Wednesday, October 17, 2012

A Need for Standardisation of Rules for Mahjong Tournaments in Singapore

I had started a series of articles on Singapore Style mahjong last year in an effort to document properly all the rules (well-known and obscure) of Singapore Style rules so that I can draft a more coherent set of rules for use in tournaments (see post on Tournaments and Singapore Style Mahjong Rules). This is necessary for me as a referee and technical advisor to community centres (CCs) organising mahjong tournaments, but this project is also a one-man show, and I hold no special authority other than the fact that I have done all my research on mahjong, as can be seen on this blog.

Anyway, a few incidents recently prompted some reflection on the state of mahjong tournaments in Singapore and the definition of Singapore Style mahjong: a few tournaments my students/associates (and I) participated in, a tournament where I served as referee and technical advisor, and ogiuemaniax's question (addressed here in Relative lack of low-value scoring elements in Singapore Style mahjong?).

Mahjong Tournaments in Singapore
Mahjong has an awkward position in Singapore: it is simultaneously reviled as an instrument for gambling and a social evil, and lauded as a tool to help active ageing. Up until recently, playing of mahjong was considered illegal (although apparently, it is not; even the police cannot really do anything about people playing mahjong, unless there is evidence of illegal gambling, or if the activity caused too many noise and disturbance), and playing of mahjong in public was outright disallowed. Mahjong tournaments, even if they were harmless and not associated with gambling, were also disallowed in public organisations such as the People's Association (PA) (see this ST report from 2008). Things have changed quite a bit since, and there are many mahjong tournaments being organised these days, and mostly by the CCs under the PA. Even so, the staging of mahjong tournaments is regulated by the Ministry of Home Affairs via the police, and permission has to be sought for such events.

So, while mahjong tournaments are becoming more common, the most important thing has not: the creation and adoption of a standardised mahjong ruleset. In Singapore, unfortunately, there is little interest in more standardised rulesets such as Mahjong Competition Rules (MCR), Zung Jung, or Japanese riichi. MCR, Zung Jung, and riichi of course also represent rather different variants of mahjong which Singaporeans are not familiar with, but they do have properly codified practices. Very well, Singaporeans may well ignore such standardised rulesets since they are foreign, but should they at least then look towards a standardised version of their favoured local rule variant, that is, Singapore Style Mahjong? From the looks of it, as experienced at various tournaments organised by different CCs, by NTUC U Live, and by private clubs, the answer seems to be a pathetic 'no'.

Rules? What Rules?
First off, some mahjong tournament organisers do not even bother to draft out proper rules, print them out and let the participants know beforehand of the rules to be used at the tournaments. Participants come to the tournament, clutching their own understanding of the rules and hope to be able to do well enough in an uncertain environment. Needless to say, without clear direction from the organisers, players often just arbitrate amongst themselves, leading to some rather divergent settlements. This is not exactly fair to participants who are sometimes disadvantaged by such arbitrary and non-standard decisions.

Worse, the rules at the tournament turn out to be quite different from the commonly used ones, so players have to adjust their strategies and styles of play on the spot. For example, at the recently staged mahjong tournament at Kampong Ubi CC (by the way, they called it a 'Friendly Mahjong Match'), the rules turned out to be quite different from expected. The notable differences were: no animals were used (!!); there was no one-double minimum, the flowerless pinghu value was adjusted from 4 doubles to only 3.5 doubles (actually, 10× the base stake; this is an ineffective adjustment anyway, for without animal tiles, this kind of pinghu is too easily achieved, so the value is highly inflated). Basically, these rules now resembled a hybrid form of Hong Kong Style and Singapore Style mahjong. Strangely, despite chicken hands (i.e. 0-double hands) being allowed, there was no properly allocated rate for 0-double wins. If the base stake was supposed to be 0.5/1, then the 1-double payment should be 1/2 and so on. But since a 0.5 chip was not feasible, the organisers decided to round up the value to 1, and so a 0-double win earned 1/1, which is hardly different from a 1-double win at 1/2. There goes any justification to even try for a 1-double win! This is patently a poorly thought-out solution because it closed the difference between 0 double and 1 double to almost nothing. A more logical solution would be to change the base stake to 1/2 (i.e. 0-double wins score 1/2) and double from there (i.e. 1-double wins score 2/4 and so on).

