game recording for analysis. Good or bad idea?

[Benjamin]

* I\'m not trying to imply that doing well on a free internet site is much in the grand scheme of things.

I certainly don\'t think I am god and it seems that most would agree you\'re the one coming off more that way at the moment! You have to forgive us for that.

The comments Hirohurl and I wrote weren\'t directed at you. I honestly think the video you made is great; it\'s easy to understand and it\'s a good introduction to pushing/pulling. I don\'t agree with all of the content, but that hardly matters since I didn\'t take the time to make my own video or post a detailed reponse to yours. That was the point I was trying to make.

I certainly don\'t hope I come off as God and I\'m sorry if what I said sounded condescending. I certainly don\'t consider myself an expert player over here, and even if I did it I don\'t think its that important since I don\'t put as much time into explaining the game to beginners as people like Jenn, Ryan (www.japanesemahjong.com) and, yes, you do. So keep up the good work.

[HotelFSR]

Yttrasil:

Sure, the 3s is good! I never claimed my choice was necessarily superior, just gave the reasons for it. I also don\'t remember from the video whether the 3s keeps tenpai or not.

I also definitely agree that what I show is a bad strategy for a beginners table, but I had to make do with it for the video. I didn\'t want to risk playing and talking in a rated game, it\'s very hard to even think straight. I\'ll do that another time on an old Toupaiou account.



Why are you getting so worked up? I am not out to get you. You must have a rough time making a living at poker if you find it so difficult being dispassionate...

It does make me wonder how good your poker game is.

I\'m not an obstinate person. If you look at my post history on this site you will see that my thinking about this game has evolved radically since I joined.

I\'m not sure why having a calm logical discussion seems so painful for you and you have to resort to caps lock insisting that \"I just won\'t get it\".


Why the attitude from you? Notice that I never said that my choices were better than yours! I simply gave you the rationale behind mine, and why I chose them for the sake of the video. I never claimed that these choices were superior.

I was trying to force a certain style on a random hand just to illustrate a certain point. Nobody should look at it as the perfect game by any means.

I suppose I need to make better organized video topics and take the time to base them off good replays to avoid offending you!


By the sounds of it you would also be offended by some of the play you\'ll see on pro replays which many of us often use to study from, because they don\'t always follow the maximally safe path. If they did everyone\'s game would look identical at the top levels and nobody would have an edge to speak of.



P.S. The god comment was not directed at you at all Benjamin!! It was for Yttrasil. You have been just great.

[Benjamin]

Also do agree with you there Benjamin but also hope you will go watch the video on those parts I commented on, it would be appreciated to clear things up.

Thanks for a details and levelheaded response :) I guess if I\'m pulling a deus ex-machina here (am I really, Hotel?) I should comment on the details too.

Commenting on HotelFSR\'s comments regarding reads and bluffs is not easy for me to do, because for the most part I just don\'t use these concepts when I play. In general, the two most important concepts in playing a hand are balancing speed/value and knowing when to fold. If I\'ve already decided to fold it rarely matters what I think the other person is waiting on.

Bluffing doesn\'t work most of the time for the same reasons. Either the other person is going to push his hand anyways, or he\'s already decided to fold. Mahjong is unlike poker in this regard.

There are situations where I try to mawasu (walk the line) but they\'re not very common. It\'s interesting to try to put someone on a wait (and I think Hotel\'s call in his first video was impressive) but this is isn\'t reliable enough to be very useful. It\'s much more useful, for example, to put someone on a *value.*

By contrast, Yttrasil bases most of his arguments on concepts I think are sound, so I tend to agree with his arguments. Things that ELIMINATE possible waits, like kabe (4 7-dot out makes 8-dot safe), cheap/expensive suit (lots of bamboos out means no one is using them), and suji (1 & 7 make 4 safe) are much more reliable than trying to put someone ON a wait.

Just because I tend to agree more with one of you doesn\'t mean he\'s right or that I\'m right. It\'s just my opinion and I gave it because it was asked for. I really do think it\'s great that that video is online; back when I learned Japanese Mahjong in 2004 there was nothing like that and it would have helped me a lot!

[Benjamin]

Ah. Well it seems I misread your comments. I guess that tends to happen when posts get a bit emotional. I do wish you guys would turn it down on the ad-hominem stuff. You both are posting because you really like the game. If you feel that the other person ignored a point you made, then that means they didn\'t rebut you and most people who read the most will think you won the argument :P

Oh, and I\'m directing that at both of you. And really, myself as well, god knows I\'m as guilty of getting emotional about internet stuff as anyone. And as long as Yttrasil mentions it, HotelFSR\'s reasoning used to drive me a little bit nuts too. But he\'s good about posting his thoughts and he\'s right that his thinking about the game has changed, so my attitude towards him has changed too. But I still disagree with a lot of what he says :)

Just a suggestion--for your next project, why not pick an interesting pro game and try to annotate it? That way you have a good measure of confidence that the play was top level.

[HotelFSR]

Best suggestion yet.

