Suomeksi

KJMJM

Majiang

Rules

Strategy

Miscellanea

Calendar

Board

History

Symbolism

Basics

Equipment

Offense

Defense

Links

Credits

Rules

History

Divination

Scoring

Variants

Folklore

Jargon

Misc. misc.

Ranking

Other rules

Offensive Majiang Strategy On this page, some basic majiang strategies for offense are given. After reading this, you should definitely take a close look at the other half of the strategy hints section which concentrates on defensive strategies. These two aspects of majiang strategy shouldn't be considered very separate (despite the way these pages, too, have been organized) but recognized as to sides of the same thing, and be allowed mingle into one, balanced whole. It is also worth keeping in mind that the hints given on this page are by no means a comprehensive analysis of anything, nor do they provide a sure path to becoming a majiang expert. True skill is only achieved only through experience and creativity - that is, through playing, and what could be more fun that that? It is imperative to note that the majiang rules system profoundly affects the viability of various strategies. These hints have been written with the Japanese-style rules used by Kejimajiangmi in mind, and are only in part generalizable to other rules systems. However, since these hints are pretty basic, they embody many of the fundamental principles that are common to most majiang systems. It might be useful, for the beginner, to actively keep in mind the most important points (in bold face) while playing - they are easy to forget in the excitement of a game. And of course, it should be remembered that these hints are just guidelines. The wall provides endless surprises to which every good player must be able to adapt. The Key to Winning The first thing about majiang strategy that one should understand and remember is that the player who goes out first wins - not the one who would have had the most valuable hand. Every dealt hand will definitely not turn into a hand worth three or more doubles, let alone that one would be able to complete such a hand before someone else goes out. A bird in the hand (no, not bamboo-1, silly) is better than three in the bush. Elements and Becoming Ready A good majiang player is able to maximize his possibilities for drawing useful tiles from the wall. This is accomplished by carefully choosing which tiles and tile combinations, elements, to discard, which in turn requires an ability to discern and appraise the elements that one has. Simple suit tiles are the most useful as they are the most versatile for forming new sets, whereas honor tiles are irritatingly inflexible; they can only be used for making triplets (and fours). Terminals are generally a bit better than honors since they can be used in sequences, albeit less easily than simple tiles. For some reason, two identical tiles often seem to the novice player an extremely good element. This is, however, a misconception: to be completed into a set of three, a pair requires a third identical tile, i.e. one of the only two copies still available in the game. For comparison, a sequential element of two suit tiles with consecutive number values becomes a full set with two different tiles, with all four copies of each remaining in the game. This is a fourfold difference in the probability of completing the set with a drawn tile, and it is reasonable to say that the sequence is the most imporant type of set. A pair is, however, a unique type of element: each winning hand has to contain a pair, so the only pair in one's hand should not be discarded or melded as a triplet without good justification! A pair is valuable as it is. Many beginners tend to 'pung' anything that moves. This is hardly ever to be recommended (unless a four-triplet hand seems very likely), as melding triplets (especially triplets of simple tiles) is one of the surest ways of destroying the elements forming in one's hand and causing most of the tiles that will be drawn from the wall to become useless. It is usually an extremely bad move to meld a triplet out of a good element, say, 5-5-6, from which a now useless part (the six) has then to be discarded. A large number of tiles drawn from the wall can make a 5-5-6-element into something better (possibly a set), not to mention that the just-discarded five guarantees that one of the fives in one's own hand is safe, and after discarding it one would still be left with a good 5-6-element. However, melding any set may be a good idea if by melding one becomes ready to go out, i.e. one (good) tile away from going out. With respect to winning the deal, the most crucial element is the one that is completed (turned into a full set or pair) last, i.e. the one that is the only incomplete element after one becomes ready. In addition to becoming ready quickly, it is essential to become ready on a large number of (different) potential winning tiles. Through collecting simple suit tiles in a concealed hand, one can almost always become ready on at least two different tiles, sometimes more. A bad ready hand is one where only one tile can be used to complete the hand: the last element is either a terminal pair (e.g. 1-2) waiting for a single third tile, a separated sequential pair (e.g. 5-7), or an isolated tile waiting to be paired with an identical tile (this last case is generally brought about by carefree melding of sets). Often, players are ready on two suit tiles separated by three numbers (e.g. with a 5-6-element waiting for a 4 or a 7 from the same suit) or on two pairs one of which has to be completed into a triplet. The most typical elements that can be used to become ready on more than two tiles are the run of five (e.g. 3-4-5-6-7) which transforms into two sequences with three different tiles, and the