Tenhou (Mahjong): Promoted to 6th dan for the third time. Using the number of successful sequences as a basis for deciding whether to push or fold.

After dropping to 5th dan in Tenhou around October 2018, I went about six months without playing.
I was busy with work, and I also wanted to solidify my fundamentals, so I diligently reviewed books like the Uzaku book.
Then, around April 2019, I gradually started playing Tenhou again, and after 111 games in May and June, I was able to be promoted back to 6th dan.
I'd like to reflect on my performance and see what areas I've tried to improve.

Points of concern regarding my detailed performance record

If we start listing all the details, there's no end to it,What I'm most concerned about is the rate of open tiles.
I'm aiming for a 35-40% chance of opening a tile per hand, but I'm currently at 33.2%.
The rate for parents is around 37%, and for children it's 31%, so I think it might be just as important to go early for children as it was for their parents.

I believe that the low rate of calling tiles might be due to the slow transition from a closed hand to a hand with open tiles, and the slowness in aiming for a formal tenpai.
自分の放銃率を見る限り守備型というわけでもなさそうなので、もう少し子のときにも攻めなければと思っています。
The number of games played varies considerably, around 100, so it might be difficult to draw conclusions based solely on this.
However, since the average rate of open calls (melding) at the 7th dan level and above is about 35%, I think that at least that level might be necessary.
(Based on the "Average Match Results for Ranked 4-Player Matches" on the Tenhou official website)

However, since I only play East-South games, if a large proportion of people mainly play East-only games, then perhaps we don't need to directly compare the rate of calling tiles.
(In East Wind games, the game is shorter and more decisive, and even a 1000-point win is considered more valuable.)

Thinking about when dealing tiles

The way you clasp your hands is like a living thing; there isn't really a set way to do it.
However, when I receive my initial hand, I try to get a rough idea of whether I can go for a Tanyao-type hand or if I have no choice but to go for a Yakuhai-type hand.
If there are many middle tiles among the number tiles, or if it looks like you can make a two-sided reach, you should clear away the non-wind tiles.
Even then, I do check for overlapping honor tiles, but I generally aim for a simple hand with open tiles.
If you get a pair of honor tiles, you have the option of using them as honor tiles or as safe tiles in the later stages of the game.

Furthermore, if there are many 1 and 9 tiles in the numbered tiles, and it is anticipated that declaring Riichi will result in a weak hand, the value of isolated honor tiles increases.
Furthermore, even if you don't get any matching honor tiles, you should still try to create hands that will at least increase your score, such as dora, dora soba, red tile receiving, sanshoku, and ikkitsuukan.
Even if you can't make a good-looking reach, you can still push if your hand is high-scoring, so I think it's important to create high-scoring hands at the expense of a slight delay.

In a regular game, it's necessary to get a sense from your hand and think, "This doesn't look like I can win," but I still feel it's quite important to try to win anyway.
By making a move early on, you can put pressure on other players and thwart their opportunities.
With a weak hand, it's difficult to push when another player has declared a reach, but the goal is to develop a sense of balance in how close you can get to winning even with a weak hand.

Reviewing the push-pull strategy

When your opponent has declared a reach, the question of what kind of hand you should have to be in order to push back has been largely established in various strategy books.
However, I hadn't quite settled on a strategy for myself, such as how to push or play from one shanten onwards, or even before that.
I'll re-examine the thinking process I adopted when I dropped to the 5th dan this time.

- Is that hand worth pressing?
If your opponent appears to be in a ready hand early on, you often need to proceed cautiously, even if you don't completely fold.
しかし、自分の中で回すことにこだわってしまい、筋引っ掛けや、ダマテンの他家に刺さってしまうという勿体ない放銃があることが気になっていました。
When I felt that attacking directly was risky, I realized I needed to be more aware of how many points my move would ultimately be worth, as well as the progress of the game and the score situation.
I've become much more aware of what rank I would drop to if I made a mistake, and whether it would be a number of games that I could recover from.
If my parents are still in the running, I still have a chance to make a comeback, but in the later stages of the game when that's not the case, I've become much more cautious.

Also, especially towards the end of the game, even if I only score 1000 points, if the point difference between me and the other players widens to 3900, 8000, or more, I've come to find value in that.
For similar reasons, I am very aware of the risks of putting out a reach stick in a close game, as this could narrow the point difference.

• Suji Counter Theory
I learned a method of counting the number of tiles that have been played, mainly in the early stages of the game, when I have a hand that seems to be worth a decent score, but my opponent has declared a reach, making it difficult to decide whether to push or fold.
First, we consider there to be 3 types of sequences, 6 of each (14, 25, 36, 47, 58, 69) = 18 sequences in total. We count the sequences that have been played and estimate the risk level of the remaining unsequential tiles.

For example, if the discards of the player who has declared riichi include number tiles 1, 4, 6m, 1, 9p, and 2s, then the number of sequences would be 6.
残りのスジは12本ですから、残りの通っていない無筋の123789牌を捨てるときの放銃率は、単純計算で、12分の1（約8.33％）です。
A 456 without any strings carries the risk of hitting two strings, so the probability is 2/12 (approximately 16.7%).
（実際は、同じ無筋でも19と28と37では放銃率が異なりますが）
In other words, the more tendons that have passed through, the more dangerous the remaining tendons become.
ですので序盤にリーチを掛けられた場合でも、スジのカウントが9本くらいまでなら放銃率は10％前後なので、そこそこ押せる場合があります。
Even so, other calculations are still necessary, such as whether or not that hand is the right one to push with, but I felt that by incorporating these kinds of judgment criteria, it would be possible to make more precise decisions about pushing and pulling.

There's a YouTube channel I've been watching for a while now that I find very educational, and it's called "Unimaru Channel."
I've been using their live commentary and analysis videos of Tenhou (an online mahjong game) that they've posted on Nico Nico Douga for some time now as a reference.
The following video is very helpful in understanding how to apply this counter-strategy theory in actual gameplay.

This video explains how to make very subtle decisions about pushing and pulling.
For now, I think it's enough to just focus on the key points you need to remember to put them into practice.
Incidentally, this channel features many videos aimed at players of various skill levels, making it both entertaining and incredibly helpful.

In conclusion

I've been browsing various blogs and other resources online written by Tenhou players, but most of them are high-ranking players, around 8th to 10th dan, so I don't really get a sense of what it's like for players like me, at the 5th or 6th dan level.
However, the first thing I can do is to learn from people at a higher level than myself and put that knowledge into practice.

While you can download game records from advanced players on the Tenhou official website, you won't know their intentions behind discards or calls unless you ask them directly.
In that respect, I found Tenhou's live commentary videos to be very easy to understand, as they clearly show what decisions are being made in real time, what factors are being considered, and what precautions are being taken.
For people like me who are struggling to improve their performance, I think it's important to actually put new knowledge, techniques, and decision-making methods into practice in real-world situations.