Common Mahjong Mistakes: 10 Errors That Cost You Hands

Mistakes Are Cheaper to Fix Than to Learn

Improving at mahjong is usually taught as accumulating knowledge — more scoring rules, more patterns, more variants. In practice the fastest improvement comes from removing errors, because most beginners lose consistently to the same short list of mistakes rather than to a lack of exotic knowledge. See also Mahjong Strategy: How to Actually Win More Hands.

The reason is compounding. A single wrong early discard costs you a tile of efficiency. The same wrong instinct applied five times in a row costs you the hand. None of the individual decisions looks decisive while it is being made, which is exactly why the errors are so persistent. See also Mahjong Defense: How to Stop Dealing Into Winning Hands.

This guide collects the ten that do the most damage, ordered roughly by how many hands they cost. Each one gets the mistake, why it seems reasonable while you are making it, and the concrete alternative.

A note on how to use this list. Do not try to fix all ten at once; that produces a player who is thinking about ten things and playing slowly. Pick the one or two that describe your last few games, correct those until they are automatic, then move down the list.

It is also worth saying what this list is not. None of these mistakes are about luck, and none of them are about knowing an obscure rule. They are all about decisions you make on ordinary tiles in ordinary hands, repeatedly. That is precisely why they are worth fixing: the payoff is frequency, not cleverness.

Mistakes 1–3: Calling and Keeping

Calling too often is the single most common beginner error. Calling feels productive because it visibly builds your hand, but every call opens your hand, forfeits concealed bonuses in most rulesets, and tells the table what you are collecting. The tile you call must genuinely accelerate the hand, and often it does not — it completes a set you could have finished by drawing while keeping your options open.

Keeping too many pairs is the second. Pairs are good, so beginners keep all of them, and a hand with five pairs has no room for the sequences a completed hand needs. Pairs are worth keeping when they have a path to a triplet or a clear role as the finishing pair. Beyond about two or three, extra pairs are crowding out the blocks that would actually finish the hand.

Keeping isolated honours too long is the third, and it is the mirror image of over-valuing tiles that look important. A lone honour is the most disposable tile in the game — no neighbours, no sequence, needing two specific copies that may be nowhere. Beginners hold them hoping for a triplet and discard useful middle tiles instead. The correct order is honours first, then isolated terminals, then middle tiles.

These three are grouped because they share a root cause: each mistakes visible progress for real progress. A call, a kept pair, a held honour — all three make the hand look more developed than it is. The correction in each case is to ask what the tile or the call is actually contributing, rather than whether it feels like progress.

A concrete illustration of the calling error. You hold a partial hand and someone discards the third tile of a sequence you were building toward. Calling completes that set immediately and feels like a clear gain. But the set you completed was going to finish by drawing anyway within a few turns, and by calling you have opened your hand, forfeited the concealed bonus, and shown the table exactly which suit you are in. You traded a small time saving for a large amount of information and a scoring penalty.

The pair error has a similar shape. Two pairs in a starting hand are good — they give you flexibility and a path to the finishing pair. Five pairs is not five times as good; it is a hand with no sequences and no room to develop one. The moment to stop collecting pairs is the moment you no longer have space for four sets alongside them.

Mistakes 4–6: Value and Speed

Chasing a big hand from the start is the fourth mistake. A player decides they want a flush, or an all-triplet hand, and then bends every later decision toward it. The result is a hand that would have completed in a normal shape but never does, because the constraint was imposed before the tiles justified it. Value should be noticed when it arrives, not ordered in advance.

Playing only for speed is the fifth, and it is the opposite error — rarer among beginners but damaging among players who have just learned efficiency. Always building the fastest hand means always collecting the smallest payment and never exploiting the hands where value was genuinely available. Speed is the default, not the only option.

Ignoring the minimum score threshold is the sixth, and it is technical but costly. Several rulesets require a hand to be worth a minimum before it can be declared a win at all. A beginner who completes a hand with no scoring pattern at all discovers, too late, that it cannot be declared. Knowing what your ruleset requires before you start building is not optional information.

The thread through these three is that value and speed are both legitimate, and the error is treating either as a rule rather than a judgement. The player who always chases value and the player who always chases speed are making the same mistake in opposite directions — fixing the decision before seeing the hand.

