This video was super interesting! I'm fairly new to majhong, and I'm only expert rank, but I think you played this hand as close to perfectly as possible. But in Washizu's words, "To play perfectly and still lose. Thats what mahjong is all about"
6:15 I actually thought about discarding 9s in this situation but I am not completely sure. If we're not calling kan immediately, then the 4th 9s can be seen as the worst isolated tile in the hand. Discarding 2m loses 3 tiles of acceptance over discarding 9s. Cutting 9s also gives the option to use 9s as a pair later if we draw 3m and need to fix 123m with the haku pon. Maybe I'm underevaluating the loss of 24/32 fu this causes, although kan is a double edged sword... This was very interesting, thank you for the video :D
In terms of the 7m vs 3p for Mortal at 7:05, I suspect it's because Mortal's analysis is mostly if not fully digital. If I'm not mistaken, Naga is trained on human play, which will inevitably have some analogue reasoning being used at some point, being incorporated into the models. Honestly I agree with Mortal more here. Even if we draw 4p, I value a 23p block more than a 466p block because waiting on a middle tile that can extend to dora (and the wait is not that good either) is kind of scary, don't think people will throw it. And as evidenced by early pinzu throws, if we riichi and someone pushes, it's not unreasonable that they don't need lower pinzu and will toss a 14p.
Thanks for the great video as always. At 5:19 you give a long string of unlikely waits quickly. Can you give a little more detail on your reasoning for those particular unlikely waits?
Do you mean 25P, 36S, 36M and 47m? These are the more likely waits. 36s is most dangerous because it is also a lot more dangerous than others to the right player.
@@MrFeng sorry for not being more specific. Just before that line you say the 8p and 7p are unlikely (then you mention the likely waits, then some more unlikely waits which you explain better). The 8 and 7 were the ones I was interested in understanding more about. You say it's due to the 2 9p being out, but I feel like I'm missing something obvious about why only the 789 or single waits are relevant Edit: nevermind - all 6s visible. Whoops.
This video was super interesting! I'm fairly new to majhong, and I'm only expert rank, but I think you played this hand as close to perfectly as possible. But in Washizu's words, "To play perfectly and still lose. Thats what mahjong is all about"
6:15 I actually thought about discarding 9s in this situation but I am not completely sure. If we're not calling kan immediately, then the 4th 9s can be seen as the worst isolated tile in the hand. Discarding 2m loses 3 tiles of acceptance over discarding 9s. Cutting 9s also gives the option to use 9s as a pair later if we draw 3m and need to fix 123m with the haku pon. Maybe I'm underevaluating the loss of 24/32 fu this causes, although kan is a double edged sword...
This was very interesting, thank you for the video :D
In terms of the 7m vs 3p for Mortal at 7:05, I suspect it's because Mortal's analysis is mostly if not fully digital. If I'm not mistaken, Naga is trained on human play, which will inevitably have some analogue reasoning being used at some point, being incorporated into the models.
Honestly I agree with Mortal more here. Even if we draw 4p, I value a 23p block more than a 466p block because waiting on a middle tile that can extend to dora (and the wait is not that good either) is kind of scary, don't think people will throw it. And as evidenced by early pinzu throws, if we riichi and someone pushes, it's not unreasonable that they don't need lower pinzu and will toss a 14p.
Thanks for the great video as always. At 5:19 you give a long string of unlikely waits quickly. Can you give a little more detail on your reasoning for those particular unlikely waits?
Do you mean 25P, 36S, 36M and 47m? These are the more likely waits. 36s is most dangerous because it is also a lot more dangerous than others to the right player.
@@MrFeng sorry for not being more specific. Just before that line you say the 8p and 7p are unlikely (then you mention the likely waits, then some more unlikely waits which you explain better).
The 8 and 7 were the ones I was interested in understanding more about. You say it's due to the 2 9p being out, but I feel like I'm missing something obvious about why only the 789 or single waits are relevant
Edit: nevermind - all 6s visible. Whoops.
Why didn't you take the win on the 1 man? 5:55
he couldn't, his hand had no yaku with 1 man. He could only win on a 3rd haku.
@@kas00078 ohhh i see