[ZJ] Zung Jung v3.01 update
Alan Kwan
which indicates the tentative changes for the next official release (v3.1).
The pattern list document can be found on the Zung Jung official website.
Have fun playing!
--
"大牌之所以大，就是貴在於能搶在小牌之前和牌。強加「起和」規定，
小牌不准和的話，便誰也懂得和大牌，沒有甚麼值得稀罕的。
要求大牌要能搶在小牌之前才能和，這才是真正的技術挑戰。"
"The true challenge of skill lies where big hands have to beat small
hands in speed in order to win. With a Minimum Requirement rule, anybody
can make big hands with no impediment; they cease to be extraordinary."
- Alan Kwan / ta...@netvigator.com
Zung Jung mahjong official website: http://www.zj-mahjong.info/
Archon_Wing
> v3.01 is not an official release version, but rather a test release version
> which indicates the tentative changes for the next official release (v3.1).
> The pattern list document can be found on the Zung Jung official website.
>
> Have fun playing!
I suppose I'll do some experimenting next week. I feel the changes are
a bit odd, but people never like change. :D It's best if you can offer
some explanations
[Btw, have you thought of anything to settle the concerns of "chicken
addicts" in the WSOM?)
Here's what I think:
==
1.2 Concealed Hand (門前清) : 5
A regular hand which is concealed, without melding any exposed
sets before winning. Winning on discard is okay. Concealed kong are
okay.
===
At first I didn't understand the purpose of this, but this mainly
applies to seven pairs. This makes thinks clearer; just scoring it as
30 points is what I do anyways instead of just 25+5.
==
2.1.2 Pure One-Suit (清一色) : 90
==
If I'm correct, the value of this has been dropping throughout the
versions. I think it is ok; it is definitely overrated, however 100
was a nice and even number.
==
4.1 All Triplets (對對和) : 30
==
I thought 40 was nice, being consistent with the other medium "core"
patterns. I personally rarely complete it because of other options.
With four pairs it seems easier for seven pairs, and less I'd rather
make sequences. Given the dominance of sequence based hands in the
more practically completed patterns, I feel All Triplets serves as a
nice counterbalance. Sequences are the fast way out, triplets are the
long way around.
==
9.4.1 Blessing of Heaven (天和) : 155
9.4.2 Blessing of Earth (地和) : 155
==
I guess they don't really affect much in the long run
==
10.1 Thirteen Terminals (十三么九) : 160
==
That's intresting-- I suppose it's not really that rare.
Alan Kwan
Many thanks for your support of Zung Jung.
Archon_Wing wrote:
> [Btw, have you thought of anything to settle the concerns of "chicken
> addicts" in the WSOM?)
It's a non-issue in the first place - chicken addicts will always lose.
We just need to schedule the tournament format better, so that the
chicken addicts get eliminated before they can bring their
opponents down with them. We just need to make sure that the
first (and perhaps the second) effective round is 8 cycles long,
so that good players get enough chance to build their good hands before
being eliminated.
> Here's what I think:
>
> ==
> 1.2 Concealed Hand (門前清) : 5
> A regular hand which is concealed, without melding any exposed
> sets before winning. Winning on discard is okay. Concealed kong are
> okay.
> ===
> At first I didn't understand the purpose of this, but this mainly
> applies to seven pairs. This makes thinks clearer; just scoring it as
> 30 points is what I do anyways instead of just 25+5.
This is just a change of the "presentation"; the actual hand values are,
of course, not really changed.
This change has been inspired by the revelation that the non-ZJ pattern
"Five Suits" should only apply to regular hands, and not to Seven Pairs,
because of the "causal basis" of the pattern. Thus I do a similar thing
to Concealed Hand.
> ==
> 2.1.2 Pure One-Suit (清一色) : 90
> ==
> If I'm correct, the value of this has been dropping throughout the
> versions. I think it is ok; it is definitely overrated, however 100
> was a nice and even number.
Yes, 100 was a nice and even number, but as you agreed it was clearly overrated.
> ==
> 4.1 All Triplets (對對和) : 30
> ==
> I thought 40 was nice, being consistent with the other medium "core"
> patterns. I personally rarely complete it because of other options.
> With four pairs it seems easier for seven pairs, and less I'd rather
> make sequences. Given the dominance of sequence based hands in the
> more practically completed patterns, I feel All Triplets serves as a
> nice counterbalance. Sequences are the fast way out, triplets are the
> long way around.
That's what I have been feeling, and when I play, I still don't go for
All Triplets often. But others play differently, and it was a fact that
during the WSoM, All Triplets was a dominating pattern, with a very high
frequency of completion. Also, some felt that the value of Seven Pairs was
too low when compared to All Triplets. So while I do not personally feel
the need to change this, I'm changing this in response to overwhelming
"popular opinion".
> ==
> 9.4.1 Blessing of Heaven (天和) : 155
> 9.4.2 Blessing of Earth (地和) : 155
> ==
>
> I guess they don't really affect much in the long run
These two patterns are more cosmetics and theme than anything.
