This article was written by Alan Kwan explaining the design rationale behind the Zung Jung scoring system. Because several older Zung Jung resources are now difficult to access outside archived webpages, this republication preserves the historical and theoretical basis of the system for modern players and organizers.

• original author: Alan Kwan
• Justifications for Zung Jung System v2.3
• Wayback Machine capture link
• Original spelling/formatting preserved
• Editorial note: When Alan Kwan refers to “Chinese New Style,” modern readers can broadly understand this as the rules family that later evolved into systems such as MCR (Mahjong Competition Rules/Chinese Official Mahjong).

Why the Zung Jung System?

The game of Mah-jong has evolved a lot since its creation. The strategies and concepts in some modern versions are so different from the Classical game that they are virtually different games.

When we look at the various modern systems, we can see a common direction in the evolvement of the game, across different places and cultures. Most modern versions are very much centered on patterns in their scoring. The Classic game is centered on triplets and points, but in modern versions, point counting has either been dropped (as in Modern Chinese Mahjong, i.e. Old Style and New Style), or has lost most of its significance (as in Modern Japanese Mahjong). It seems that triplet-point counting, though contributing to a very well-balanced scoring system, is too complicated a task for the masses. Today, the most widely played version of Mah-jong is Hong Kong (Cantonese) Old Style with a stringent minimum faan requirement, which is a very pattern-centered game, despite the scarcity of the patterns that are there. What conclusion can we draw from this? Mine is that patterns have their magic on modern mah-jong players, and that patterns are the central element which constitutes the playability of (most versions of) Modern Mah-jong. Triplet-point counting (in the Classical game) gave us a deep game, but unfortunately not a simple one, and patterns look like a hopeful candidate which can take its place of providing depth to the game, while also giving us a game that is easy to learn and to get into.

For this reason, the Zung Jung System has been designed as a very pattern-centered version. The selection of the set of patterns is loosely based on Modern Japanese; this gives us a large enough set to play and have fun with, without being overwhelming. Aside from that, I have studied closely the effects of various distinctive elements in various modern systems, and boldly throw out any elements that I find to be distracting from pattern-centered game play. As a result, there is no emphasis on the “Concealed” hand (as in Modern Japanese), nor on self-draw (as in Modern Chinese), because I find these elements to be distracting from my intended emphasis of the game. I’ll go more into these points later.

Why an additive scoring system?

The faan-doubling system has its origin in Classical Chinese mahjong, in which hands of more than 3 faan are rare. The system has originally been designed with small number of faan in mind. With the increase of the faan values of patterns in Old Style, the system is thrown out of balance, and this has brought about the artificial and complex laak system. I feel that as we evolve to a pattern-centered style, there is no longer any reason to retain the faan-doubling system, which has been designed for the triplet-point game, and becomes broken as we use more patterns and increase their values. An additive system is far simpler and neater.

The fact that the exponential faan-doubling system is not very appropriate in the pattern-centered context is made apparent by the upper limit issue. I have never been impressed by the artificiality and the complexity of the “laak” system used in Old/New Style (or a variant of it in Japanese). Its adoption plainly tells the truth that faan-doubling doesn’t work when we are getting a large number of faan. Also, there is a problem of continuity (consistency). For example, suppose that 6 faan is 1 laak, and 9 faan is 2 laaks. This means that, in a hand, the 6th faan is very valuable, the 7th and the 8th faan are pretty worthless, and the 9th faan is very very valuable. I feel that this discontinuity is unnatural, and distracts from the simple objective of building more, valuable patterns. For this reason, the Complementary System (the previous name of the Zung Jung System) v1.0 used a quadratic system, in which for large hands, the score for the hand is simply the square of the number of faan. However, that system is still unnecessarily complex, and it does not satisfactorily solve the problem detailed in the next paragraph, so in v2.0 the simple additive system is adopted.

There are other reasons too. Unlike Classical and Old Style, here there are many scoring patterns. Many patterns have a high rate of concurrence, in a sense. For example, most “3 Similar Sequences” or “9-tile straight” hands are “All Sequences”, and a large proportion of “Terminal in Each Set” hands also have “3 Similar Sequences”.

