Reachmahjong.com

Here is this week's みんなで何切る？from the Mahjong Fight Club website. I'll post a translation of the pro commentary when it's released next Monday.

It leaves the widest path to tenpai. 357s are all bad discards since they kill two kanchan waits. 9s is ok, but it still loses 3 tiles to tenpai, all ponnable and mostly yakuhai. All 1s loses are two 2s.

The only alternative I can see is the East, but that leaves the hand in a pretty restricted shape with kanchans for 2s and 6s. Going chinitsu about doubles the value of the hand, but it also about halves the wait. I'd settle for the honitsu.

seems good but why not the ? If you can draw either of , you are tenpai and can riichi for seven pairs.

You are indeed right. I was thinking from the 13579 point of view, so 3s discard registered in my mind as "24s become backfire". But they don't because chiitoitsu. Even 7s gets added. It's the best move then.

is my choice. The only tiles you can’t use afterwards are the 8s and another 7s. You keep the iishanten for seven pairs, and also ishanten for ipeiikou. Most other bams will change your hand shape and which yaku you’ll get but you’ll still be ishanten.
or lets you drop the 2s or 4s and be ishanten for san ankou / toi toi.
or gives you an ipeiikou tenpai by cutting the non-red 5s or the 1s. I’d take that over seven pairs as you have a chance at 3 han over the 2 you’d get from seven pairs if you get East as your winning tile (on top of the honitsu, and reach as well if you reach) and you’re waiting on three tiles either way.
doesn’t get you to tenpai, but gives you a shot at ryanpeiikou, albeit with two kanchan waits. You could also couple it with chinitsu as well if you want to risk that and cut the East.
is probably the worst of the lot as you stay at ishanten without any increase in possible yaku. Most likely cut the non-red 5s and hope for a 2s for ipeiikou or just tsumo-giri it.
or give you the choice between either the ishanten san ankou / toi toi option or the tenpai ipeiikou option.

Also Iapetus: I agree settling for honitsu is better, as going Chinitsu actually wouldn’t necessarily double the value of the hand. If you stay honitsu and pon the East you still have honitsu, double East, and 1 dora for mangan, (and you can in theory aim for toi-toi to finish it to go even higher) while open chinitsu would only be 6 han (with the one dora). So unless you stay closed and add another yaku, it’s only a 50% increase (12,000 to 18,000).

I am amazed at how you can come up with all of this in the time allowed to make a decision on what to discard. I would need about an hour, with paper and pencil.
Seriously, I am impressed with your tenacity and diligence.

or2az wrote:I am amazed at how you can come up with all of this in the time allowed to make a decision on what to discard. I would need about an hour, with paper and pencil.
Seriously, I am impressed with your tenacity and diligence.

While I made my initial decision on the 7s in a short period of time, though likely slightly longer than the standard discard window (said window varies a lot depending on where you play) and not all of the above information would have been actively considered in that time. And I’ll admit that when I made the decision I did miss the seven pairs option.

One of the issues with WWYD problems is that they don’t really represent an in-game situation, as you have as much time as you’d like to think about your decision. But on the flip side of that you’re missing the greater understanding of the hand that comes from actually playing it from the first turn. As happens with all games and sports, those who are actually in the driving seat at the time of the decision will likely be seeing things very differently from an observer. With Mahjong in particular you’re usually considering you hand while the other players are drawing and discarding, so you’ll have a rough idea of the various possible shapes that could arise even before the next tile is drawn. This means that you can select your discard a lot faster than someone who looks at the hand as a WWYD problem, since they will have to assess the current situation of the board, and figure out the shape of the hand from scratch.

With WWYD problems I feel you can approach them in two different (and not necessarily mutually exclusive) ways. The first is to treat it purely as a discard question. You’re playing the hand and this is what you’ve drawn, now you need to make your discard so the game can continue. In this case you try and pick a discard quickly as most games you will not have huge amounts of time to think things through. Though the chances of being able to answer in the actual time required in a game is relatively low due to the stuff mentioned above. The other way to view WWYD is as more of an analysis problem and take the time to really figure out what the best discard would actually be. You look at all the possible discards, look at the different shapes the hand could become, and consider future draws and how you’d deal with them to hopefully complete the hand. This helps you to learn to read / assess a hand faster in a game as you’ll start recognising certain patterns in the tiles, and make your own shortcuts for reading. Eg I have trouble reading runs when there are pairs or even triples involved. For example 233445 or 23334445, but have managed to figure out that the first shape is two complete runs, and only one tile off ipeiikou. While the second is 2 runs and a ryanmen wait on a third. In the past I always had to look at the shape hard and figure out: “ok so that’s 234, and 345 so what’s left is a 34 half finshed run” where as these days I just see the shape and recognise that the middle two tiles will be a ryanmen while the others make up two runs. I guess it’s similar to reading a language where you have to spell out each letter to understand the word and where you can read the word as a whole. Eg I still have to sound out Katakana words when reading them for example, but some tests show that people can read English even if the letters are jumbled up, so long as the first and last letters are correct.