> ==
> 10.1 Thirteen Terminals (十三么九) : 160
> ==
>
> That's intresting-- I suppose it's not really that rare.
It has been alarmingly frequent in the satellite tournaments;
other (non-ZJ) sources also indicate a high frequency;
my own combinatoric works have also indicated that it is
easier than some other patterns (namely Nine Gates and
Mixed Greater Terminals + Seven Pairs). It has been
traditionally overrated, just because it is an irregular hand -
overrating hands just for being irregular has been a common problem
in many scoring systems.
Archon_Wing
> Dear Archon,
>
> Many thanks for your support of Zung Jung.
>
> Archon_Wing wrote:
>
> > [Btw, have you thought of anything to settle the concerns of "chicken
>
> > addicts" in the WSOM?)
>
> It's a non-issue in the first place - chicken addicts will always lose.
>
> We just need to schedule the tournament format better, so that the
> chicken addicts get eliminated before they can bring their
> opponents down with them. We just need to make sure that the
> first (and perhaps the second) effective round is 8 cycles long,
> so that good players get enough chance to build their good hands before
> being eliminated.
chicken hands since I know if someone just goes for them they will
lose at the end of the session. Having sessions of good length is
necessary to emphasize the skill aspect.
That being said, I strongly believe the "chicken addicts" have not
read the scoring cards well at all! Or maybe they are too accustomed
to playing with a minimum limit and got too enthusastic. One of the
basis for many who argue for a minimu limit is that going chicken is
some invincible strategy and a chicken addict will always beat those
who try to build "real" hands. The WSOM may have proven itself to be
an argument that this simply isn't true, as long as you don't give a
large base score or easy to get luck factors that a fan based system
simply cannot handle.
> > ==
> > 4.1 All Triplets (對對和) : 30
> > ==
> > I thought 40 was nice, being consistent with the other medium "core"
> > patterns. I personally rarely complete it because of other options.
> > With four pairs it seems easier for seven pairs, and less I'd rather
> > make sequences. Given the dominance of sequence based hands in the
> > more practically completed patterns, I feel All Triplets serves as a
> > nice counterbalance. Sequences are the fast way out, triplets are the
> > long way around.
>
> That's what I have been feeling, and when I play, I still don't go for
> All Triplets often. But others play differently, and it was a fact that
> during the WSoM, All Triplets was a dominating pattern, with a very high
> frequency of completion. Also, some felt that the value of Seven Pairs was
> too low when compared to All Triplets. So while I do not personally feel
> the need to change this, I'm changing this in response to overwhelming
> "popular opinion".
IMO, it is not because the scoring is broken or All triplets are too
easy. I think it is the natural result of having many players from
nearby Hong Kong in the tournament. Most of them have probaly played a
lot of some version of HKOS and all triplets is a big pattern simply
because it is one of the few viable patterns that one can build. When
faced with a new system, one would stick with what they know best.
Since mixed one suit and all triplets score well in the system, an
HKOS player could really just stick to their regular way of playing
with some adjustments. [Not that it is a bad thing-- it's desirable
for one person to not have to radically change their play when playing
another variant. The key to learning is usually not a radical one.]
Indeed, in my group where we played HKOS before, all triplets is a
definite fallback and those scary "alien" patterns like seven pairs
are too hard, or harder then they really are. I wonder who would be
complaining about it; I suspect it might include some riichi players
who are trying to quietly build their 15 point hand might be annoyed
at all the ponging going around and not using multiple waits to go
out. I have certainly had that feeling before. ;)
> > ==
> > 10.1 Thirteen Terminals (十三么九) : 160
> > ==
>
> > That's intresting-- I suppose it's not really that rare.
>
> It has been alarmingly frequent in the satellite tournaments;
> other (non-ZJ) sources also indicate a high frequency;
> my own combinatoric works have also indicated that it is
> easier than some other patterns (namely Nine Gates and
> Mixed Greater Terminals + Seven Pairs). It has been
> traditionally overrated, just because it is an irregular hand -
> overrating hands just for being irregular has been a common problem
> in many scoring systems.
Yes, most people don't believe me when I tell them 13 terminals isn't
really that special, and it's almost always been a matter of
tradition. It does get kind of awkward trying to score a non-limit
irregular hand though.
Alan Kwan
> That being said, I strongly believe the "chicken addicts" have not
> read the scoring cards well at all! Or maybe they are too accustomed
> to playing with a minimum limit and got too enthusastic. One of the
> basis for many who argue for a minimu limit is that going chicken is
> some invincible strategy and a chicken addict will always beat those
> who try to build "real" hands. The WSOM may have proven itself to be
> an argument that this simply isn't true, as long as you don't give a
> large base score or easy to get luck factors that a fan based system
> simply cannot handle.
Yes, the elimination of the base score and all other side benefits of
merely winning (such as additional payments or East advantage rotation)
reduces chicken hands to their face value: 1 point and the termination
of the current hand. I believe ZJ is the first system to try this.