concealed triplet plus adjacent tile -combo (e.g. 5-5-5-6) which transforms into a set and a pair also with three different tiles. It is almost impossible to become ready on more than three tiles without having at least one concealed triplet in one's hand. During the game one should of course also pay attention to the tiles on the table and their effects on the possibilities for developing one's elements into better elements and full sets. For instance, if one copy of a tile has already been discarded right at the beginning of the deal, it is considerably less likely for a pair of that tile in one's hand to be completed into a triplet. And rather less likely indeed if two copies of that tile have already been discarded or have shown up in other melded sets. Similarly, an otherwise good 2-3-element is not that great after all if someone has, say, melded a four of the ones in that suit. Hand Versatility In order to improve one's hand, it is essential that at each stage of the deal, as many obtainable tiles as possible can be used either to bring it one tile closer to completion, ultimately to become ready and then go out, or at least to increase the likelihood of drawing a useful tile during the next few turns. Developing the hand is a process that lasts through the whole deal, with the elements in the hand shifting and, with the player's assistance, form better and better combinations. To enable this development, one must take care to keep the hand as versatile as possible. This means that one should not, at an early stage, decide upon which tiles will be used for which element - not even if some three tiles already form a full set. Let the tiles find their own places. When composing even medium-sized elements out of simple suit tiles, the role of any given tile can change completely as new useful tiles are drawn from the wall. There are often tiles which are somehow connected with an element, but seem useless, such as the seven in a 4-5-7-combination. These tiles should not be dismissed outright as worthless as they may prove quite valuable if the wall provides a suitable tile. For instance, when a six is drawn to the 4-5-7, a 4-5-6-7-element is created, which easily forms into two sets with two tiles or into one set and a pair with one tile. In general, it's advisable to collect tiles that are numbered close to each other in one suit even if they don't put you closer to winning right away. If there are several otherwise equal options, one should often discard tiles from the 'shortest' suit (i.e. the suit of which one has the least tiles in hand), though generally defensive considerations outweigh this one. As noted above, it's important to recognize good elements as they form in one's hand. Developing the hand is not, however, a passive observatory process but an active preparation for the future which requires the player not only to recognize elements but to understand the probabilities for creating new elements and to choose his tactics so that the likelihood of being able to form good elements is as high as possible. The Wall is Your Friend   (usually) Melding sets is usually a bad idea. This may sound completely ludicrous to the novice player, who have not yet observed that by keeping one's hand concealed and only claiming tiles when the are really needed (say, to get a single tile necessary to become ready), the hand usually develops faster than if it is torn to pieces by melding. Both sequences and triplets break up your hand very efficiently, destroying its options for development. The elements of a concealed hand may be modified as new tiles are drawn, which makes it possible to achieve a high number of potential victory tiles, to increase the scoring value of the hand, and to play a much better defensive game. Further, a concealed hand makes it considerably more difficult for other players to defend well and guess as to the value of your hand. Of course there are situations where claiming tiles is worth it. For example, lucky winds and 'dragons' are often worth claiming simply because of the valuable doubles they give. Other triplets and sequences can certainly be useful in specific situations as well, if the hand has reached a point where the number of useful tiles is small. However, aggressively melding everything will lead to disaster sooner or later. Try it yourself: make up your mind not to claim any tiles at all for a few deals, and you will probably find that your hand develops unexpectedly well despite not melding. Claiming a tile for a four is a particularly often a bad idea.. Only if you already have a melded set (no concealed hand), have safe discards in your hand, and stand to gain an extra turn by melding the four (the discarder is not the player on your left), should you consider claiming a tile to make a four. It's almost never worth it just for the point value (though a 32-point concealed four is certainly good). Melding fours of simple suit tiles is particularly prone to breaking up your hand if done carelessly. Hand Value Winning hands are not equal. Quite the opposite: a valuable hand may earn dozens of times more than a cheap one. It is therefore self-evident that any majiang player should take care to remember the scoring rules when choosing his tactics. This certainly does not mean that one should always try to win with an extremely valuable hand, but that improving the hand and taking risks must be done prudently, analyzing one's situation. Becoming ready quickly and with many potential winning tiles should usually be the main priorities when developing one's hand, but realistic opportunities for higher scores must be recognized when they exist, and they must be made use of in order to win in the long run. Victories in majiang come as a combination of both low- and high-scoring