Damage control beats hero hands
The 2012 world champion plays small on purpose.
Recreational players chase the biggest hand on the card because it pays the most. Competitive players do the opposite. A hand worth very little that you actually complete, over and over, outscores a spectacular hand you almost never finish. This is not caution — it is arithmetic.
Duan Yanbin won the individual title at the 3rd World Mahjong Championship (Chongqing, 2012) after years of being labelled the "eternal runner-up." Asked about his approach, his answer was disarmingly plain: keep the hand small, because at a table of experts the big hand simply does not come in often enough to justify the risk.
个人喜欢尽量把牌做小，少做或不做大牌，高手较量中大牌和牌的几率太小。
I personally prefer to keep hands as small as possible, and rarely or never chase the big ones — when experts compete, the odds of actually completing a big hand are far too low.
At an American table
Every NMJL card hand has a printed point value, and the tempting ones are almost always the awkward ones. Before you commit, estimate probability × value, not value alone. A 25-point hand you complete twice beats a 75-point hand you complete never. When your Charleston leaves you with a scattered rack, the disciplined move is to aim at a modest line you can actually finish — not to salvage a dream.
To join the top 1%, oppose the majority
Why a career professional attacks — and defends — against the grain.
Every strategy book preaches aggression, so most players become aggressive. That consensus is exactly where the edge hides: if the whole table is attacking, the profitable position is defence, and the reverse. Takahashi Ooi (RMU representative and captain of the M-League team Shibuya ABEMAS) puts this bluntly in his own book — reaching the very top means being willing to stand against the majority.
Notice the second half of his point, though. He is famous for "enduring" at the table precisely because everyone else is pushing. Contrarian does not mean reckless; it means reading what the room is doing and taking the other side of it.
如果想成为最顶尖的那 1% 的人群，那么就要成为多数派的对立面。如果大家都在防守的话那我就进攻。
If you want to be among the top 1%, you have to stand in opposition to the majority. If everyone is defending, then I attack.
At an American table
Read your room before choosing a style. At a table where everyone grabs every exposed tile and races for the flashy hand, the winning adjustment is patience: hold safe tiles, let them knock themselves out, and take the hands they hand you. At a timid table where nobody commits, the edge flips — be the one who actually finishes something.
"Fastest" is a trade-off, not a speed limit
Build the best hand you have room for — then cut your target to beat them to it.
Takahashi's nickname is "fastest and strongest," and he has explained what he actually means by it. It is not a licence to rush. It is a two-step test: when you have room, build the strongest hand available; but if someone is likely to finish before you, downgrade your target and take the win first.
That is the real skill — not speed for its own sake, but estimating how much time you have and resizing your ambition to fit it. Players who cannot do this either overreach and get beaten to the punch, or underreach and win small pots they could have won large.
我所说的最速是指在一局牌中，在有余裕的条件下去做最强的手牌……如果有人会在自己之前和牌的话，那就改以 5200 之类的牌为目标尽可能抢先和牌。
What I mean by "fastest" is: within a single hand, build the strongest hand you have room for … but if someone looks likely to win before you, then downgrade your target and win first.
At an American table
American mahjong has no riichi declaration, so you read speed from what is exposed: how many tiles a player has picked up, whether their exposed groups cluster around one card line, how close their rack looks. If a neighbour is visibly one tile from winning, abandon the ambitious line and switch to a smaller hand you can complete in the next few picks. Winning something now beats losing everything later.
Own the variance: luck is real, and that is the point
A world champion publicly credits luck for half the result. He is not being modest.
Immediately after winning the world title, Duan Yanbin said the game is roughly half luck and that he had simply run well. Coming from the man who just won, that sounds like humility. It is actually the clearest possible statement of how advanced players think about results.
If luck is a large share of any single session, then a single session tells you almost nothing. What you can control — and therefore what you should judge yourself on — is the quality of your decisions. Judge the process over hundreds of hands, not tonight's score.
打麻将还是很靠运气，大概占了 50%，我这次就是运气不错。
Mahjong still depends a great deal on luck — maybe 50%. This time, my luck was simply good.
At an American table
Stop grading yourself on whether you won. Grade yourself on specific decisions you can replay: did you pass sensibly in the Charleston, did you abandon a dead hand early enough, did you take the hand the card actually offered instead of the one you wanted? Those are the only levers you own.
Rule mastery is not beginner work — it is the whole floor
The champion who bought ten rulebooks; the foreigners who won by knowing one rule set cold.
Duan Yanbin did not touch a tile until he was twenty. When China published its standard competition rules in 1998 he bought ten copies and studied every scoring pattern until he could recite them. He has also pointed out something uncomfortable: at international events, Chinese players who grew up on local house rules were at a disadvantage against foreign players who had only ever played the official standard — because the visitors knew that one rule set cold.