>>>==
>>>10.1 Thirteen Terminals (十三么九) : 160
>>>==
>>
> Yes, most people don't believe me when I tell them 13 terminals isn't
> really that special, and it's almost always been a matter of
> tradition. It does get kind of awkward trying to score a non-limit
> irregular hand though.
Well, the 160-point patterns are limit-class patterns (in other scoring systems).
The balance now is much better, with the three (13 Terminals, 4 Concealed
Triplets, Big 3 Dragons) together at this level.
Archon_Wing
> Archon_Wing wrote:
> > That being said, I strongly believe the "chicken addicts" have not
> > read the scoring cards well at all! Or maybe they are too accustomed
> > to playing with a minimum limit and got too enthusastic. One of the
> > basis for many who argue for a minimu limit is that going chicken is
> > some invincible strategy and a chicken addict will always beat those
> > who try to build "real" hands. The WSOM may have proven itself to be
> > an argument that this simply isn't true, as long as you don't give a
> > large base score or easy to get luck factors that a fan based system
> > simply cannot handle.
>
> Yes, the elimination of the base score and all other side benefits of
> merely winning (such as additional payments or East advantage rotation)
> reduces chicken hands to their face value: 1 point and the termination
> of the current hand. I believe ZJ is the first system to try this.
Out of curiosity, were there detailed stats taken during the WSOM? I'm
interested in knowing the frequency of patterns completed and how it
compares with my own experiences.It might help with my knowledge of
strategy because of course, there's not written much about it
regarding Zung Jung ^_^
Alan Kwan
> Out of curiosity, were there detailed stats taken during the WSOM? I'm
> interested in knowing the frequency of patterns completed and how it
> compares with my own experiences.It might help with my knowledge of
> strategy because of course, there's not written much about it
> regarding Zung Jung ^_^
Unfortunately, we could not afford to record the data.
We only have records of a few hands, for the TV production.
Tom Sloper
> Archon_Wing wrote:
>
>> Out of curiosity, were there detailed stats taken during the WSOM?
>
> We only have records of a few hands, for the TV production.
Of course. To record each and every hand at a tournament would be well
beyond the bounds of practicability today. The number of video cameras
and/or recordkeepers/analysts would be most impractical indeed.
Or, to require each player's hand to be notated on paper after each hand
would slow things down and greatly impede the enjoyment of the tournament.
Unless someone can suggest a method that could be fast and affordable, it's
not reasonable to expect detailed stats to be taken during a tournament.
Tom
Archon_Wing
Ah yes, I guess it'd be nice for some of it to be on TV, although
unfortunately, since I live in the states, I probaly won't be able to
watch it.
Benjamin Boas
wrote:
>
> Of course. To record each and every hand at a tournament would be well
> beyond the bounds of practicability today.
I disagree. Perhaps this would be overly impractical at many
tournaments, but certainly not all. In Japan, recording each and
every hand is standard practice in the upper pro leagues of the JPML,
as well as during major pro tournaments and the last rounds of major
open tournaments. How do you think a Mahjong magazine like Kindai
Maajan can run regular articles on tournaments? Their writers have
complete game records of every hand (or at least every decisive
hand.) This is particularly important for riichi, since opponents'
discards have a great impact on one's play.
The number of video cameras
> and/or recordkeepers/analysts would be most impractical indeed.
All you need is one recordkeeper per player to sit behind that payer
and take down the initial hand and all draws, discards and claims, as
well as the final hand structure. If this is done for all players,
you have a complete game record. All pros are trained to do this sort
of recording and it is not difficult to learn. (Anyone CCed to this
email can teach you, Tom). If anyone reading this has a copy of Mai
Hatsune's book, you can see this method used in the strategy articles
near the end of the book. It is the standard in Japanese mahjong
media.
Video cameras are entirely unnecessary and analysts can do their work
after the game is over.
> Or, to require each player's hand to be notated on paper after each hand
> would slow things down and greatly impede the enjoyment of the tournament.
This is also unnecessary, although if you assume the level of players
to be high enough I don't think it would take very long as you say.
All you need is a quick transcription method. The most common one
used in Japan is circled numbers (1) (3) for dots, chinese numerals
for characters, regular numbers for bamboos and graphical
representations of honor tiles. Memorizing this simple method will
enable anyone to quickly write down any hand.
Alternatively, you could write b,c, or d before every numeral and
write the first English letter of each honor (E, S,W,N,R,W,G).
This is an efficient and effective system that I hope will spread
outside of Japan. If it does, we may start to see more English-
language strategy articles.
> Unless someone can suggest a method that could be fast and affordable, it's
> not reasonable to expect detailed stats to be taken during a tournament.
I agree that this cannot be done for every hand at every mahjong
tournament and I also agree that taking detailed stats at every
tournament would be very impractical (and also, in my opinion, not of
much use).
