Why do some people stay at the same rank?

[RobertJasiek]

Besides the already mentioned reasons, there are more:

- The general level of skill at a given rank increases.

- Doing too little at these skills demanding the greatest effort: a) life+death reading, b) endgame value calculations.

- Neglecting some other skills: positional judgement, fighting, joseki or whatever.

- Neglecting some means of learning, e.g., learning too little from one's own mistakes.

- Having too little access to strong opponents or too little money for running AI well.

- Having a learning / thinking style not met by any available teachers.

[John Fairbairn]

- Having too little ... money for running AI well.

- Having a learning / thinking style not met by any available teachers.

I think that the first line of that quote illustrates, although in an unusual way, a problem that afflicts nearly all of us. Given a piece of jam and bread, we lick the jam off and leave the bread. Or, if forced to eat the bread as well, we cut the crusts off. I fell victim to the same syndrome, also in an unusual way. Once I learned about the history of go, I stopped studying the game and tried to get my hands on anything and everything about players of the past. With no regrets.

But even if you really want to remain a dedicated jam licker, you can go too far. You don't need jam that is made exclusively from fruit picked in remote valleys high in the Andes and cooked by virgins who have undergone a 20-year purification process. Marmalade will do. In fact, for people like Go Seigen, Huang Longshi, Nakamura Sumire and Ogawa Doteki, just basic bread has done very nicely, thank you.

As to the second half of the quote, it's obviously a personal choice whether to try something different, but one of the joys of my life has been observing, and learning from, how differently other people think. You too have given me joy, Robert!

[jlt]

Teachers can explain mistakes in human terms but are not always available: this is where AI can help. However, AIs can't replace human teaching for the moment. Even people who have access to a powerful computer are still willing to pay a teacher.

[CoolBoots]

My teacher says many are taking their effort on a wrong path.
While improving, their strength, in terms of winrate, will go back (by the way Cho Chikun 9p have said the same).
Why?, because you are a beginner on such new skills/new ways of thinking. By how much?, even two or three stones(2-3 kyu/dan). For how long?, as long as 6 month, sometimes a year. And these players, including myself, revert to their old ways.

I think I really needed to read this. I'm the kind of person your teacher is talking about. I take short term feedback (E.g. the result of a single game) far too seriously, often dropping experimentation if it starts to affect my rank.
My question is: Is it enough to focus on new ideas (like a chapter of a theory book), hoping through time they overwrite your old knowledge, or do you also need to somehow unlearn old (incorrect) knowledge? If so, how?

[jlt]

I take short term feedback (E.g. the result of a single game) far too seriously, often dropping experimentation if it starts to affect my rank.

A suggestion: open a separate account on which you try crazy things, e.g. Black hole opening, joseki variations that you don't know well, playing with a completely different style, etc. Since you will go into paths that you haven't explored well, the rank of that secondary account will necessarily be lower than your main account. If not, then try crazier things, until your mind is completely free from trying to increase your rank.

[CoolBoots]

Thank you for the advice. I think I am going to go with the second option, as the issue isn't so much with protecting my accounts, but with short term thinking, like "this isn't working right now, so I need to stop doing it". I'm really looking forward to playing more interesting games and gleefully tanking my main account. I think just the idea of disconnecting rank from improvement on a timescale as long as a year has blown my mind a bit.