For the next video I\'ll annotate an online (Tenhou) pro replay.



I definitely agree that you don\'t necessarily have to wait-read or bluff to get to a high level. I was also pretty skeptical when I first read about these kinds of strategies.


However, I tried a little experiment to test it out that converted me somewhat on this point.

I watched an absolute ton of replays while trying to read the discard pools (with the hands open so that I could always check my answers). I started to notice that reading could be remarkably accurate if you took into account all the factors available to you (dora tile/suit, number of turns, suji, hand discard/tsumokiri, tile order, reach-tile suit). I learned a lot I never expected by looking closely at what the discards reflected about what was going on in the hand (viewing it open during the replay).

I started making it a habit to memorize these things and take note of other players\' actions during play, and it\'s something I\'m still working on. I still practice by \'reading\' as many replays as I can get my hands on, which has helped enormously and is a lot of fun. Try it!

It has definitely paid off more than I expected. I\'ve become able to hit waits all the time. I rarely fall into two-sided waits anymore even when pushing. I also can detect closed waits a fair bit, which I\'ll talk about in another post. Reading waits, final ones or sets, is not nearly as difficult or unreliable as it may seem.

You will naturally fail a fair amount, but it will succeed more often than it fails and it can really be worth it. It\'s just another tool in your box for when you really need to push or pull, and can be used either defensively or offensively.

You get to know what sets your opponents are making/waiting for, and thus what tiles they are holding that you might want, even if you don\'t necessarily get the final wait nailed (or decide not to go with your final read). This has a lot of value.

The only thing I have absolutely no real way of detecting (although I want to learn it and I\'m sure it can be done, I think Konno even wrote about it) is detecting shanpon waits. I figure though that my rate of success in reading is still high enough to justify using it, even if I screw up on pretty much all the closed shanpon waits I get.

Tanki is obviously pretty much impossible to stop but I do sometimes hold on to lone honors and terminals late game if I have decided to fold.



As far as bluffs go, I also find they often work well- especially against very defensive players. I sometimes check player stats in-game and try to determine if they play this way. There\'s also value in making a player doubt themselves. I think bluff plays can work well in Mahjong. Imagine yourself on the other end- would you play a little differently if I made a suspicious move?

Probably.

Whether a bluff will work also depends on how late it is in the hand, and whether other players look weak enough.

Obviously, all this depends on your style. I personally have chosen to take the path of learning and trying to perfect a more read/bluff oriented style for the simple reason that I find it more fun and exciting!

There\'s a great live replay where Takeo Kojima detects and clearly avoids a fully concealed dama-ten shanpon wait (no reach). Not sure how he does it.


I wish I could get my hands on more poker-style literature that discussed these kinds of things in depth and with detailed statistics, commentary and situations.


ALSO:

(from Ben)

There are situations where I try to mawasu (walk the line) but they\'re not very common. It\'s interesting to try to put someone on a wait (and I think Hotel\'s call in his first video was impressive) but this is isn\'t reliable enough to be very useful. It\'s much more useful, for example, to put someone on a *value.*

True, value is always more important and more useful.

Still...

Especially in inflated mahjong I do feel that wait reading has its place because low value hands are not only rare, but rarer than big hands. So while you are going to fold more than you are going to push, it\'s a good option to have- and possibly slightly more reliable than you think. In my videos alone I pretty much hit them all (even if the play isn\'t ideal or the read isn\'t needed in those cases).

I also find it can often be harder to read value than to read waits, and either way the two are closely related (e.g. detecting dora use).

[Benjamin]

There\'s a great live replay where Takeo Kojima detects and clearly avoids a fully concealed dama-ten shanpon wait (no reach). Not sure how he does it.

Because the other 999 times, he loses the hand? :P I\'m real skeptical about this stuff. The most famous defensive player in Japan is a pro named Igarashi and I\'ve won hands off of him.

Especially in inflated mahjong I do feel that wait reading has its place because low value hands are not only rare, but rarer than big hands

Rarer than big hands? Like there are more mangans on average than there are 1000-3900 point hands? I don\'t think this is true. Do you have stats to back this up?

[HotelFSR]

Ben:

Yeah, Takeo Kojima is a bad example. I take anybody who believes in \'nagare\' with more than a little grain of salt. I\'m not trying to imply that I\'m wooed by \'magic\' or that play has to be flashy to be good.

As for Igarashi, I\'m sure you and many others have won plenty of hands off him. Not sure that proves a lot. He only has to be successful enough to make his strategy worthwhile in the long run- which I would assume it should be if he is famous.

Although wait-reading I would often characterize as an agressive, not defensive, style. You can use it to fold or to attack. Not sure what Igarashi does or how it would classify.



Rarer than big hands? Like there are more mangans on average than there are 1000-3900 point hands? I don\'t think this is true. Do you have stats to back this up?

On the hand size, yes, as far as I know. 1-han is the rarest (by a wide margin), then 2-han, with 3 & 4 being the most common. For some players 4 han is even slightly more common than 3. Obviously we are talking about inflated games here.

