New go tool

[John Fairbairn]

I was sitting on a train this week looking out at tracts of snow in the countryside. I find virgin snow strangely stimulating in making me think new thoughts, but I was astounded by the thought that came up unbidden this time. It was whether there was not a brand-new way of looking at go that was simple and useful. I have no idea what might have brought this on, because I've had no other thoughts in that direction before that I can recall, and I don't study go or teach go.

I was even more astounded when an answer came to me instantly. I can only assume the answer was there already bursting to get out, and it told my brain to ask the question for form's sake. If so, I likewise have no idea from whence or why the answer came.

The answer seemed so simple and useful that I assumed there had to be a major flaw I'd overlooked. I therefore waited till the GoGoD executive lunch today to bounce the idea off my colleague T Mark Hall, who is not only rather stronger but is also very, very quick at spotting flaws. He not only spotted none here, but even gave the whole idea a provisional thumbs up. Given that seal of approval, I will describe the idea here. Since I don't play or study, the idea is entirely untested, but if anyone does try it out I'd be pleased to hear the reaction.

The new tool is simply to make a list of all your groups in order of priority at various stages in the game, and a similar list for the opponent's groups. You do this mentally during a game, but you can also make notes if you are studying.

Let me list some benefits:
1. Any player of any grade can do this - and I maintain it is useful for any grade.
2. It can be done in less than a minute the first time, then can be refreshed in seconds.
3. The number of groups alone gives you useful instant information: if you have more than five, the proverb tells you the sixth will die. Or it can tell you that you maybe have too few groups and are in danger of being overconcentrated.
4. You can use this information in various other ways, e.g.
(a) During play, you can make decisions where to play or whether to answer on the basis of your priorities
(b) During study, you can look at groups in the final position and compare with your list of priorities earlier in the game. E.g. if you made a group a high priority but it ended up with just three points, that tells you not just that something went wrong but where.
5. You can apply the method to your own games and to pro games.
6. The method is not limited to any one part of the game. Tracking your groups' rises and falls actually maps out the game in a useful way.
7. It provides instant feedback.
8. You can refine it a la carte. E.g you can add a value to a group - either a number or a category such as major, medium, minor.

Any grade can use this method. Of course, a stronger player will make more reliable early assessments which accord better quantitatively with the final result, but qualitatively the weaker player gets the same sort of benefit. The method can even be used between vastly different grades. A teacher asking a pupil to list groups in order of priority gets useful and instant information that essentially means the same to both parties, and he can give advice using the same medium.

If anyone ends up using this method, I'd like to call it the Virgin Snow method, so as to make it sound mysterious.

As far as I can see, there are no drawbacks to the system and no interference with any other mental tools you may use.

Over to you.

[EdLee]

I was sitting on a train this week looking out at tracts of snow in the countryside.

Very Snow Country.
国境の長いトンネルを抜けると雪国であった。

[daniel_the_smith]

I like this idea, I will try to use it next time I play and report back.

[billywoods]

The new tool is simply to make a list of all your groups in order of priority at various stages in the game, and a similar list for the opponent's groups.

Can I ask what you mean by "priority"? I suspect that this term is deliberately vague, but...

[EdLee]

Can I ask what you mean by "priority"? I suspect that this term is deliberately vague, but...

If you have a group where if you lose it, it's game over, that's pretty high priority.

[daal]

The new tool is simply to make a list of all your groups in order of priority at various stages in the game, and a similar list for the opponent's groups. You do this mentally during a game, but you can also make notes if you are studying.

First of all, this is such an elegant idea, it's astonishing that it hasn't been presented in this form before (though if you could find a way to express it without the word "list" it would be even more elegant).

I immediately tried it out, and the first instant benefit was that throughout the entire game, I found myself looking purposefully at the whole board.

Can I ask what you mean by "priority"? I suspect that this term is deliberately vague, but...

At our level, this is a good question, yet at the same time, trying to answer it ourselves seems to be the task at hand. What are the functions of our groups? Are they making territory, are they keeping an opponent's group weak, are they providing support for a later attack etc? We have to be the judge of what our groups are doing to help us win the game, and the "virgin snow method" is a tool which allows us to better grasp what we are comparing when we consider places to move.

[billywoods]

Can I ask what you mean by "priority"? I suspect that this term is deliberately vague, but...

If you have a group where if you lose it, it's game over, that's pretty high priority.

Then all of my groups (apart from sacrifice stones and probes) are high priority after move 20. If I'm thrashing out groups that I could afford to lose, either I'm dividing my efforts or I'm winning by a hundred points. I can see how there's use in noting whether stones are light or heavy, or strong or weak, but I don't see what a "medium priority" group is in your understanding.