When a pattern occurs always with a certain other (for example, “7 Pairs” hands are always “Concealed”, and “Double of Two Identical Sequences” are always “All Sequences”), there is no problem: one determines the value of the former pattern taking this into account. However, when a pattern occurs often with another low-value pattern which is relatively easier (as in all the examples cited above), it becomes impossible to assign a value to the former pattern, under the traditional exponential system, which satisfactorily reflects the value of the hand both when it includes the latter pattern and when it does not. If a high value is assigned which adequately reflects the efforts in building the hand which does not include the latter pattern, the value of the hand which does becomes too high relative to others (including the very hand which does not!). That high value is not desirable, because the latter pattern concurs with the former pattern often, and is not that much extra effort above the former pattern. Let’s look at “3 Similar Sequences” for example. Does the hand deserve to have its value doubled or increased by 50% or so for a 4th sequence (“All Sequences” hand)? I don’t think so. Adding a mere 5 points, in the Zung Jung system, would more accurately reflect the efforts and difficulty involved in these hands and similar ones.

Why no ‘diagonal’ patterns and few 2-set patterns?

Players who are familiar with Chinese New Style may feel that there are too few patterns included in the Zung Jung System. In particular, there are patterns in the categories of Identical Sets, Similar Sets and Consecutive Sets, but no ‘diagonal’ patterns such as “9-tile straight in 3 suits”, and no 2-set patterns in the Similar and Consecutive categories such as “2 Similar Sequences”. Why not include these patterns?

One of the main design goals of the Zung Jung System is to keep it simple and easy to learn. Patterns are grouped in a few categories, so that they are not only easy to learn, but are also easy to deal with during play. I have observed that many players stay well away from New Style because of its complexity, and because playing it can be rather “tiresome”. Thus I exclude ‘diagonal’ patterns, which are not as easily recognizable as the others.

One problem which makes New Style difficult to learn, besides the numerous patterns, is that one has to figure out the ‘exclusions’ and ‘conflicts’. I am referring to combinations of patterns which cannot be both counted for scoring in the same hand because they share common elements in the hand. With a large set of patterns, exclusions and conflicts can become a very complex puzzle. The exclusions in the Zung Jung System are kept as simple as possible: patterns in the same series are mutually exclusive, and no others are. (This is actually the original concept in Chinese Classical mahjong.) I observe that, when I try to add diagonal patterns or 2-set Similar and Consecutive patterns, this simple rule ceases to suffice, and there will be a large increase in the complexity of the rules. I deem this undesirable for a game which is supposed to be played by more than a select few.

Why no minimum hand limit?

Every (or most) beginner learns Mah-jong first as a game in which one’s primary objective is to build a hand with 4 sets and a pair (a “Regular Hand”) to go out. I feel that there is some elegance in the simplicity of this rule. I have realized that most versions of minimum hand limits in fact prevent most possible hands (“Regular Hand” combinations out of the 136-tile set) from going out. In other words, I feel that minimum hand limits disrupt the simple, elegant objective of the game by sticking to it a not-so-simple condition.

The original Classical game does not have any minimum hand limit. The Classical game is a well-balanced system: high-value hands are ‘difficult’, in the sense of being infrequent. It is only a matter of course that in the Classical game, and its direct descendent the Old Style game, high-value hands are not completed often, and the majority of hands that go out in games played are fast, low-value hands. In a matter of time, players who long for the excitement of pursuing, and completing, high-value hands become frustrated and bored. This phenomenon is even more marked among Old style players, because the Old Style, without counting points, tends to ‘flatten’ hand values even more and gives rise to even fewer high- or medium-valued hands. (This is alleviated, but not adequately, by the practice of raising the values of the existing patterns.) As a result, players begin to employ minimum hand limits, sometimes stringent ones, in their games to try to recover their pleasure of completing high-valued hands. But this practice, as I have indicated, is disruptive to the playability of the game.

The Zung Jung System (following the New Style) attempts to solve this problem with an entirely different approach. By adopting many new high- and medium-value patterns, players can naturally try for, and complete, high- and medium-value hands more often, without being forced to struggle towards some unrealistic goal (and without being overwhelmed by the complexity of full-scale New Style). The Jung Zung System (following the New Style) tries to solve the problem at its root.

