A guide for every stage of your game.
- Learn
- Short explanations and hands-on examples to help you get started. Each lesson names the games it applies to and links to further reading.
- Dictionary
- Plain-language definitions, alternate spellings and tile examples. Each entry identifies the games it covers and links to its references. The dictionary uses WRC 2025 for Riichi and the WMO 2006 Green Book for MCR.
- Rulesets
- Scoring references for the versions played in Hé Fēn. Tournament editions and house rules can differ; Taiwanese Club follows our own club sheet.
- Strategy
- Ideas for building a hand, judging a call and reading the table. Choose the guide for your ruleset, then weigh the advice against the hand in front of you.
- Hé Fēn Lab
- Experiments with computer players that explore specific strategy questions. Each report explains the comparison and its limitations. These are simulations, not studies of human players.
Explore the original references.
These links take you to the publishers’ own material. For a tournament, check the edition and updates chosen by its organiser. Hé Fēn is not affiliated with the organisations listed here.
About our content
Our lessons are original explanations, created with AI assistance. They have not been independently reviewed by the organisations we reference. The links let you explore those organisations’ rules directly.
Mahjong rules vary by edition, club and family. Our guides name the version they describe so you can compare it with the rules at your own table.
American dictionary entries also draw on the American Mah Jongg Association’s independent guidance. It is separate from the National Mah Jongg League; disputed rulings should be checked with the card publisher or tournament organiser.
Our American mahjong introduction explains how card-based play works and where to find the National Mah Jongg League’s guidance. You’ll need your own annual card; we do not reproduce its hands.
The Lab includes inconclusive results as well as useful findings. The latest reports link to their data and explain uncertainty; older studies are labelled as archived and describe what their results can support.
Spotted something we should check?
Please tell us which page you were reading and what seems wrong or unclear. If you have a rulebook, edition or example that would help, include it too.
You don’t need to settle the question first. We’ll check the explanation and whether a different set of house rules accounts for the difference.
Send us a note →