Recordkeeping would have been impossible to do during either of the
OEMCs, for example. The level of play was not high enough and there
were not enough non-players to serve. But for the WSOM, it may
actually be possible since it is structured as an elimination
tournament. By the 2nd round, half the players have been eliminated
so there are enough recordkeepers to go around. Of course, there are
complications. Making sure that eliminated players would be willing
to do this and that recordkeeping was accurate would require a good
deal of preparation but I think it's entirely possible.
Keeping complete game records may be impractical in many cases but in
cases where it is possible it would provide a big benefit to the
development of Mahjong culture in the English speaking world. I
imagine a day when many of discussions on this newsgroups are about
strategy and what to discard in specific situations. I think this day
will come much sooner if simple and efficient hand transcription/
recording methods spread.
Edwin Phua
professional), but I think it should be quite easy to use a digital
camera to take a photograph of every winning/completed hand.
It would take just a few seconds to snap a quick but serviceable
picture of the winning hand, and transcribing out the hand can be done
at leisure later. If a full game/session has only 16 hands, then 16 or
less photographs would be needed. This would be enough to gather data
on winning hands. Granted, the analysis would be less complete than
the way done by the Japanese players, but this method is good enough
for all levels of play and need not require recordkeepers. This
photographing can be done by each table of players.
Cheers!
Edwin Phua
Tom Sloper
>> To record each and every hand at a tournament would be well
>> beyond the bounds of practicability today.
"Benjamin Boas" <ben...@gmail.com> wrote
> I disagree. Perhaps this would be overly impractical at many
> tournaments, but certainly not all.
>...
> I agree that this cannot be done for every hand at every mahjong
> tournament and I also agree that taking detailed stats at every
> tournament would be very impractical (and also, in my opinion, not of
> much use).
>...
> Recordkeeping would have been impossible to do during either of the
> OEMCs, for example.
In other words, you disagree. But you agree. (^_^)
The tournament in question is the 2007 World Series of Mahjong (the one in
Macau in June, not to be confused with the one in Chengdu in November or the
one that didn't happen in Australia this year or the one that ...) which
was played with a starting lineup of 256 players.
THAT is the scenario I was addressing.
Also, the original question was about "statistics" (and "the frequency of
patterns completed") - which would assume not only hands that were won, but
also attempted hands that weren't won. When someone asks for statistics,
they're probably asking for a lot more data than is reasonable to collect at
a tournament with hundreds of players. As we already agree. ... I think.
> In Japan, recording each and
> every hand is standard practice in the upper pro leagues of the JPML,
> as well as during major pro tournaments and the last rounds of major
> open tournaments. How do you think a Mahjong magazine like Kindai
> Maajan can run regular articles on tournaments? Their writers have
> complete game records of every hand (or at least every decisive
> hand.)
Sure.
> All you need is one recordkeeper per player to sit behind that payer
> and take down the initial hand and all draws, discards and claims, as
> well as the final hand structure.
Impractical with 100 or 256 players, as you already agree. So what are we
disagreeing about again?
> All pros are trained to do this sort
> of recording and it is not difficult to learn. (Anyone CCed to this
> email can teach you, Tom).
This is an email? I thought it was a newsgroup post. (^_^)
And of course I use MJ notation when I take written notes of tricky hand
situations I encounter. But I always yearn to learn!
> If anyone reading this has a copy of Mai
> Hatsune's book, you can see this method used in the strategy articles
> near the end of the book. It is the standard in Japanese mahjong
> media.
Got a page # for me? Flipping through it just now I saw tile images on every
page...
> This is also unnecessary, although if you assume the level of players
> to be high enough I don't think it would take very long as you say.
> All you need is a quick transcription method. The most common one
> used in Japan is circled numbers (1) (3) for dots, chinese numerals
> for characters, regular numbers for bamboos and graphical
> representations of honor tiles. Memorizing this simple method will
> enable anyone to quickly write down any hand.
Sure, I've seen that in there. Wasn't hard to figure it out.
> Alternatively, you could write b,c, or d before every numeral and
Well, some folks would say put the letter before the numeral, some prefer
putting it after. Some would say put the letter before /each/ numeral, some
prefer using one letter for each grouping in one suit. Can't we all just get
along? :-D
> write the first English letter of each honor (E, S,W,N,R,W,G).
No, that doesn't work - W = W, so you need W and Wh (that's standard
notation I usually see in English materials).
> This is an efficient and effective system that I hope will spread
> outside of Japan. If it does, we may start to see more English-
> language strategy articles.
If you just mean MJ notation for writing out a list of tiles in a hand or
whatever, it's already been widely used here on this newsgroup for many
years. Hmm, I just realized, this newsgroup is 10 years old this year!
> But for the WSOM, it may
> actually be possible since it is structured as an elimination
> tournament.
The Chengdu event? I hadn't noticed that in the signup materials.
Interesting.
> By the 2nd round, half the players have been eliminated
> so there are enough recordkeepers to go around. Of course, there are
> complications. Making sure that eliminated players would be willing
> to do this and that recordkeeping was accurate would require a good
> deal of preparation but I think it's entirely possible.