hands, through knowing when to aim for which. Putting one extreme before the other will cause defeats in the end. Anyone who's played just a little bit, or spent even a couple of moments with the scoring rules, will notice that the easiest way of obtaining a high score is not collecting points but doubles. For instance, the points from non-lucky triplets have very little positive effect on the value of the hand, and even often ruin the 'no points' double. Along with the obvious doubles from 'dragon' and wind triplets, the most common (i.e. easiest to get) doubles are 'no points', 'all simples', and perhaps - depending on one's playing style - 'concealed hand'. These doubles form with remarkably little effort in the hands of players collecting concealed simple suit tiles, and often even a player who's ready to go out is able to alter his (concealed) hand so that it receives one or more of the common doubles. One could say that practically all other doubles and valuable hands are originate from some exceptional circumstances. There are types of dealt hands with inherent potential for high-scoring hands. With a dealt hand with five or more pairs it's a good idea to keep in mind the all-triplets double. However, aiming for this high-risk, low-reward double generally means that you must meld several sets, which is very risky if you don't win quickly (which you probably won't). In addition, since there are only two 'free' tiles of each type to add to your elements, four-triplet hands are hard to complete, especially if the pairs are formed of (more rarely discarded) simple suit tiles. The double for three concealed triplets (one sequence allowed) is also good to remember even though it's a rarer double that requires reasonable luck). If the initial hand contains one or more good elements of honor tiles and lots of (simple) tiles from a particular suit, it may be worth trying for the 'one suit with honors' double (which, like the all-triplets hand, is, however, rather hard to complete compared to the rewards). Discard the tiles from the other suits, but if you happen to accidentally form a set or a good element in another suit it's probably better to forget about the extra double, and keep the good element or set. If the initial hand contains very many (at least nine, preferably at least ten) tiles from one suit, you should definitely consider going for a pure one-suit hand (three doubles), which is one of the easiest extremely valuable hands. If the wall favors you with good tiles, a one-suit hand will form conveniently into a concealed hand, ready with at least two tiles. When aiming for just one suit, you should be patient and wait for what the wall has to offer, and not rush into melding all tiles of that suit discarded by others. This is essential both to keep your elements flexible and to keep your opponents in the dark. However, in the later parts of the deal, it might be necessary to meld a hard-to-complete set when opportunity arises. It is important, but - when the frenzy of gaming hits you - difficult, to remember that less valuable hands are also good. The most important thing is to be able to win something. Don't try for too fancy winning hands. Unless your hand is exceptionally well-suited (duh) for it, it is very unlikely you'll be able to get a hand of more than a hundred points before the other players. Not to mention that you'll probably end up losing to someone else while eagerly trying to complete your unlikely hand. The Changing Role of Offense Some players (and strategy hints) perceive a deal as consisting of two or more 'phases' where developing better elements, melding, and defending are, in turn, given the emphasis. This kind of division may make it easier to select ones tactics, but as one becomes a better player, it only restricts one's thinking. Ideally, a player would be able to see, at each point of the deal, the game situation from the viewpoint which is real at that very moment, and not to consider himself as being in a specific 'phase' where particular types of strategy should be used. The tiles that are drawn and discarded create an environment which is constantly shifting for the better and for the worse. A good player constantly re-evaluates his situation and acts accordingly. As the deal goes on, it does of course have an effect on what kinds of decisions are generally better. At the start of the deal, a hand usually consists of more or less isolated tiles and elements, and most of the tiles one draws are somehow useful as they easily form elements with the scattered tiles already in one's hand. This natural tendency should be encouraged by carefully judging which elements and tiles are the least useful and discarding them, and by declining to claim tiles. Usually only after the hand is close to going out is the time right for more aggressive playing. So, from the beginning, one must be able to see things from the perspective of the whole deal. When the hand is close to being finished, it is possible to recognize those few elements, which have the strongest potential for becoming sets in the complete hand, and those which are the most difficult to improve. It is these difficult elements (in particular, elements that only become sets with one specific type of tile) that one should consider claiming a tile to. When the other players also close in on winning, time runs fast and it may be necassary to get that rare useful tile through melding. With good, flexible elements, however, one should wait right until the very end of the deal. In majiang, some of the most problematic decisions arise in situations, where discarding a certain tile would take the hand one tile closer to victory (than it was before drawing the latest tile) but would not provide good chances for further developing