There\'s a link on Tenhou I need to find for you where you can look at this stuff for yourself (sorted by lobby level). If I recall correctly, the numbers are something like:

1 han 10%, 2 han 20%, 3 han 22%, 4 han 23%, 5 han 18 %, 6+ han 7%

i.e. about 2/3 hands are 3+


You can also check on something like MFC.

Ultimately, the wait reading I\'m referring to is not about getting the lowest furikomi % (hands thrown %). It\'s about getting the highest possible hands won % with a low furikomi %, it\'s the gap between the two numbers that\'s the most important. It also lets you cut your opponents out of more hands, reducing their tsumo chances and win size.

[Ruro]

I\'m not really sure if i should make a video, my english IS bad, and i\'m a really low level player. As for these % things, i\'m pretty interested. Because it\'s seems that chance in mahjong is rather different from the chance factor in cards. What i mean is that when you pick a card from a deck you will almost likely not get the very same card in the 3rd or 4th draw after that, but in mahjong it\'s really common to draw the same tile that you picked last turn, is there any kind of \"special\" reason for it? And how do you even calculate for tsumos? It seems (almost) completely random for me.

[HotelFSR]

By all means make a video. Doesn\'t matter at all if your English isn\'t the best (although it seems excellent to me!).

I\'m not too sure about your grasp of statistics, though, if I understand you correctly.



There is definitely nothing \'special\' about mahjong that means that you are more likely to draw a matching tile per se.

The difference with cards is simply the number and type of cards of the deck. There are four suits, for example, whereas mahjong has three. Both games are of course governed by the normal laws of physics and probability.

Another thing to note is that in mahjong every tile has four copies. In cards there is one of each per deck, so in that sense you will draw more similar tiles in mahjong.


You can also calculate \'outs\' and chances in mahjong just like you can in cards. To figure out your tsumo chances, for example, just compare how many tiles are left in the wall (deck) with how my possible winners you have.

Then adjust the number depending on factors such as how many of your winners have already been discarded/used, how many turns are left, etc.


Mahjong for all intents and purposes \'is\' a card game.

[Yttrasil]

Just out of curiosity what do you play in poker HotelFSR? Here is your chance to earn my respect hehe ;)

[Tha_Reaper]

Ruro: make a video!
My commenting on my playing style also wasnt good, but helpfull people here still managed to help me a lot by giving useable tips, and pointing out what i should watch for.
Im sure everyone will have some flaws in their playing style that they dont know about. even the best.. you will only figure out those flaws by letting someone else analyse your style.

[HotelFSR]

Yttrasil:

Mostly holdem, mostly live games. I tend to play at the empire leicester square in london. Couple places in mayfair as well, where you find some very wealthy people playing poker very badly =)

Fortunately I have a good sized bankroll, partly from playing the stock market for the last decade or so (pre poker), having invested in videogame-related companies like nintendo and activision, which have done very well.

I\'m otherwise a researcher at UCL so my salary is pretty much nonexistent. I wouldn\'t be able to afford my current rent without poker, especially since I\'ve refused to allow myself to trade stocks for the past year or so- which has been a good call I think, considering the economy.

I\'m not really interested in getting into big poker discussions on this board because, frankly, I play mahjong to get away from poker.

[Yttrasil]

Okay ic agree on the mahjong part just want to know how far you\'ve taken your game, the rest is also quite interesting also I\'ve lived in Japan a total of 1.5 years so have been playing here some at least =)

If you play online what do you play and what limits? I play NL200 SH but can surely beat NL400 though I don\'t feel comfortable playing there so I rather not...

Must be tough I myself wouldn\'t play the stock market at all with any money I can\'t afford to lose think it\'s impossible without great knowledge of the market and different calculations that only firms have access to.

Edit: Also can add that I don\'t find poker fun at the least really but I find mahjong loads of fun... poker is for paying the rent only and giving me a decent lifestyle but also has alot of downsides for sure, studying in Japan at the moment... then there is mahjong for entertainment and I love the game still.

[HotelFSR]

I basically never play online, not seriously.

I tried it a little but I prefer live games, and the competition is usually easier as well. I also really like to act and look for tells, etc. so that goes out the window online.

As for the stock market, I definitely would not recommend it at the moment. However, once upon a time I think it was a perfectly safe thing to do. Personally I invested in videogames because it was an industry I know a lot about, I certainly wouldn\'t invest in companies I don\'t have knowledge of.

I don\'t think you need complex calculations or insider information to do well at stocks, especially if you are an individual with the flexibility to get in and out when you please. Detailed knowledge of a specific industry can take you a long way in good market conditions, I think.

[Yttrasil]

Hmmkey ic, very true that I don\'t really like live games with much money also it\'s too slow for my taste just one table but can understand why you prefer it. :blush:

Maybe so if you are very knowledgeble in a specific market but then again you are at a disadvantage compaired to the big companies imo.

Also edited my first prior post so if you feel like it you can check that out =)