>
> Keeping complete game records may be impractical in many cases but in
> cases where it is possible it would provide a big benefit to the
> development of Mahjong culture in the English speaking world.
No argument there. I'm mindful of televised majan matches from Japanese TV,
with like 6 cameras - one looking at each player's hand, one bird's-eye view
looking down on the table, and one that gives a view of the 4 players
sitting around the table, as if viewed by someone standing nearby. 5 or 6
cameras seems like a lot for one game but it's really the way to go. Trying
to do that for 256 players - to keep accurate data, from which to build
statistics - would be impractical.
> I
> imagine a day when many of discussions on this newsgroups are about
> strategy and what to discard in specific situations. I think this day
> will come much sooner if simple and efficient hand transcription/
> recording methods spread.
See, we don't disagree! So to your statement "I disagree," I have to say "I
disagree." Because we do, in the end, agree.
Tom
Tom Sloper
That's a thought. I presume each table would need to have a camera, or each
referee would need to have a camera and run to a table when it's finished a
hand. It would be necessary to check each photo to make sure it isn't
blurred or insufficiently lit, and to collect all photos to the
competition's central computer.
Planners would have to work out the logistics of all that.
But also it only captures the winning hand - so it doesn't make a full set
of statistics that would provide usable statistics on the frequency of
making particular hands (how many times attempted vs. how many times
successful)...
Recordkeeping is reasonably easy to do for small events, at least.
It's gonna take quite some time to have a history of mahjong games as
extensive as chess games. Chess is so much easier to notate - the starting
position is always the same, so you only have to record which piece moves
where. Would that MJ notation to record an entire game could be developed.
Cheers,
Tom
>
Benjamin Boas
responded to a lengthy post of mine, so lets see how I do at the
newsgroup game.
On Aug 13, 8:38 pm, "Tom Sloper" <tslo...@DONTsloperamaSPAMME.com>
wrote:
> > On Aug 12, 10:36 pm, "Tom Sloper" wrote:
> >> To record each and every hand at a tournament would be well
> >> beyond the bounds of practicability today.
>
> "Benjamin Boas" <ben...@gmail.com> wrote
>
> > I disagree. Perhaps this would be overly impractical at many
> > tournaments, but certainly not all.
> >...
> > I agree that this cannot be done for every hand at every mahjong
> > tournament and I also agree that taking detailed stats at every
> > tournament would be very impractical (and also, in my opinion, not of
> > much use).
> >...
> > Recordkeeping would have been impossible to do during either of the
> > OEMCs, for example.
>
> In other words, you disagree. But you agree. (^_^)
> The tournament in question is the 2007 World Series of Mahjong (the one in
> Macau in June, not to be confused with the one in Chengdu in November or the
> one that didn't happen in Australia this year or the one that ...) which
> was played with a starting lineup of 256 players.
>
> THAT is the scenario I was addressing.
Well, if we're nitpicking then why did you say "a tournament" instead
of "the tournament"? (^_^) If I took your statement out of context
then I apologize, but from your own words it seems you're generalizing
to ALL tournaments.
>
> Also, the original question was about "statistics" (and "the frequency of
> patterns completed") - which would assume not only hands that were won, but
> also attempted hands that weren't won. When someone asks for statistics,
> they're probably asking for a lot more data than is reasonable to collect at
> a tournament with hundreds of players. As we already agree. ... I think.
The original question was about statistics but your post seemed to be
entirely about recordkeeping, hence my reply (and topic change) on
recordkeeping.
>
> > In Japan, recording each and
> > every hand is standard practice in the upper pro leagues of the JPML,
> > as well as during major pro tournaments and the last rounds of major
> > open tournaments. How do you think a Mahjong magazine like Kindai
> > Maajan can run regular articles on tournaments? Their writers have
> > complete game records of every hand (or at least every decisive
> > hand.)
>
> Sure.
>
> > All you need is one recordkeeper per player to sit behind that payer
> > and take down the initial hand and all draws, discards and claims, as
> > well as the final hand structure.
>
> Impractical with 100 or 256 players, as you already agree. So what are we
> disagreeing about again?
Well, for 256 I agree (to agree) but for 100 players I think it's
certainly possible and practical in some circumstances. But its
probably impractical in general so maybe I have to disagree with
myself that I'm disagreeing. (Do you agree?)
>
> > All pros are trained to do this sort
> > of recording and it is not difficult to learn. (Anyone CCed to this
> > email can teach you, Tom).
>
> This is an email? I thought it was a newsgroup post. (^_^)
> And of course I use MJ notation when I take written notes of tricky hand
> situations I encounter. But I always yearn to learn!
You are correct, it was a newsgroup post. It was also sent by email.
Therefore I think we're both right on this one (even though we
disagree.)
I wonder though, how common it is for players on this newsgroup to
take notations of a *complete* round for all players (discards
included). Anyone?