To get back on track / conclude: A lot of the information above I won’t have consciously considered when I made my decision. I went with what looked at the time the most flexible discard in terms of using the bams below 9s, and knowing that I would want to pon the East if it came since it is worth 2 yaku. As a result I would hopefully not be able to use the East as the head, so keeping the 9s as a pair helped sort that out. It is likely that if you were playing this in a real game you would have considered a lot of the possible draws already and some draws don’t affect/interfere with each other (eg drawing a 2s would likely result in the same action regardless of whether it happened before or after the 9s draw, if that makes sense) meaning that considering it a few turns ago prepares you for when it actually happens, even though some other parts of your hand may have changed. So while I made my choice back on Tuesday (right after I posted this thread) I only had a very vague reasoning behind my choice. So today I did the second method of approaching WWYD problems I described above and went through and actively considered the problem in more detail, looking at the possible shapes and yaku that could come from the hand. It’s quite possible that during this stage that I’ll discover that I was wrong with my initial discard choice and that there was a better option that I didn’t see at the time. In this case though, I stand by my choice. I’m not saying it is definitely right, but it’s still what I’d choose as my discard. So while I try to make my initial selection is a reasonable discard time frame, I will always go back and figure out the details of what the discard would mean for the hand and whether there might have been a better option.

As usual, a very good explanation, and I understand what you're saying, because I didn't realize at the time I made my decision, until you mentioned it just above, that ponning EAST would be worth 2 yaku. I've been approaching WWYD by your method one, game time decision, as opposed to situational analysis.
For this particular problem, I first had to decide whether I would open the hand or stay concealed, and proceed from there. Sometimes, even that's not so easy.
How long have you been playing? (found it, April 2013, really??)

or2az wrote:As usual, a very good explanation, and I understand what you're saying, because I didn't realize at the time I made my decision, until you mentioned it just above, that ponning EAST would be worth 2 yaku. I've been approaching WWYD by your method one, game time decision, as opposed to situational analysis.
For this particular problem, I first had to decide whether I would open the hand or stay concealed, and proceed from there. Sometimes, even that's not so easy.

I try and do both methods these days, for the reasons I mentioned. But to be honest a lot of what I said above I hadn't consciously thought about till I was writing the post. I just knew that I hated the first 5-10 seconds of looking at a WWYD problem because I always felt so lost as to what was going on and pressured as I knew that in a game I would have to make my decision quickly (the first method I described). It's only really been since I've been posting WWYDs here that I've been doing the second method in as much detail as this, and also actually writing it down rather than trying to do it in my head.

The first thing I always do with WWYD is check the round number, my seat allocation / score, and the turn number. So straight away I know which winds I can use to make Yaku and which I can't, as well as how creative I can get with long shot waits / yakus. Eg if it's still early game (first 6 turns or so) then kanchan and penchan waits aren't as much of a worry as they might be if it was more like the 12th+ turn, where the ryanmen wait is usually going to be king.

Actually thinking about it I'm not sure whether I really consider whether to open or not directly. I look at the possible yaku and then from there decide if and where I want to open. Eg for the above problem I would probably only open on an East pon at this point. And then only look at pon/chi -ing bams if after a few turns I'd drawn nothing. Also you lose the chance at ipeiikou or seven pairs if open, so as there are still yaku possible that required a closed hand I think my mind didn't consider opening on anything other than the East which would would make up for any of the lost ones since it's worth two han in this case.
This case is also a bit special since it's not hugely common to have a closed honitsu hand, usually I find that you'll open to complete a pon or a chi before you get rid of all your other suits.