Then, one may ask, why not a minimum hand limit in conjunction with the Zung Jung System? One reason is that, a minimum limit is confusing to the beginner. One of the most frequently asked questions I’ve seen from beginners is the one concerning the minimum limit. The point is that a minimum limit makes it impossible for the beginner to “learn as one plays”; it forces the beginner to familiarize himself with the entire scoring system before he can play. This is especially noticeable in the case of computer mahjong games, where a beginner is often frustrated because he can’t understand why he is not allowed to go out. While, in the absence of a minimum limit, a player can play the game just by knowing the (more or less) universal rules of mah-jong; even if he doesn’t know anything about the scoring system at all, he can still play and he will gradually pick up the scoring system as he observes how the computer evaluates the score for the winning hand. In fact, in real games around a physical table, it is not uncommon that some of the players do not understand the scoring system being used, and rely on others to count the score for them. A minimum limit makes such playing much harder.

A minimum limit will bring about other playability problems, whether the limit is a high one or a low one. If we adopt a high minimum limit of say 50 points (roughly equivalent to the 3-faan limit common in Old Style), the hands which can possibly go out will be very restricted, since the 50-point patterns are by no means easily completed nor viably attempted every hand. A good number of dealt hands will become hopeless since the start. On the other hand, if we adopt a low minimum limit of say 5 or 10 points (comparable to the 1-faan limit in Modern Japanese), this will place an unnatural emphasis on the easy, low-value patterns such as “No Terminals” and “Scoring Honor Triplet”. These are supposed to be easy patterns with cheap rewards which are only that much significant, but with the minimum limit, they turn into your all-too-important express ticket to going out. Is “No Terminals” really such a noble attainment, that a hand which satisfies the pattern deserves to be allowed to go out, while the very similar hand which contains a 789 instead of the 678 is a crime which must be prohibited?

Instead of a ‘real’ minimum limit, here I try to use a ‘virtual’ minimum limit: all pattern scores are in multiples of 5, but a “Chicken Hand” scores only a token 1 point. The idea is to discourage, but not prohibit, hands which do not contain any patterns. This 5:1 ratio is a radical and bold approach; I would like to hear from players how it works out.

My practical experience is that, even though there is no minimum limit, “chicken hands” (1-point hands) do not occur too often. There may be one per 5 or 6 hands, on average. With a good list of patterns and (hopefully) balanced pattern values, a player cannot win (in the long term) if he passes up good opportunities for valuable hands and always goes for cheap hands.

If the players would still prefer a minimum limit, it is preferable that a low one of 5 or 10 points be adopted; a high minimum limit would really stagnate the game. I really don’t like even a low minimum limit, because it attaches a much higher significance to low-value patterns (such as “No Terminals”) than warranted by their difficulty and face values.

Why the deemphasis on “Concealed” and “Self-draw”?

Why no kui-sagari and no riichi?

Players used to the Modern Japanese system may have noticed that in the Zung Jung System, all one gets for keeping a hand Concealed is 5 points (except for hands with the “Concealed Triplets” patterns under 4.2). After all, many patterns are much harder to be completed Concealed than otherwise. Why not have something like the kui-sagari system (lowering of pattern values for melding) in Modern Japanese?

The answer is that I have determined the kui-sagari system to be downright undesirable. Mah-jong is a 4-player game with player interaction, not a 4-player solitaire. Giving much higher values to Concealed hands encourages everybody to just sit, draw and discard, with little Chi or Pong action.

Unlike the older systems, the Zung Jung System also has an explicit goal to reward effort, besides luck. I feel that whether a hand can maintain being Concealed or not is more a matter of luck than effort. In fact, the player who is not luckily blessed with a very good deal and good draws, but who nevertheless tries for a high- or medium-value pattern, is likely to end up having to meld when advancing his hand. Emphasizing Concealed hands rewards the lucky and discourages such efforts, violating the very first intention of a pattern-centered scoring system. Which is more interesting: a game in which players actively attepmt various patterns, such as Three Similar Sequences and Pure Terminal in Each Set, even if they are likely to have to claim some discards in doing so, or a game in which players tend to develop their hands along the smoothest path, usually sticking to patterns which can easily be completed concealed, such as “Concealed + No Terminals + All Sequences”?

The original concept of the game is to form sets with your tiles towards building a winning hand. Sets are supposed to include both concealed and exposed sets; exposed sets are every bit as ‘proper’ as concealed sets, in terms of their status as sets towards going out. Devaluing hands which contain exposed sets (which is a different thing from assigning different values to exposed and concealed triplets and quartets in Classical) disrupts the original balance of the game, resulting in a slow-paced game ending in lots of draws.

The kui-sagari system also adds significantly to rules complexity, in effect doubling the number of pattern values which need to be remembered before one can play with reasonable competence. It is also a lot of trouble to explain to a new player.