>
> > If anyone reading this has a copy of Mai
> > Hatsune's book, you can see this method used in the strategy articles
> > near the end of the book. It is the standard in Japanese mahjong
> > media.
>
> Got a page # for me? Flipping through it just now I saw tile images on every
> page...
I don't have the book on me right now, but the diagrams are pretty
easy to spot. The article that has the most of these is the one thats
either 2nd or 3rd to last in the book. It's on different playing
styles, or more specifically what different players do when they have
the same initial hand and the same draws. Look for diagrams that show
tiles in the initial hand, a row of drawn tiles, a row of discards,
and then the final hand. The row of drawn tiles and the initial hand
will be the same in all diagrams. If you see a picture of Ryan Morris
(the only non-Japanese) smiling in a diagram, you're in the right
spot.
>
> > This is also unnecessary, although if you assume the level of players
> > to be high enough I don't think it would take very long as you say.
> > All you need is a quick transcription method. The most common one
> > used in Japan is circled numbers (1) (3) for dots, chinese numerals
> > for characters, regular numbers for bamboos and graphical
> > representations of honor tiles. Memorizing this simple method will
> > enable anyone to quickly write down any hand.
>
> Sure, I've seen that in there. Wasn't hard to figure it out.
>
> > Alternatively, you could write b,c, or d before every numeral and
>
> Well, some folks would say put the letter before the numeral, some prefer
> putting it after. Some would say put the letter before /each/ numeral, some
> prefer using one letter for each grouping in one suit. Can't we all just get
> along? :-D
I think we all get along. I thought we were talking about whether we
all agree, although I'm personally open to disagreement on this (^_^)
>
> > write the first English letter of each honor (E, S,W,N,R,W,G).
>
> No, that doesn't work - W = W, so you need W and Wh (that's standard
> notation I usually see in English materials).
You are absolutely correct. I'm used to using the Japanese characters
for writing down honors and I got hasty. My mistake. This is kind of
like chess notation where you have to use "N" for knight.
>
> > This is an efficient and effective system that I hope will spread
> > outside of Japan. If it does, we may start to see more English-
> > language strategy articles.
>
> If you just mean MJ notation for writing out a list of tiles in a hand or
> whatever, it's already been widely used here on this newsgroup for many
> years. Hmm, I just realized, this newsgroup is 10 years old this year!
I meant the way that drawn tiles and discarded tiles were also
recorded, not just how you transcribe a hand. But you're right, I
could've (and should've) clarified. Perhaps the difference between
the system I've seen in Japan and the newsgroup is that there isn't
much variation in what I've seen. Probably doesn't matter so much.
Happy birthday, MJnewsgroup.
>
> > But for the WSOM, it may
> > actually be possible since it is structured as an elimination
> > tournament.
>
> The Chengdu event? I hadn't noticed that in the signup materials.
> Interesting.
I think we both know which tournament I'm talking about. (The WMO-
WSOM contains tournaments other than Chengdu so I think the nitpick-
factor favors me here.)
>
> > By the 2nd round, half the players have been eliminated
> > so there are enough recordkeepers to go around. Of course, there are
> > complications. Making sure that eliminated players would be willing
> > to do this and that recordkeeping was accurate would require a good
> > deal of preparation but I think it's entirely possible.
I'm surprised you didn't reply to this paragraph. Im interested in
other people's opinions on this, especially Alan's. If you're going
to eliminate players every round, it'd be nice to put them to good
use, right?
>
> > Keeping complete game records may be impractical in many cases but in
> > cases where it is possible it would provide a big benefit to the
> > development of Mahjong culture in the English speaking world.
>
> No argument there. I'm mindful of televised majan matches from Japanese TV,
> with like 6 cameras - one looking at each player's hand, one bird's-eye view
> looking down on the table, and one that gives a view of the 4 players
> sitting around the table, as if viewed by someone standing nearby. 5 or 6
> cameras seems like a lot for one game but it's really the way to go. Trying
> to do that for 256 players - to keep accurate data, from which to build
> statistics - would be impractical.
This I do actually disagree with. In terms of recording game records
and statistics I don't think video is the way to go at all. The
reason they use so many cameras for that show, and I'm assuming you
mean the show on Mondo21, is to make it easy to see everyone's hand
and to make things interesting for the viewer. If you want to keep
accurate game records of entire games I think you need multiple
'witnesses.' If you want records of only final hand shapes (or
winning hands) I think it's best and easiest to have the players right
them down, as in chess. If you want to keep statistics, I'm not sure
this can be easily done except on matches played by a computer. Even
if you got the game records who would want to go through all of
them?
>
> > I
> > imagine a day when many of discussions on this newsgroups are about
> > strategy and what to discard in specific situations. I think this day
> > will come much sooner if simple and efficient hand transcription/
> > recording methods spread.
>
> See, we don't disagree! So to your statement "I disagree," I have to say "I
> disagree." Because we do, in the end, agree.
Oh, but I think we still do disagree, both in our individual points
and in terms of the discussion as a whole. So I disagree with your
disagreement that we disagree.