So I guess what I'm saying is don't worry about staying open or closed, aim to stay closed unless you can make a roughly equal score by opening. (or you have some really bad waits that you get a chance to call to complete) I actually had some issues a few months back where I realised I called for a lot of my hands. As in A LOT. MFC still actually thinks I do but I read an article on this site actually that really stuck in my head. http://reachmahjong.com/en/2011/06/half-flushes-part-1/ Funnily enough it's about half flushes But it was more Garthe's comments at the bottom that stuck with me, in particular the men-tan-pin mantra. The previous article he mention he also talks about it (http://reachmahjong.com/en/2011/05/yaku-hand-points/). I'm not going to rehash what Garthe said in the articles since he's said it better than I ever could and you're better off reading them yourself, but those comments actually had a huge impact on my play. Instead of looking to open to complete a hand, I started trying more to just stay closed and NOT open unless I absolutely had to. This also opened (no pun intended) me up to more expensive hands as I had more access to the yaku that required the hand to be closed, and the higher values for the ones that could be opened (ikkitsukan, chanta, jun-chan etc). Before that point I very very rarely had pinfu in my hands, and actually usually forgot about it when scoring (and it was a nice bonus if it happened to come up when playing on a video game), where as these days it's one of the first yaku I consider if I don't see a double of a wind or dragon I can use.

To be honest I've learnt a lot about how I think about my hands just from writing the last couple posts And in all honesty posting these WWYD problems and answering questions when I can elsewhere on the board is partially for my own benefit as well as they force me to actively think about the problems and what I want to say and then be able to write it down in a coherent manner that people other than me can understand. I also get to see how other people think as well and hopefully learn new things that way too.

or2az wrote:How long have you been playing? (found it, April 2013, really??)

Yeah it's been about that long. It was the Easter long weekend (so March 30th or so I guess?) I think, that I actually sat down and watched HanaYoriUta's videos on YouTube (http://www.youtube.com/playlist?list=PLDC9453A66D5D2CA7) and found the flash game on gamedesign.jp (http://www.gamedesign.jp/flash/mahjong/mahjong_e.html).

I'd discard both tiles and go for Chinitsu. There's still a lot of unplayed sou tiles in the discards. The only real question is how far along the other players are, North's pile indicates that he's either in tenpai or close to it

"Hey Ozball! Have you forgotten to post the Pro answer for this?!"

Nope, I'm just blind and didn't read the line at the top saying that the answer would be released on the 27th. Looks like MFC does these in roughly 10 day rotations rather than weekly. Anyways, here are the results:



Here are the results from other players.



Pro's Response:
***Read Masa's Translation in the next post. It's more accurate than the translation below***
"I’d cut the 1-sou. My first instinct would be to cut the 7-sou and aim for Suuankou, while cutting just 1 of the 1 or 9 sou would make Suuankou too difficult. Because there are pairs like the 1-sou and 9-sou, if you pon them, then East becomes the obvious discard, and makes winning the hand a lot harder. At this stage the only trap is the double East. If you can overlook Suuankou and those higher scoring hands, and be happy with a maximum of 8,000 all, then a balanced play style is the best for the highest chance of winning."
Sugiura Kansuke Pro – 4th Dan

I'm not entirely sure about some of my translations in this one, particularly the first sentence about suuankou. Also the first half of the last sentence felt a bit wording wise. So if anyone has any suggestions for improving the translation, let me know.



Finally here is a break down of who chose which option based on their MFC play styles.
Brief run down of the styles:
Dark Blue: Lots of dora in their winning hands.
Green: Very low average of dealing into other people's hands.
Light Blue: High average han value in winning hands.
Red: High hand winning percentage.
Brown: Total.

He says.(I don't translate as it is, I make complement his sentence)

I would discard 1b.

My first impression was aiming suuanko. but it would be hard to do due to one 1b and one 9b can be seen in the pond already.
If I make pon for 1b or 9b due to last tile for them, Other players notice that I'm aiming double East(2 yaku) to check my pond. No one discard East. then the probability of winning is going to less.
So that I would discard 1b to make wide acceptance and proceed with concealed hand. At this moment what I should make open is only East pair.
Suanko(or pon hand) can not be made, but if I can expect 8,000 from each as the maximum, 1b is best balanced tile to discard to have secure winning and wide acceptance.

Remarks:
He is saying "四暗刻等の縦の手役" . If you translate "縦の手役"into English, it going to "vertical hand". But in mahjong "縦の手役" means "pon hand"(like suuanko, sananko,toi-toi)
Other side is "横の手役" which is "horizontal hands". it means "shuntsu hand"

And his first name is not Kankai but Kansuke.

Thanks Masa! Some bits of my translation didn’t quite feel right, but couldn’t figure out what exactly. The name was tricky as well as I wasn’t able to ask my workmates this time to double check I had it right.