At our level, this is a good question, yet at the same time, trying to answer it ourselves seems to be the task at hand. What are the functions of our groups? Are they making territory, are they keeping an opponent's group weak, are they providing support for a later attack etc? We have to be the judge of what our groups are doing to help us win the game, and the "virgin snow method" is a tool which allows us to better grasp what we are comparing when we consider places to move.

I agree that this is all very important to think about, but it doesn't sound like the "simple" advice that John was intending. For a start, he suggests making a list of groups in order. How do I rank some cutting stones, a thick wall, a 20-point corner and an invasion in order? The point is that they're all important, in different ways (at least until they're suddenly not any more, at which point I may sacrifice them).

John's point 4b sounds very much like he is talking (perhaps among other things) about choosing to prioritise the development of certain groups over others, not working out how strong or important or heavy they are (which Ed and daal seem to be talking about). My question stands.

[John Fairbairn]

Asking me how to sort out your priorities is asking me to play your game. The rank of your groups depends on what you have been trying to do so far, what you want to do next and what you think the opponent is doing. The whole idea of the method is to give a fix and get some feedback on what you are doing.

If you are in doubt how to rank one group above another, just ask yourself the Desert Island question. If you could only take one group with you to a desert island, which one would it be?

I'm sure there are many parables exposing the folly of saying all your groups are important, but I'll leave them to others.

[billywoods]

Asking me how to sort out your priorities is asking me to play your game.

No, I'm simply asking you to define what you meant when you said "priority". I understand priority as a complex, multi-dimensional, intertwined, constantly changing, horrendous beast, which I of course try to evaluate, but usually fail to evaluate correctly due to my poor shape, direction, reading, etc. On the other hand, you describe it as "simple", and suggest that I put it into a linear order. That's why I think I'm misunderstanding you, and am asking for clarification.

[Javaness2]

Sounds like a reasonable enough analysis method to me.
I tend not to think in a group based way during a game though.

[Javaness2]

I think that you should be able to define priority yourself. If you cannot, then I'd leave the theory on the shelf.

Asking me how to sort out your priorities is asking me to play your game.

No, I'm simply asking you to define what you meant when you said "priority". I understand priority as a complex, multi-dimensional, intertwined, constantly changing, horrendous beast, which I of course try to evaluate, but usually fail to evaluate correctly due to my poor shape, direction, reading, etc. On the other hand, you describe it as "simple", and suggest that I put it into a linear order. That's why I think I'm misunderstanding you, and am asking for clarification.

[ez4u]

I tried this today. So far the new heuristic did not survive the first 8 or so plays. I need more practice at thinking in terms of groups rather than other possible images. Try again tomorrow.

[daniel_the_smith]

Those of you who don't think in terms of groups, what do you think in terms of?

How do I rank some cutting stones, a thick wall, a 20-point corner and an invasion in order? The point is that they're all important, in different ways (at least until they're suddenly not any more, at which point I may sacrifice them).

Sounds like making the attempt will be especially good for you!

Everything in go is eventually reducible to points. So you can ask yourself how much komi you'd want in exchange for removing a group from the board. EDIT: don't forget that you need to factor in the probability that your opponent is going to forcibly remove it anyway!

[Unusedname]

Those of you who don't think in terms of groups, what do you think in terms of?

How do I rank some cutting stones, a thick wall, a 20-point corner and an invasion in order? The point is that they're all important, in different ways (at least until they're suddenly not any more, at which point I may sacrifice them).

Sounds like making the attempt will be especially good for you!

Everything in go is eventually reducible to points. So you can ask yourself how much komi you'd want in exchange for removing a group from the board. EDIT: don't forget that you need to factor in the probability that your opponent is going to forcibly remove it anyway!

i recently (like a month ago) started thinking in terms of groups.

but it's strange because i don't remember what i used to think before.

I think 4b is the most important step

cause then you can figure out why you value poor groups so highly and then stop doing it.

[billywoods]

you can ask yourself how much komi you'd want in exchange for removing a group from the board

Sure, this is very sensible advice - "how many points will this group get me?" is an obvious question, and it's one you have to be able to answer well in order to get good use out of your groups. Doesn't everyone already do this? Isn't it really difficult? I'm still struggling to see where this "new", "simple" tool is. In fact, I won't nitpick: I don't care if it's new or not, but I would be grateful for an explanation of why it's simple. I find it as hard as anything else.

(I hope I'm not coming across as obnoxious here, because I'm getting some rather curt replies. I just want to understand what John had in mind.)