Edwin Phua
> "Edwin Phua" <fant...@pacific.net.sg> wrote...
> > Well, my suggested method is not as complicated as Benjamin's (nor as
> > professional), but I think it should be quite easy to use a digital
> > camera to take a photograph of every winning/completed hand.
>
> referee would need to have a camera and run to a table when it's finished a
> hand. It would be necessary to check each photo to make sure it isn't
> blurred or insufficiently lit, and to collect all photos to the
> competition's central computer.
>
> Planners would have to work out the logistics of all that.
>
> But also it only captures the winning hand - so it doesn't make a full set
> of statistics that would provide usable statistics on the frequency of
> making particular hands (how many times attempted vs. how many times
> successful)...
With digital cameras (and camera-enabled mobile phones) so cheap and
ubiquitous, I think getting cameras is not a big problem. Perhaps the
problem would in coordinating the collection of the data (i.e. either
the photos or the transcribed hands).
As for the winning hand versus attempted hand, I am skeptical how
practical recording attempted hands is. After all, non-winnings hands
may not have any clear patterns compared to a winning hand. For
example, Player A may be attempting either Thirteen Orphans or one of
the Honours and Knitted Tiles hands but he may not draw any useful
tile to advance his hand for say eight draws when Player B wins. Now,
is his 'attempted' hand counted as Thirteen Orphans, Greater Honours
and Knitted Tiles, or Lesser Honours and Knitted Tiles? It seems to me
that such statistics are not necessarily helpful.
Another situation: Player A may have been attempting to do Greater
Honours and Knitted Tiles, but decides to win on a less optimal tile,
to give him a Lesser Honours and Knitted Tiles. Does his attempt at
Greater Honours and Knitted Tiles count as a statistic, or is only his
final winning hand more important? The same situation could be applied
to a player whose hand is a potential Big Three Dragons, but he wins
with a Small Three Dragons instead, because he suspects one of his
opponents are holding onto the remaining two Dragon tiles and he
decides to win with a cheaper but guaranteed hand. His winning hand
can be recorded as a Small Three Dragons, but is his attempted Big
Three Dragons counted?
Unless we adopt the complex recordkeeping like what Benjamin is
saying, it does seem very impractical to expect attempted hands to be
recorded (because such hands probably need a lot of comments/
annotation/notes), and are not neat and pretty statistics anyway. No
one can argue that winning hands cannot be classified since winning
hands are not possibilities (unlike attempted hands) but definite.
Cheers!
Edwin Phua
Tom Sloper
> then I apologize, but from your own words it seems you're generalizing
> to ALL tournaments.
All large tournaments (of, like, 100 or more players). There's a size
beyond which it becomes impractical, but I can't pinpoint the number. I
think it's 100, you say it's 136, let's call the whole thing off.
> The original question was about statistics but your post seemed to be
> entirely about recordkeeping, hence my reply (and topic change) on
> recordkeeping.
I drew the connection between recordkeeping and statistics. ME! MY fault!
Don't try hogging all the blame here.
> probably impractical in general so maybe I have to disagree with
> myself that I'm disagreeing. (Do you agree?)
Um, sure.
>>> All pros are trained to do this sort
>>> of recording and it is not difficult to learn. (Anyone CCed to this
>>> email can teach you, Tom).
> I wonder though, how common it is for players on this newsgroup to
> take notations of a *complete* round for all players (discards
> included). Anyone?
OK, so either everyone reading this post can teach me, or very few of them
have actually done it. Gotcha. Hey, ain't I anyone? (Answer: yes, I am, but
I haven't notated a complete round without tile images - would have to work
my poor little head to figure out a notation to do so.)
> I don't have the book on me right now, but the diagrams are pretty
> easy to spot. ... If you see a picture of Ryan Morris
> (the only non-Japanese) smiling in a diagram, you're in the right
> spot.
Then perhaps you're talking about page 149? Smiling Ryan on that one.
Don't see any notation there - just hands formed with tile images, with text
labels on each hand. Starting 30 pages later I see rows of tile images, with
arrows - that looks interesting.
BTW, I did see those circled numbers representing dots, kanji #s
representing craks, and Arabic #s representing bams, in Ryan's translation
of the strategy chapters. Hmm, I just went to
http://museum.takeshobo.co.jp/kokusai/index.html and I see tile images
instead. Honest, I did see that notation in there at one time (perhaps
before it was put online). And it's used in some books in my collection.
> Happy birthday, MJnewsgroup.
Yippee! [toot!] Banzai!
>>> But for the WSOM, it may
>>> actually be possible since it is structured as an elimination
>>> tournament.
>>
>> The Chengdu event? I hadn't noticed that in the signup materials.
>> Interesting.
>
> I think we both know which tournament I'm talking about. (The WMO-
> WSOM contains tournaments other than Chengdu so I think the nitpick-
> factor favors me here.)
