop

The book alone won’t make you a master. Books rarely do. The loop is:

1. Challenge — generate today’s drill with si-daily-challenge. It picks a spirit × adversary × level × learning-goal based on your play history.
2. Read — spend 15 minutes on the spirit chapter (Mode 1 above).
3. Play — the challenge. Solo, multi-handed, or at-the-table.
4. Log & Debrief — run si-post-game. It logs to data/playlog.csv and surfaces the 1–2 mistakes most likely to be what cost you this game.
5. Iterate — tomorrow’s challenge shifts based on what you logged.

Repeat. After ~20–30 sessions you’ll notice you’re reading chapters less, because the patterns are in your head.

Notation you’ll see everywhere

This book uses the community-standard notation popularized by latentoctopus. Memorize this table:

The admonition palette

Six kinds of callout blocks, each with a specific meaning. See ADMONITION_STYLE for the style guide.

• Pro Tip — actionable in-game advice.
• Common Mistake — frequent trap + why.
• Stat Insight — data-backed claim with source/date/sample.
• Rules Clarification — from Querki FAQ, not opinion.
• Sources — citations at section or chapter end.
• Math — probability calculations, worked examples.

Confidence ratings (inherited from latentoctopus)

Opening-strategy subsections sometimes carry a confidence badge:

• 🟥 Red — tentative / theorycrafting.
• 🟨 Yellow — somewhat confident, limited play-testing.
• 🟩 Green — well-tested, consistent success.

If a chapter doesn’t carry a badge, assume the content is in the “yellow” tier.

How to give feedback

This book is a living document and feedback changes it.

• If a chapter’s advice failed you mid-game, note it in your si-post-game entry. Batched across sessions, these become revision tickets.
• If a stat is stale (digital patch shifted things), open an issue on the repo or DM Brett.
• If a common community pattern isn’t covered, same.

See Meta Changelog for the running edit log.

Last revised: 2026-04-19