I suspect that the kui-sagari system has its origins in something other than an attempt to enhance playability. It is not in most scoring systems other than Japanese. Therefore the “Concealed Hand” receives only minimal credit in the Zung Jung System.

Where on earth has the all-important bonus for Self-draw gone?

Players used to Old Style and Taiwanese may have noticed the absence of any credit for self-draw in the Zung Jung System. Do I seriously mean that the all-important self-draw (which typically triples the score in the two systems) is not worth anything here (again, except sometimes for hands with the “Concealed Triplets” patterns under 4.2)?

My opinion is that the arbitrary self-draw bonus in Old Style and Taiwanese has led players to become obsessed with the concept. In Classical, the merit of self-draw is minimal: 2 points. Most players also recognize a double for an entirely Concealed hand which goes out on self-draw, but rarely do they recognize a double unconditionally for self-draw by itself.

What has always been in the Classical system is the double payoff (both plus and minus) for the dealer. When the dealer goes out, he receives 2 parts from each other player; when some other player goes out, he receives 2 parts from the dealer and 1 part each from the others. In Old Style, instead of to the dealer, this doubling is applied to the “chung” player, the player who discarded the final, winning tile. That is a good idea, but a side effect is that the doubling is applied to all players when the winner self-draws, as if he is the dealer in Classical. Coupled with the universal doubling in Old Style for self-draw in all situations, players soon become obsessed with the self-draw.

In Taiwanese, the “chung” player pays all: 1 part of the score. If the winner self-draws, everybody pays 1 part each. I have found no way of interpreting that other than as complete arbitrariness.

My stand is that the emphasis on self-draw in these systems (and also New Style, as a descendent of Old Style; and Chinese Official, which apparently has been inspired by the Taiwanese scheme) has been arbitrary. I find no reason in assigning a triple, or a double, or even a 1.5 multiplier to a hand for a mostly random event that occurs no less than 1/4 of the time. An important part of mahjong strategy is to estimate the values of the opponents’ hands, and to base the play of one’s hand on that. For example, if one notices that an opponent is attempting a big hand, it might become more advantageous for one to try to go out quickly with a small hand, rather than risking losing to the big hand while developing one’s own. However, excessive random fluctuations (inflation) to the value of a hand, such as an excessive self-draw reward, disrupt this strategy aspect of the game. Also, another aspect of the strategy is to avoid letting a big hand go out by discarding carefully. But a large bonus to self-draw renders this strategy meaningless (thus conflicting the very stated purpose of those payment schemes, namely emphasis on defense through punishing the “chung” player). In my humble opinion, the customary self-draw bonus in most modern Chinese systems is too disruptive to the playability of the game to be acceptable.

Well, one may ask, doesn’t the potential bonus for self-draw benefit the big-pattern builder as much as the speed-hand chicken? Unfortunately, with the inclusion of some New Style patterns in Zung Jung, the answer is a definite no, at least in some cases. Let’s consider a hand which is calling for a “5” to complete “Three Similar Sequences” in 567. The player draws a “8”. With no excessive self-draw bonus, the player may decide to discard the tile instead of going out, preferring to wait for the “5” in order to realize the full potential of the hand. However, a large self-draw bonus will sway the decision towards going out on the self-drawn tile, since even if he waits and later completes the valuable pattern on a discarded “5”, the merit (relative to going out immediately on the self-drawn “8”) is not as big as in the case when we don’t give the large bonus for self-draw. This retroactively means that all that effort in building the pattern is wasted. And that in turn implies that it is not that useful to try to build such patterns in the first place, after all, since it is less likely for the pattern to be actually completed. This is certainly an undesirable conclusion.

Looking at the vast ignorance among Hong Kong mahjong players of the ‘correct’ Chinese characters for mahjong terminology as given in Millington’s book, I believe in the conjecture that the current form of Hong Kong Old Style mahjong has once been propagated mostly by common people, who are not particularly intellectually oriented, and many of them illiterate, by oral tradition. (Editorial note: Some readers may find portions of Kwan’s language dated or overly dismissive in tone. However, the underlying historical point concerns the oral and decentralized transmission of Hong Kong mahjong, which contributed to significant regional variation and rules fragmentation.) Hence it is possible that the unconditional double for self-draw has come about not as an intentional ‘improvement’ to the game, but rather as a result of misunderstanding either the faan-pattern for the entirely Concealed hand, or the double payment for the self-draw according to the Old Style “chung” payment scheme. Thus I feel that it is perfectly reasonable to abolish it.

