I'd like to work on my counting skills. In particular, to practice precise counting so I can more easily do approximate counting in games.

Is there a practice resource available for this? For example - show a group/board, you count then it shows you or you pick the answer. Like goproblems.com but for counting.

I looked around and couldn't find any thing on the internet (maybe most people would find this tragically boring) but I would be willing to pay a small fee for it.

Maybe one way is to load random games on KGS, go to the end, count then compare to Score Estimator - seems a but clumsy though.

Any other suggestions?

-Brendan

There are 9x9 endgame tsumego.

This isn't exactly what you're looking for, but you might find it useful anyway. It's a series of lessons on counting by DrStraw that he gave a few years back.

http://senseis.xmp.net/?SteveFawthrop%2FCounting

_________________
Patience, grasshopper.

If you can't wait and want mostly territory counting, read Cho's Positional Judgment High-Speed Game Analysis http://senseis.xmp.net/?PositionalJudgm ... meAnalysis
If you can wait two or three months, the good news is that I just happen to write a detailed book on positional judgement and evaluating territory, thickness, influence, options, endgames etc.

Practicing counting is the easy part. The difficult part is to know what to count and under which conditions it may be counted or when simple counting is misleading. Then helping methods such as quiescence become necessary.

The general ideas contained in it are useful but I disagree to which intersections he assesses as territory, to the counts and to his heuristics for these purposes.

Nice to hear! Sounds like you're on a roll ^^

_________________
Patience, grasshopper.

Maybe I'll have my cake and eat it too by doing both! :p

You might pick up a copy of the Train Like a Pro series, if it's still available. There are a number of excellent full-board endgame problems on 13x13 boards. Your job is to read out the last 20 or so moves in the game and figure out the final game result. Highly recommended.

I would recommend positional judgement by cho chikun

Why not just get a random pro game that you know ends within 5 or so points and count that? You can count it at the end position to start, and slowly take some moves off and estimate. As a bonus, you also get to learn good endgame moves.

It is good to see that that book determines Current Territory.

Interesting, I think I will give this a trial.

That is one groovy cover, man.

_________________
a1h1 [1d]: You just need to curse the gods and defend.
Good Go = Shape.
Associação Portuguesa de Go

Endgame problems from senseis:

http://senseis.xmp.net/?EndgameProblems
http://senseis.xmp.net/?PracticalEndgameTests

I love it! Laughed a lot and for long time when I saw it. Awesome.

An interesting idea would be to do a search over your old games and find all the ones that were 0.5 point wins/losses, or at least pretty close, and take the final board position and make those in to problems. I still find it possible to count the same -finished- board position 2 or 3 times and get 2 or 3 slightly different counts, especially when you factor in captured stones and komi; it's just really easy to lose track of a point here or there (on that subject, if I have to count playing cards to make sure I have all 52 I usually count twice because about 25% of the time I'll count wrong and have to do a best 2 of 3, so the core problem might be that I just suck at counting things in general).

I've been meaning to do this with my own games at some point but I've been super lazy. If I manage it at some point I might post it here.

_________________
1k KGS