You mean, people have to qualify in other events before attending the final
event? OK, that part I knew - guess I got confused. I confuse easily,
especially when everything is brown (even when it's written in white, or
black on white).
>>> By the 2nd round, half the players have been eliminated
>>> so there are enough recordkeepers to go around. Of course, there are
>>> complications. Making sure that eliminated players would be willing
>>> to do this and that recordkeeping was accurate would require a good
>>> deal of preparation but I think it's entirely possible.
>
> I'm surprised you didn't reply to this paragraph.
I'm surprised you're surprised. My not responding was due to the fact that I
found nothing to argue with. Sure - anything is possible. And sure - there
are complications. So? (^_^)
>Im interested in
> other people's opinions on this, especially Alan's. If you're going
> to eliminate players every round, it'd be nice to put them to good
> use, right?
Sure. In fact, we should also have them act as referees, ushers/security,
commentators, and garbage can emptiers. Of course, there are
complications - like what if someone says "I don't wanna"?
> Even
> if you got the game records who would want to go through all of
> them?
Someone who wanted to create the kind of statistics that were inquired about
originally, instigating this thread. I sure wouldn't want to do it, but I
suppose I could be persuaded...
> Oh, but I think we still do disagree, both in our individual points
> and in terms of the discussion as a whole. So I disagree with your
> disagreement that we disagree.
But if I agree, then we really do disagree. So if I disagree, then we really
do agree? I'm so confused. (-_-)
Dewa
Tom
Tom Sloper
> As for the winning hand versus attempted hand, I am skeptical how
> practical recording attempted hands is.
It definitely multiplies the challenge. The reason I brought that up is
because this question - about statistics regarding what hands to go for,
what the chances of success in attempting a particular hand - has come up on
other forums at least twice within the past couple of weeks. So I wasn't
only addressing the question as it was worded here - I was also addressing
it as it's come up elsewhere recently as well. The only way to come up with
reliable statistics of that sort would be by increasing the amount of data
collected, and that data can't be solely about the hands that won. If I'm
wrong, maybe someone more mathematically minded can correct me in this
assumption.
> After all, non-winnings hands
> may not have any clear patterns compared to a winning hand.
Sure. The simple solution is to add two "hands" to the checklist: "multiple
options" and "a mess of unrelated tiles."
For that matter, how do you record a hand that was All Chows, Pure Straight,
Mixed Double Chow, One Voided Suit (for example)? List all those fan for the
hand, or just list the highest-scoring one? The database would need to allow
for multiple patterns and characteristics for each hand, in order to get the
most complete data from which statistics could be gotten. And of course
statistics that apply to MCR wouldn't apply to Zung Jung or Riichi/Dora or
HKOS.
Tom
Archon_Wing
wrote:
> "Edwin Phua" <fant...@pacific.net.sg> wrote
>
> > As for the winning hand versus attempted hand, I am skeptical how
> > practical recording attempted hands is.
>
> It definitely multiplies the challenge. The reason I brought that up is
> because this question - about statistics regarding what hands to go for,
> what the chances of success in attempting a particular hand - has come up on
> other forums at least twice within the past couple of weeks. So I wasn't
> only addressing the question as it was worded here - I was also addressing
> it as it's come up elsewhere recently as well. The only way to come up with
> reliable statistics of that sort would be by increasing the amount of data
> collected, and that data can't be solely about the hands that won. If I'm
> wrong, maybe someone more mathematically minded can correct me in this
> assumption.
>
I shouldn't have written the word detailed, but something like a
snapshot of each completed hands would be good enough for me. I was
really just intrested in the frequency of small hands vs big hands as
well as what patterns go out. Of course, I am no statistics expert but
I think that the winning hand would represent the most concrete, the
most quantifiable part of what happened. Trying to figure out
incomplete hands doesn't seem like a very good idea because we can't
read people's minds. Certainly this isn't perfect, but it would
definitely be "good enough" for us to have an idea on how balanced or
imbalanced the ZJ system is in practice regarding various situations.
Tom Sloper
> ... something like a
> snapshot of each completed hands would be good enough for me. I was
> really just intrested in the frequency of small hands vs big hands as
> well as what patterns go out. Of course, I am no statistics expert but
> I think that the winning hand would represent the most concrete, the
> ... Certainly this isn't perfect, but it would
> definitely be "good enough"
I learned a long time ago, in remaking Shanghai every two years and in
writing the FAQs, that what's good enough for one person is very often
decidedly not good enough for another. It's impossible to satisfy everyone.
But in devising a system for wide use, one must consider how best to satisfy
the majority. I wrote what I have written on this subject based on my
experience, my own judgment, as to what the majority would want and expect
from a system of statistics.
I would be extremely happy to have to record no more than only the winning
hands in such an endeavor. I'm not yet convinced that the statistics that
could be derived from that recordkeeping would be satisfactory to the
majority of players.
Then again, it may be that the majority of players are already satisfied,
with not having any hard data at all! (^_^) Maybe it's only a small minority
who think these statistics should be available right now!
Cheers
Tom