Shouldn’t “7 Pairs” get paid more?

Here, “7 Pairs” has been assigned a relatively low value of 25 (20 for the pattern itself, plus 5 for a Concealed Hand). [Editorial Note: Seven Pairs is worth 30 in Zung Jung Mahjong v3.3; this was a legacy score in 2.3, which Kwan increased in later versions.] In many other scoring systems, it is worth much more. Shouldn’t it get something like 100?

“7 Pairs” is not as difficult as it seems. In another article, I am trying to provide a more detailed explanation.

In the sequence-centered environment of Complementary, unlike the pong-centered environment of Classical, players are not so apt to pong their pairs at the earliest opportunity. Experiments show that under such environment, “7 Pairs” is not any rare occurrence. As pointed out in the other article, it is reasonable to believe that “7 Pairs” occur at a frequency no lower than that of “All Triplets”, probably much higher.

In practice, “7 Pairs” is not as often the fruit of incredible effort and patience as some might think. Actually, in a sequenced-centered environment, 7 Pairs often occurs as a bi-product, as a way for a hand that happens to get more pairs than it can use for a Regular Hand: it’s more often a back door than a noble goal to strive for. Trying to make “All Triplets” often involves more commitment and effort than ending up with “7 Pairs”.

Because of the nature in which many “7 Pairs” hands are developed, I find it undesirable to assign it a high value, for the sake of playability. The possibility that someone may luckily go out with a valuable “7 Pairs” hand would add a high risk factor to any player attempting other patterns. Even at a low value of 25, a hand with more pairs than it can use will still go out faster with 7 Pairs than with a Regular Hand. The Complementary System tries to reward effort, and rewarding convenience is contrary to that goal.

I feel that a large part of the appeal and fasciation of mahjong lies in the simplicity and elegance of the Regular Hand concept, and the focus of the game thereon. “Thirteen Terminals” is the polar opposite of the Regular Hand. Attempting it basically means that one has to give up most or all hopes of completing a Regular Hand. Such nature means that, in actual playing, the pattern does not distract the game from its focus on the Regular Hand to any noticeable degree. However, the “Seven Pairs” pattern is rather close to some Regular Hands, in particular those with “All Triplets” or “Two Identical Sequences”. It is not uncommon that one does not have to decide whether to settle for a regular hand or the irregular one until a late stage (in terms of hand development, such as “one tile to calling” or actually “calling”) in the hand. In other words, the “Seven Pairs” pattern presents a continuing distraction throughout the development of the possibly regular hand. Assigning an excessively high value to Seven Pairs would thus distract the game from its primary focus in such cases, and I feel that such is not desirable.

Players who still are not convinced may, of course, raise the value of the pattern. But it is recommended that the value be raised to no higher than 45 (which makes 50 with the 5 for Concealed Hand), for I have not found any evidence that “Seven Pairs” is any more difficult than “All Triplets”, and I strongly believe that the opposite is the case.

Why the “unit” payoff scheme?

The first reason is simplicity. In addition to being simple to learn, it also makes it easy to visualize the value of prospective hands when playing: if you are building a 50-point hand, you will collect 50 points from every other player should you complete it, and if someone else is building a 50-point hand, you are going to lose 50 points unless he is thwarted.

A couple of sections ago, I have explained why I think there is no reason to increase one’s income due to self-draw.

It is, of course, perfectly possible to assign some penalty to the “chung” player (the discarder) without rewarding the self-draw, as in the Japanese payoff system, or by using coefficients of 6-3-4 instead of 2-1-2 in Chinese Old or New Style. Defensive mahjong play is sometimes interesting. But since I am trying to emphasize patterns here, I hope to encourage aggressive, offensive play of boldly attempting valuable patterns. Emphasizing defensive play excessively will make it harder to complete valuable patterns, and this might leave some players bored and crying out for a minimum limit, something which I do not want.

Also, rules changes are not isolated items. Along with a heavier punishment for the “chung” player, sacred discard (furiten in Japanese) rules will be called for, adding yet more to rules complexity.

The above being said, I do not oppose a chung payment scheme. Players who prefer it are encouraged to try some level of chung payment scheme, and report their results and thoughts, provided that the scheme does not give any extra reward to the self-draw.