The lesson generalises past national rules. At expert level, nobody loses because they lack a clever trick. They lose because they did not know what the card, the season, or the tournament floor actually required.
外国选手因为从来都只按照这套规则打牌，反而更占优势。
Foreign players, because they have only ever played by this one rule set, are actually at an advantage.
At an American table
The current official NMJL card is the final authority on every rule, and it changes every year. The single highest-value habit for an American player is to read the new card completely before the first game of the season — not to skim it. Half of all "bad luck" at a table is a player aiming at a hand that is not on this year's card, or has different requirements than they remember.
Folding is not cowardice
When the expected value is negative, the brave move is to stop.
Beginners treat "I'm on a wait" as permission to stay in. Advanced players treat a wait as a number to be compared against a risk. If you are two tiles from winning while an opponent is already there, pushing on is not courage — it is paying for the privilege of losing more.
Related and equally common: a good shape beats an early wait. Holding a poor wait (a single-tile gap or an edge wait) just because you got there first is usually worse than stepping back one stage to build a wide two-way wait that will actually convert.
Figures circulating in mahjong instructional material (for example, that a four-tile wait converts far more often early than late) are useful rules of thumb, not official statistics. Treat them as directional.
只要不是只差一张牌听牌，且手里没有两张绝对安全牌，直接弃和。这叫"止损"，不叫怂。
Unless you are one tile from ready and hold no two absolutely safe tiles, fold outright. Call it cutting your losses — not cowardice.
At an American table
American mahjong punishes this less harshly than riichi, because a dead hand is your own loss rather than a payment to someone else — but the logic is identical. Once your hand is dead, stop building it and start playing defence. Every further tile you commit to a hand that cannot win is a tile you cannot use to keep someone else from winning.
Steal other people's thinking
There is no single correct answer — and that is the opportunity.
Asked repeatedly to solve "what would you discard" puzzles, Takahashi kept returning to the same meta point: these puzzles have many defensible answers, and the value is not in finding the one right move but in collecting how different strong players reason — then digesting it into your own judgement.
Advanced players are magpies. They read the player they disagree with most, because that is where the biggest gap in their own model lives.
何切る问题有很多答案就好……听很多人的思考，把它变成自己的养分，这很重要。哪怕是我这种怪人的思考。
It is good for a "what do you discard" problem to have many answers … listening to how many people think, and turning that into your own nourishment, is what matters — even the thinking of an oddball like me.
At an American table
Ask the strongest player at your table why they passed what they passed, or why they abandoned a hand you would have kept. Do not argue for your line — collect theirs. Over a season, the players who improve fastest are the ones who treat every game as a chance to download someone else's model.
The work happens away from the table
Review, one fix at a time, plus limits you actually honour.
The gap between a decent player and a strong one is rarely another clever tactic. It is review. Strong players replay their worst decisions, write them down, and fix one recurring mistake per week rather than vaguely resolving to "play better."
The off-table habits matter just as much: stopping when tired, taking breaks so judgement does not decay over a long session, and keeping the game inside a budget that never threatens anything you care about. Tilt — playing on after a bad beat — costs more points than any technical flaw.
每周至少复盘一次自己的牌局，记录关键决策点，这是高手成长的必经之路。
Review your own games at least once a week and record the key decision points — this is the only road to getting genuinely good.
At an American table
Keep a simple log: the card year, the hand you aimed at, the moment it died, and the single decision you would change. One fix per week — "I stop passing joker-adjacent tiles to the right," "I abandon a hand as soon as two of its groups are unreachable" — compounds faster than any amount of extra playing.
Where these ideas come from
Every attributed quote below is traceable to public reporting or the player's own published work.
Duan Yanbin (段燕斌) — individual champion, 3rd World Mahjong Championship, Chongqing 2012; reporting by CCTV and China News Service (2012-10-31) and a China News Service profile (2015-11-29) covering his "small hands" preference, his comments on luck, and his 1998 self-study of the official rules.
Takahashi Ooi (多井隆晴) — RMU representative, captain of M-League team Shibuya ABEMAS. Quotations are from his own book 《多井热》 (2018, ch. 38, on the meaning of "fastest and strongest"), his serialized "What would you discard?" columns for HeartLand (2015 and 2019), and his instructional work on defence.
General competitive principles (sections 06 and 08) are not attributed to any individual. They are common ground in mahjong instructional material, restated here and adapted for American play. Any percentages quoted in that material are rules of thumb, not official statistics.
Note: where a widely circulated quotation could not be traced to a verifiable source, it is deliberately not attributed to any named player. Unattributed "table wisdom" on this site is editorial, not reported speech.