These changes were not minor and essentially changed the dynamics of the game so much that it was hardly fair to more seasoned players of the Singapore Style rules. In fact, this ruleset can hardly be called Singapore Style anymore! The worse thing was, had players known of such rules before registering for the tournament, they may have changed their minds and not joined it at all. My students and associates had tried asking for a copy of the rules, but were fobbed off with replies like 'oh, they will explain the rules on the day of the competition'. Evidently, they did not enjoy playing at the competition at all, and some have sworn off ever participating at this CC's tournaments.

Contrary to its aims of bringing people together, this mahjong tournament had annoyed veteran players with its seemingly idiosyncratic rules. Such atypical rules were explained as 'to cater for older players to make it easier' (I presume the organisers meant older novice players) but which at the same time confuses the veteran older players. Kampong Ubi CC's ruleset simply deviated too much from the more commonly used rules, as played by most people and as used in most other CCs.

I can only conclude that the organisers did not really know the rules very well, lacked experience in running such tournaments, and assumed too much of their participants (that they will be old[er] and are novices that require rules that make the game easier to play). With so many tournaments being organised by CCs nowadays, there are players who go around participating such tournaments as often as they can. Such players are not novices and hardly naive, and so, such non-standardised and arbitrary rules just make things difficult for them and are discouraging in nature.

Badly Written Rules
Some other CCs were not so poorly organised as to not have any rules and regulations to distribute to registrants for their tournaments. However, their efforts are often not good enough. For example, their rules (whether in English, Chinese, or a typically Singaporean mix of both) can contain many strange terms, Chinese characters, and bizarre explanations (see pictures below).


Fig. 1. An example of very poor copy-editing: where one important character is written as two different characters.

Figure 1 (above) shows the poor copy-editing that is found in printed rules and regulations for tournaments. In just one section, we can find examples of two different Han characters that supposed to be the same! Of course, in my view, neither 胡 nor 糊 are correct; the correct character that represents a winning/complete hand is actually 和 (and this is correctly used in PRC Chinese and Japanese publications on mahjong). But the point is, why do these people who organise tournaments and write/edit rules and regulations not spend more effort to ensure that basic things the name of a game element is written correctly and consistently? I do not even want to touch on the English names and the mixing of the Chinese and English text in the explanation for the scoring element.

Fig.2 An example of poor knowledge and/or research. How can such a common, basic word in the Chinese mahjong terminology be wrong?

Figure 2 (above) shows the lack of research and fact-checking that should be done for official documents (even if these are just for social events). It just shows a lack of knowledge, professionalism, and commitment to the cause. There is an abundance of mahjong materials in both English and Chinese available in our public libraries as well as on the Internet! There is no excuse to omit the basic research. What is even more unforgivable in this context is that the mistake in Figure 2 came from a set of rules and regulations prepared by a youth committee. So, even less of an excuse for not even giving Google a go! By the way, 冈 gāng is not even pronounced the same as 杠 gàng (the correct character), as was the case for the confusion between 胡/糊/(湖)/和 (all pronounced as ) as in Figure 1.

In another incident, I was helping out at a mahjong tournament at the invitation of the organising committee. But this invitation came about just three weeks before the event itself, and the organising committee had prepared a set of rules and regulations taken from another CC. Now, this set of rules and regulations seem to act as a common reference document as it seems that several CCs used the same for their own tournaments, but unfortunately, this original document had myriad errors, unconventional spellings, and wrong choice of Han characters (for example, 供 gòng instead of 杠 gàng), and poorly written explanations for the scoring elements. As referee, I had to adhere to a set of rules that I felt comfortable with (i.e. logical, consistent, and balanced) but the disseminated rules had so many errors and problems in them. Since the rules and regulations had already been released for dissemination, i.e. people who registered for the tournament received the already printed rules and regulations document, I was unable to revise the document totally. So, I ended up going through the entire set of rules during the rules briefing before the start of the tournament and answered all participants' questions individually. Even then, some participants came up to us (i.e. me and members of the organising committee) and berated us for giving them such a confusing and poorly written set of rules and regulations. What was I to say? I had no control over this particular matter.