Concerning East doubling, I am rather indifferent to the concept, since it has been put out of use in many modern systems. I think that since we have a fair number of patterns here, East doubling is not quite needed. While in the Classical game one has to watch out for East, in Zung Jung there is often the opponent who is building a valuable hand who one must watch out for. East doubling would add to rules complexity, and it might also bring some inconvenience in busy playing environments where a round may have to be aborted midway. It also tends to encourage fast hands in some situations.

In an additive (as opposed to exponential) scoring system, adding some “base point value” to a winning hand would only encourage fast, cheap hands, contradicting the very purpose of having a scoring system with many patterns. Such a scheme is definitely not recommended in a pattern-centered system.

Why say that the Prevailing Wind is asymmetric?

How can the Prevailing Wind be ‘asymmetric’, when it applies equally to all players? After all, Zung Jung uses an additive scoring system, so there is not even the multiplicative advantage of the double wind as in exponential scoring systems.

The fact is that having the double wind still gives the player an advantage, even under an additive scoring system, because in mahjong play, the player chooses his plays, rather than discarding randomly.

Let’s consider a hand with 9 useful tiles and 4 wind tiles, 1 of each wind. The player draws 3 useful tiles (not winds) and discards 3 of the wind tiles. If the player has the double wind, he is now holding a double wind tile. Otherwise, he is holding only a single wind tile. From this example, we can see that the player with the double wind is indeed at an advantage.

In some playing environments, the game may need to be aborted after any hand, before a round of play (4 hands) can be completed. (One such example is playing before a banquet. If the players keep playing until they finish the round, all the food will be gone. ^_^ ) To improve ‘fairness’ (in a certain sense) in such playing environments, Zung Jung v2.21 onwards no longer recognizes the Prevailing Wind. There is another reason also, to make it easier for new players to understand the Winds. Seat Wind is a very straightforward concept, but when coupled with the Prevailing Wind, it becomes a bit more complex.

In situations where the above two concerns are immaterial, it is all right to recognize the Prevailing Wind if the players prefer, of course. After all, it is no big deal.

In fact, it can be argued that abolishing the prevailing wind would actually improve the color aspect of the game. In the Chinese Classical game, the dragons are considered the most valuable tiles in the set, since a dragon triplet is worth a faan (which is a big thing in Chinese Classical) to any player. Recognizing the prevailing wind would detract from this concept, since we now have a wind in each game which is essentially a fourth dragon. Abolishing the pattern would sharpen the contrast between dragons and winds: the dragons are always worth one faan, but the winds are worth one faan only for the corresponding player.

Isn’t it possible that a certain pattern is being scored too high?

This is possible, for the Zung Jung System is under development. If you have suggestions on specific pattern value adjustments, please do not hesitate to drop me a line.

Isn’t 10 points too high for “triplet of scoring honor”?

This is intentional, as a tribute which Zung Jung is paying to Classical mahjong.

In Chinese Classical mahjong, there are not many scoring patterns, and most of them (other than Limit Patterns) are worth only 1 faan. Mixed One-Suit and All Tiplets, in particular, are each worth 1 faan, while Pure One-Suit is worth 3 faan. In comaprison, the 1 faan one gets for a triplet (or kong) of dragon or scoring wind is quite a ‘big’ thing. Adding further to the significance of honors is that the 1 faan can be scored by non-winning players, while the faan for Mixed One-Suit, etc. cannot. Thus in Chinese Classical, the 1 faan for scoring honor triplets occupies a rather high significance. It is considered intrinsic to the game, and is never thought of as unbalancing or too high relative to other patterns.

As pattern faan values of other patterns are raised in most modern systems, the significance of the 1 faan for scoring honor, which in most cases has been staying at the same value because of the general ease of obtaining it, has become smaller and smaller. Yet, I feel that as we introduce many suit-tile patterns (in the case of Zung Jung, the Identical, Similar and Consecutive categories), it is appropriate and balancing to raise the value of scoring honors slightly, to make up for the fact that the honor tiles do not compose the new patterns. After all, unless one is luckily blessed with a pair of those honors to start with, holding a single honor tile in hopes of drawing another to form a pair does call for some sacrifice in terms of speed, for an honor tile cannot form a sequence, which is the strongest weapon in the arsenal of the speed hand. I feel that it is meaningful to remind ourselves of the importance of the honors in the Chinese Classical game, so I have given to them a value slightly higher than the 5-point patterns.

Isn’t 50 points too high for “Mixed Terminal in Each Set”?

