I am a beginner player, mostly studying tsumego and playing less that I would want.
As such I am a bit lost in the “global moves” department.
This is a wonderful example: a game which I am “winning” against a bot epanizon vs. amybot-ddk . (I resigned for reasons unrelated to the actual game).
The problem is that while I knew I was in a good position (later I learned that the AI optimistically says 99.7%!), I did not know exactly what I should have done. How should I go to limit the potential of white on the bottom half?
My defensive nature would be to keep on growing on the left side and top-right corner (attacking the white mess in the middle), and afterwards reducing from outside the bottom half of the board.
However, this seems too slow. Should I instead invade?
Do you have general tips for situations like this, or recommendations on reads/videos/material? Thanks a lot!
I took the liberty of doing a sort of game review/ small lesson for you here. I hope this helps, and if you find any of my notes a little confusing, feel free to ask, and I’m more than happy to clarify or answer any questions you have.
I suggest you play more attention to the estimated score rather than the win probability. 0.997 is a guess at the probability of the AI winning against itself, which has little bearing on your chances of beating someone else. While the score difference (22.4) is also based on AI self-play, it is a much more useful estimate of your final score. At your level a swing of 25 points is no rarity, so your win is far from assured, even if your lead is comfortable for now.
This is exactly what I mean by “optimistic” and by not knowing how to actually win
Indeed it seems that to keep my edge the invasion points and the exploitation of the opponent’s weakness were the keys. Now I have a better view how to proceed!
So, when I was around 20 kyu, one of the things that repeatedly confused me was how easily my opponents managed to grab a 30-50 point lead - usually within the first 60-80 moves of the game. No matter how well I felt I was doing in my local battles - I kept losing the global conflict.
It took me a long time to realize that there were underlying whole-board dynamics that were shaping these games, and by being ignorant of them, I was continuously giving that advantage away to my opponent.
In order to get over this hump, it’s not enough to get more skilled in the local fighting you’re doing now - you have to grow a completely new set of eyes so that you can see new dangers and opportunities you were missing before, and then develop a new set of skills (learning when to invade early, handling weak groups, using direction of play to your advantage) to handle those risks / opportunities
I’ve been working on a 19x19 FOR BEGINNERS series here on OGS for a while. I think these 2 articles might help you see all this in a different light
It’s a lot of reading so take your time. I’ve also included several demo boards in the 2nd article that you can open in a different browser window and play through move-by-move as you read my commentary and look at my images.
Hopefully, these will help you see the ways in which you will need to zoom out from the local exchanges and learn to see the underlying large-scale forces which affect the whole-board game
As such you are still lost in most departments or you ll be an SDK already.
You will have to get some experience and “culture” in every part of the game from the opening to the endgame passing by invasion/reduction and such. I feel like an interest in middle game which is the most difficult part IMHO. So most of the mastery of it will come from your time you played.
It’s always a good time to read books or watch videos as long as you enjoy it and don’t forget to play! Find some go meeting place, find friends to guide your progress and you ll be in for the best.