The Need for Standardisation
All these recent incidents merely highlighted the need for a consistent and coherent set of Singapore Style rules that are stable and not change from tournament to tournament. Now that mahjong tournaments are becoming more common in Singapore, a pool of competitive players is growing, and there is a growing expectation that the rules conform to a common understanding of what constitutes Singapore Style mahjong. Whether there is true standardisation (and regulation by a governing body) remains to be seen, but a generally identical set of rules that recurs at the various CCs would be most welcome.

I now help two different CCs run mahjong tournaments and so they use my rules, and in a little way, I have helped create a little standardisation. It would be indeed presumptuous of me to suggest that CCs use my set of rules (which, by the way, is still a work in progress), but at the very least, they can use it as a form of reference to improve their own rules and regulations. I would welcome a form of forum amongst clubs that can thrash out all the rules to create a unified set of rules for use by all participating clubs: that would be the ideal platform for standardisation. But that would remain an improbable hope for now.

Wednesday, October 10, 2012

Relative Lack of Low-Value Scoring Elements in Singapore Style Mahjong?

I have just received an interesting comment/question from ogiuemaniax, who writes:
I am someone who primarily plays Japanese mahjong, but over the past year or so I've become increasingly interested in other forms of mahjong, if only to see what variety exists out there. In this time, I've had the chance to play Singapore-style online, whether against computers or human opponents (on Viwawa if you know that site), and I just wanted to ask someone more familiar with the game to see if my observations are at all accurate.

Coming from Japanese-style, I find the most significant difference to not be so much the lack of riichi or even the animal tiles (though they do contribute to a different experience), but the relative lack of low fan/tai hands. Things like tanyao (all simples) and iipeikou (two of the same sequence) don't exist in Singapore mahjong, and so I feel like the hands are more inflexible, that they don't flow into each other quite as easily.

As a result, it seems like you have to decide from the very beginning where you want your hand to go and that, unless you draw a relevant bonus tile, you have to stick to your path much more diligently, whether that's an all sequences hand or starting out with two honor tiles and planning around getting the third or aiming for a half flush.

I see that there is the possibiliy of increasing the tai minimum in order to make the game more challenging, but my opinion (based on my limited experience) is that high tai requirements are not suitable for Singapore Mahjong because of the relatively small amount of available hands. Unlike MCR especially where there are so many hands that you can realistically stack many of them together to meet more difficult fan requirements, I feel like anything more than a 2 tai minimum in Singapore Mahjong may make the game overly stiff and unwieldy, though whether the minimum is better at 1 or 2 is something I'm undecided on.

Anyway, thanks for hearing me out. I don't pretend to be an expert in even my own preferred mahjong, so I'm looking forward to a response.

A most interesting comment, with some interesting points to discuss! Below, I have laid out my thoughts and views in response to ogiuemaniax's comment. By no means are these points conclusive and definitive; there is a lot of scope for discussion.

1. Significant differences between Japanese riichi majan and Singapore Style mahjong.
I find it actually difficult to articulate the significant difference because Japanese riichi majan and Singapore Style mahjong are so different but yet still quite similar. I will try to discuss this point from a historical perspective.

Essentially, Japanese riichi majan and Singapore Style mahjong evolved from what is known as Chinese Classical, but in different ways. Both variants actually retain many features of Chinese Classical; we can compare this to variants such as American mah jongg (almost no resemblance to any other variant of mahjong anymore!), MCR, and Taiwanese. However, I view Singapore Style to be more conservative, and this includes the retention of most, if not all, of the traditional scoring elements found in Chinese Classical, whereas Japanese riichi majan innovated many new scoring elements (examples ogiuemaniax brought up include tanyao and iipeikou). Many of the newer styles of mahjong (e.g. Shanghainese, Hong Kong New Style, MCR, Zung Jung) have similar/identical innovations pertaining to new scoring elements. Singapore Style and Hong Kong Old Style, by contrast, use few, if any, new scoring elements beyond those inherited from Chinese Classical.