[Editorial note: Kwan later reduced the value of this hand, known as Half Outside Hand in the ZJM naming, to 40 in Zung Jung Mahjong v. 3.3]

Most modern scoring systems, if they recognize “Mixed Terminal in Each Set”, assign to it a lower value than “Mixed One-Suit”. After all, you can “use” 25 tiles in Mixed Terminal in Each Set, while you have only 16 tiles in Mixed One-Suit to build your hand with, so the latter must be a lot harder.

In my experience, Mixed One-Suit isn’t that hard, and Mixed Terminal in Each Set isn’t that much easier, if at all. The numbers “25” and “16” tell only part of the story. Another comparison is the number of qualifying sets allowed by each pattern. Mixed One-Suit allows 7 sequences (123 to 789) and 16 triplets/pair (9 suit triplets and 7 honor triplets). Mixed Terminal in Each Set allows 6 sequences (123 and 789, in three suits) and 13 triplets/pair (terminal/honor triplets).

The fact is that both the number of allowed tiles and the number of allowed sets contribute to the practical difficulty and frequency of the patterns. In the case of these two patterns, I feel that the two factors just about cancel out, so it is appropriate to give them the same value.

In the case of Pure Terminal in Each Set and Pure One-Suit, the number of allowed tiles takes on a heavier weight, because 9 tiles is really a small number with which to build a hand: you can’t even have 4 sequences without repeating some tiles. Thus Pure One-Suit is given a higher value than Pure Terminal in Each Set.

Why these (English) pattern names?

In accordance with the “Zung Jung” doctrine, these names have been carefully chosen, aiming at the balance between ease of learning and ease of usage (convenience).

While traditional Chinese names such as “Three Sisters” sound very beautiful and culturally colorful, just by looking at the names, it is not immediately clear to the beginner what patterns these names stand for (Three Similar Sequences, in this example). This would create a barrier against learning the system: not only do the names offer no assistance in learning the patterns, but instead they become yet another thing to be remembered in addition to the patterns themselves! This would be undesirable when there are a fair number of patterns to be learned. For this reason, I have decided to put the culturally rich names aside, and use descriptive, scientific names. For players who are familiar with the traditional Chinese names, it is all right to use them when playing: when you disclose your winning hand containing Three Similar Sequences and claim “Three Sisters”, it is not hard for even beginners to figure out what you mean, from what they are seeing.

Another option is to use a fuller description of the pattern, such as “Three Same-Numbered Sequences in Three Suits”, as a direct translation of the Japanese name. However, while Chinese (the origin of many mahjong terminologies used in Japan) is a monosyllabic language, English is not. The above name is 4 Chinese characters, 4 syllables in Chinese, and 5 or 8 syllables in Japanese (depending on how you count; 5 if in a similar way as counting syllables in English). However, it is 7 (or 8) words, 38 letters and 10 syllables in English, and is a long, clumsy name to say or write. Such detail may be necessary if we are using all the common and uncommon New Style patterns, but since in Zung Jung we are using only patterns which belong within a few categories, the shorter ‘scientific’ names would suffice.

By ‘scientific’, I mean a naming system similar to that one learns for naming chemical compounds in elementary chemistry. Each word represents a specific concept, and the words in a name, when put together, would give a good picture of what the pattern is about. Here is a list of some of the words used in Zung Jung pattern names, and the concept each stands for:

Word	Concept
Mixed	mixed with honor tiles
Pure	does not include honor tiles
Big	contains the full thing, all as triplets
Small	one tile short of the full thing; one triplet is replaced by the pair of eyes
Identical	same-numbered sets (sequences) in the same suit
Similar	same-numbered sets across 3 different suits
Consecutive	consecutively numbered sets in the same suit

One might ask, why do I use the name “Nine-Tile Straight” instead of “Three Consecutive Sequences” for that pattern? The reason is because “Three Consecutive Sequences” is ambiguous: one may wonder whether it stands for 123-456-789, or (for example) 123-234-345. I suppose the meaning of “straight” should be known to everyone who is interested in card/tile games, so I have chosen to use the unambiguous name of “Nine-Tile Straight”.

Players who are familiar with “traditional” terminology (such as “chow”, “pong” and “kong”) and are more comfortable with those are welcome to use them during play and discussion. After all, I myself am Chinese and am totally comfortable with them.

— Alan Shiu Ho Kwan, 19 November 1997