The minimum score rule deserves one more sentence because it is the only mistake on this list that can make a completed hand worthless. In rulesets with a threshold, a hand that finishes but scores below the minimum cannot be declared at all. That means the hand is not merely low value — it is not a win. Checking the threshold before committing to a shape is therefore not bookkeeping; it is the difference between a hand that counts and one that does not.

二万 · 2 Crak三万 · 3 Crak四万 · 4 Crak五万 · 5 Crak六万 · 6 Crak七万 · 7 Crak

Mistakes 7–8: Reading the Table

Not watching discards is the seventh and arguably the most expensive on the list. Beginners look at their own hand and the tile just drawn, and treat the rest of the table as decoration. But the discards are the only reliable public information in the game, and a player who never reads them is playing with a fraction of the available data.

The specific failure is most costly in defence. A player who has not been tracking the threatening opponent repeatedly discards into a hand that was visibly nearly complete, then feels unlucky. The information was on the table for several turns; it simply was not read. A minimal version of the habit — track one committed opponent and know which tiles they have discarded — prevents most of these outcomes.

Misreading your own distance to a win is the eighth. Beginners judge closeness by how good the hand looks rather than by how many tiles it needs. A tidy-looking hand with a closed gap wait can be further from completion than a messy-looking one with a two-sided block, because tile efficiency is about how many future tiles you accept, not about how organised the tiles appear in your rack.

The practical substitute for judgement is counting: how many blocks do I hold, how many are complete, and how many tiles would improve each incomplete one. That gives a number rather than an impression, and the number is almost always more accurate than the feeling.

Both mistakes in this pair come from looking inward. The fix in each case is to redirect attention outward — to the table’s discards for the first, and to an objective count for the second. Neither requires new knowledge, only a change in what you are watching.

A quick test tells you which of the two you suffer from more. After your next hand, try to recall the last three tiles each opponent discarded. If you cannot, the discard-reading habit is missing. Then try to state how many tiles your hand needed at the midpoint. If you cannot, the distance-counting habit is missing. Most players fail one of these two cleanly, and that tells them where to start.

Mistakes 9–10: Tempo and Discipline

Rushing the turn is the ninth. Under time pressure, players discard the first tile that looks unhelpful rather than comparing their options. Mahjong rewards comparison — the difference between two candidate discards is often measurable, and a player who never pauses will never see it. Slow, deliberate play is not indecision; it is where the skill actually gets applied.

The tenth is the most abstract and the most costly: failing to accept that some hands are not winnable. A player who insists on finishing every hand they start will push past the point where folding was correct, and the accumulated deal-ins will outweigh the occasional extra win. Recognising a dead hand and steering it safely to the end is a skill, not a defeat.

These two are related. Rushing produces shallow decisions; refusing to fold produces bad ones. Both come from treating every hand as though it must be won.

The discipline that fixes both is the same: before deciding, ask what the current hand is actually worth finishing. Some hands deserve aggressive play. Many do not, and identifying which is which is most of what separates a steady player from a volatile one.

It is worth adding a note on how these two feel from the inside, because they are the mistakes most resistant to correction. Rushing feels efficient, and folding feels like losing, so both punish the player who corrects them with an immediate sensation of having done something wrong. The reward for slowing down and for folding properly arrives later and in aggregate, which is why the fix requires trusting the numbers rather than the feeling.

Fixing Them in Order

The list is long, so here is the order that pays fastest. Stop calling reflexively. That single change improves both your efficiency and the information you leak, and it takes almost no new knowledge to apply.

Then fix your discard order — honours, then isolated terminals, then middle tiles. This is mechanical and immediately reduces the number of dead tiles you carry into the middle game.

Then start tracking one opponent. Not all three. One player, the one who has shown commitment, and their discard pile. This converts defence from a theory into a habit you can actually run.

Then add the distance check: before discarding, count how many tiles away your hand is. If you cannot answer that roughly, you are choosing blind.

The remaining mistakes — chasing value, ignoring minimum score, insisting on finishing every hand — resolve largely on their own once those four habits are in place, because a player who measures distance and reads the table naturally stops over-committing.

None of these corrections requires memorising a scoring table or learning a new variant. They are changes to what you pay attention to, which is why they show up in results within days rather than months.

One last point about how to judge whether the fixes are working. Do not measure by whether you win more hands; that number is too noisy over a short run to tell you anything. Measure by whether you can describe, for your last few hands, what your distance was and which opponent you were watching. If you can answer those two questions about your own play, the corrections have taken. If you cannot, the mistakes are still running unobserved.