Because there are many more ways to obtain higher scores through hand manipulation, rather than through sheer luck (say, through the drawing of bonus tiles), Japanese riichi majan, MCR, Zung Jung have all developed to emphasise hand-construction and to de-emphasise luck. This is the reason, perhaps, why Japanese riichi majan use no flowers. Contrastingly, Singapore Style mahjong added four more bonus tiles. This makes obtaining the minimum of one double (one tai) relatively easier, and drawing more useful bonus tiles is not that difficult. In terms of play, a Singapore Style player can expect to win a lot of hands with one double, two doubles, three doubles, just using the bonus tiles alone.

So, if you were to ask me, the significant difference between Japanese riichi majan and Singapore Style mahjong is then not truly about the relative lack of low-value scoring elements, but the presence or absence of bonus tiles that fulfil the role of obligatory scoring element (i.e. the equivalent of yaku, not han). Since there are no bonus tiles that easily allow any player to meet the minimum requirement to win in Japanese riichi majan, the player is forced to use any of the available yaku to compete for the win. By necessity, a lot more yaku had been developed to allow the player more opportunities to complete the hand.

2. Flexibility and hand development
Well, it is true that without bonus tiles, the hand development in Singapore Style mahjong usually takes only one of three paths: All Chows (pinghu), All Pungs, and Mixed Suit/Half Flush. But this is not really very different from Japanese riichi majan. I doubt it is easy to 'flow' from an opened hand with a pung of Seat Wind back into something else other than All Pungs and/or Half Flush anyway. People tend to think of Japanese riichi majan as 'flexible' but only if the hand remains concealed (which then allow the completion of obligatorily concealed scoring elements such as riichi, pinfu, menzen, iipeikou etc. and some optionally concealed scoring elements). Singapore Style mahjong is as flexible as most other variants; opening up a hand always create a loss of flexibility. It is less of a hindrance in Singapore Style mahjong though, since bonus tiles usually shoulder the requirement for more doubles for high-scoring wins.

3. Scoring system dictates the strategy
Sometimes, it is rather difficult to compare different styles of mahjong. The very differences in the scoring system will create different conditions that then constrain the kinds of strategies usable. Singapore Style mahjong does not require the kind of flexibility demanded in Japanese riichi majan. If a player who is used to a style of play best suited for Japanese riichi majan play that way in Singapore Style mahjong, then he may not be able to do well. Since there is no restriction on concealment for pinghu in Singapore Style, players will readily open their hands to take chows to quickly advance their hands towards completion (moreso if such players have not drawn any bonus tiles and wish to aim for a 4-double pinghu win). The presence of bonus tiles affects the strategies of players in Singapore Style and cannot be discounted.

4. Minimum and maximum
Actually, as typically played in Singapore, a minimum of one double and a maximum of five doubles is 'standard' (well, 'standard' means 'very common' here) after decades of game evolution and equilibrium. The scoring system (with all the associated bao penalties etc.) have stabilised to be most balanced under the conditions of a minimum of one double and a maximum of five doubles.

If there is no minimum (thus allowing Chicken Hands), then there is not enough challenge for players; to meet a minimum of one double, players must know how to construct a pinghu hand correctly in the absence of a valid bonus tile. If the minimum is raised to two doubles, players cannot pre-empt dangerous players late-game with a cheap one-double pinghu hand; therefore any minimum above one double is typically unpopular with the more experienced and expert players.

The maximum of five doubles fits nicely with limiting the role of luck (especially for luck-based limit hands like Heavenly Win, Eight Flowers etc.), and at the same time making the role of bao penalties relevant and interesting. If the limit is too high, say, at 8 doubles, then a baoda penalty, where a player becomes penalised for discarding a tile allowing a dangerous player to go from 7 to 8 doubles (i.e. to reach the Limit; based on the exposed bonus tiles and Honour tile melds on the tables) becomes all too rare and thus pointless. It is common to see 4 doubles on the table, so a baoda penalty is quite likely in such situations; much less common to see 7 doubles on the table.


ogiuemaniax, I hope this answers your query in some part. Feel free to carry on this discussion with your comments.