the hand, whereas discarding another tile would not take the hand closer to victory as such (or even would take it one tile away from winning) but would improve the elements in such a way that good tiles could easily be drawn later. A simple example of this is a player, who's wondering whether to discard an isolated number five in one suit or a bad 1-2-element in another suit. A five is, as such, not as good as a two-tile element, as no single tile will make a set out of it. On the other hand, however, it has more useful neighbors than the 1-2-element. As a rule of thumb, one should initially aim far away and emphasize the future potential for development of the elements, and a bit later start concentrating on moving towards victory one step at a time. As the deal progresses ever further, and other players are close to winning, defensive considerations begin to outweigh offense. This does not mean that it is always impossible to consider both at the same time. Observing One's Opponents Majiang is not 'solitaire for four', where each player works by himself to build the necessary sets. Observing other players and analyzing their game plays a particularly important role in defensive majiang, but it has its uses when drawing up one's plans for winning as well. By reading other players' discards one can draw conclusions as to what kinds of tiles they don't need. For instance, if a player discards a lot of tiles in a certain suit early in the deal, it is not likely that tiles close to the discarded ones will be of use to the discarder. Therefore, it's sensible to try and become ready on tiles from that suit, if possible. One must also pay attention to the 'safety' of the tiles one hopes to get: if the tiles you're waiting for haven't been discarded at all, it is rather unlikely, especially towards the end of the deal, that those tiles will be discarded by other players (this depends heavily on the skill level and playing style of our opponents). If there are options to choose from, try to become ready on a tile which would more likely seem like safe discard. Of course, it is always possible to draw the winning tile from the wall, which is an important factor especially when playing against cautious opponents. The '1-4-7-rule' that is generally applied in defensive play can also be used conversely to intentionally become ready on tiles which are removed by three numbers from recently discarded ones. A hand which is ready on a single tile waiting to be paired was deemed unlikely to succeed above. Its only good side is its possible unpredictability: it's easy to change the single tile to another, perhaps even one that has been just discarded (esp. an honor tile that's already been discarded) so it seems like a safe discard to the others. Other players will be trying to read your game, so you should try to delude them as much as possible. To this goal, you can use a variety of means, which are largely player-dependent. The obvious rule of thumb is that you should get the other player to believe you're holding something completely different from what you actually have. The most common means to accomplish this bluffing are using your discards, and displays of interest, disinterest or puzzlement expressed through body language as tiles get discarded. The deductions others may make about your playing should be remembered even when they're wrong. For example, if you haven't, by chance, happened to discard tiles of one suit at all, the others will probably consider that suit dangerous, and may even think you're going for a pure one-suit hand. It might therefore be prudent to try and become ready on a tile which is not in this suit, or to somehow alleviate the others' fears through discards or melds. Miscellaneous Tips At the start of the deal, you should generally get rid of single honor tiles, as an isolated honor is the weakest of all elements. This is especially true of those honors which won't give you a double even if you managed to get three of them. Discard other players' winds first, followed by isolated dragons. It's not really worth it to keep waiting until you manage to pair up honor tiles by drawing from the wall; this rarely happens. Instead, you might find yourself having to either make a very dangerous discard later or to give up thoughts of winning the deal. With a lousy dealt hand, which has many different terminal and honor tiles, one might be tempted to go for the 'Thirteen Orphans' special hand. The attempt is, however, almost doomed to failure unless you have at least nine (preferably at least ten) orphans' in the dealt hand. Don't divide the tiles in your hand into groups or arrange them in numerical order. At least, avoid arranging them in the same way all the time, as observant opponents might be able to figure out your system and gain some very important information about your hand. This does of course depend on the skill level of one's opponents. If you do wish to arrange your tiles somehow (which can, on the other hand, be of advantage to a beginner, since it helps to learn to recognize some more complex tile elements), try to do it in a sort of 'semi-random' way, e.g. by having sequences running in different directions, and not having all the tiles in one suit right next to each other. There are situations where it's a good idea to discard a tile from a triplet, and use the remaining two as a pair. This can lead to a situation where you have more potential winning tiles when you're ready, and it may also increase your score. As a special case, note that the double for 'terminal()s and/or honor(s) in each set' is disproportionately difficult to get compared to the measly award you receive for it. Unless the hand quite naturally becomes like this, it's not worth aiming for.

Kejimajiangmi

Feedback