Yesterday I went through my recent tournament losses to compile a set of my mistakes. This compilation focuses on the decisions that cost me the game, confirmed by analysis with KataGo.
The result are roughly 40 board positions which I have started to transcribe to my new, first, puzzle collection here. I might finish the collection in the coming days as time allows.
With some care, the positions fit into OGS’ problem format and I hope some of you find it entertaining or useful.
Now I don’t know how to link a whole collection. Even for other puzzles, a click on “the collection” just opens some puzzle in it. Anyway, I fixed my link, please try again.
Yes, I would put a circle (as last move marker) on n14 in problem 71338, otherwise you read “Black aims to reduce the right-side potential and simultaneously swallow White’s weak group on a large scale.” and wonder what move that’s talking about unless you have enough go skill and memory/prediction from previous problem to figure it out. Or add in text "With n14 black aims to reduce the right-side potential and simultaneously swallow White’s weak group on a large scale.
And then “Continuations at the circle mark.” confused me as I was expecting to see some circle marks on the board immediately, only after I managed to get one of the answers did I understand what it meant. And use a different shape from circle. So “Black continuations at the triangle mark.” clearer.
Problem 3 doesn’t seem like 5k to me! As a 4d I can think of the answer and the continuation you give is all very plausible, but as black’s cut white at f11 and it’s still not clear how white is going to manage the groups so I’d want at least another 10 moves or so to reach a quiet enough position I can make a confident positional judgement that white is better than simple nobi.
Thank you, I implemented your feedback on the markings.
With ranks and variations, it is always hard to say exactly how much the problem should contain. I prefer fewer moves because the longer the sequence, the less realistic it is that it will appear in an actual game. Especially once even the computer starts spitting out multiple viable branches.
Another option that I see is to give the “naive response”, i.e. show how black, instead of resisting, jumps on the single stone and gets over-concentrated. This could be educational, but also create false expectations. I decided to exhibit the stronger variation because the curious reader might ask “why doesn’t Black just capture?” and figure it out. On the other hand, when we are in White’s shoes and get to see Black dominated, we are less likely to question the methods. (Racist connotations not intended )