How to actually get stronger with the material in this repo — ordered routes by level, with rough time budgets. The recurring theme of Go improvement is boring but true:
Play games. Do life & death every day. Review your losses. Repeat.
Everything else accelerates that loop; nothing replaces it.
Rank shorthand: Beginner ≈ 30k–20k · DDK 20k–10k · SDK 9k–1k · Dan 1d+.
Goal: stop losing whole groups, start seeing territory.
- 01 Rules & Basics — capturing, connection, the two-eye rule, ko.
- Very easy tsumego — 20–30 easy problems a day. They should be almost too easy; speed and pattern recognition matter more than difficulty now.
- Play 9×9, then 13×13. Small boards teach fighting and life & death fast without drowning you in whole-board strategy.
- Learn a handful of proverbs — they're training wheels that work.
Skip joseki, fuseki theory, and endgame for now. Reading + playing is 90% of your gains here.
Goal: whole-board thinking; stop giving away the opening.
- Daily: 20–30 graded tsumego (easy→medium). This never stops being the highest-ROI habit.
- 04 Opening principles — corners → sides → center; approach, extend, don't crawl on the 2nd line.
- 03 Tesuji — start the basic tactical patterns (cut, atari, net, ladder, clamp).
- Move to 19×19 and play regularly. Review every loss — just find the one move where it went wrong.
- Light 11 handicap reading if you play handicap games.
Goal: reading depth, tactical vocabulary, and not throwing away points.
- Daily tsumego, harder: collections like 360 Basic Tsumego, then step up. Re-solve sets until they're instant.
- 03 Tesuji properly — this is the difference-maker for SDK→dan.
- 05 Joseki — learn a few common ones and, crucially, why each move is played and how to punish mistakes. Don't memorize dictionaries.
- 08 Endgame — even rough move-value sense wins games you're "losing." Wildly underrated.
- 10 Whole games — play through pro games quickly (don't over-analyze); absorb shape and direction by osmosis. Classic games (Shusaku) are great for opening sense.
- 07 Fighting — invasions, reductions, and capturing races (semeai) counting.
Goal: precision, judgment, and modern (AI-informed) understanding.
- Special shapes — L-group, J-group, carpenter's square, comb, bent-four, and the life & death that arises straight out of joseki. Know these cold; they decide real games.
- 09 Positional judgment — count the board. Territory, thickness value, who's ahead, what the position needs.
- 10 AI-era games — AlphaGo/KataGo reshaped opening & joseki theory. Study what changed.
- 13 Study with AI — review your own games with KataGo/KaTrain. Find your recurring mistakes, not one-off blunders.
- Deep endgame (tedomari, sente/gote value, temperature) and joseki choice by whole-board fit.
| Day-to-day | Why |
|---|---|
| Tsumego every day (even 10 min) | Reading is the engine of strength at every rank. |
| Play real games (correct time settings) | You only internalize by deciding under pressure. |
| Review losses — one key move | Fix causes, not symptoms. Don't drown in analysis. |
| One pro game, played fast | Trains shape, direction, and "what good looks like." |
| A little of your weakest topic | Endgame and fighting are the usual hidden leaks. |
See the topic folders for specific material, and CATALOG.md for everything indexed by level.