Winning a lost game

Tuesday, March 13. 2007


Yesterday I found myself on a bad situation after having played very bad moves on the opening and the middle game. Luckily for me I could find a path which would lead black to the defeat.

Where is it?

Did you find it?

You can watch the full game without commentaries on the "continue reading" with the zGo applet. You'll find lots of mistakes from boths sides.

The situation shown above is movement 201, black at K4. But I'd like to call your attention to the move 70 by White, the incredibly bad sequences played by white on the smaller groups and the uncomfortable obstination of black on the left side.


Continue reading "Winning a lost game"

Lectures and Forum

Tuesday, March 13. 2007


As you can see I've been working on the blog features and the results are quite obvious. There is a new Forum going on which you can visit by clicking on its link, and there will be lectures on weekends.

You'll find more details about this features on the corresponding sections.

Currently I'm spending most of my spare time in tsumegos and Bruce Wilcox Go Dojo, so that's why I don't post much lately.

Please, let me know your opinion on the added features and remember that any suggestion will be welcomed.

Don't keima while escaping

Saturday, March 10. 2007


First of all: keima is the correct word for the "knight's move" on Go.

Keima can be a very good move while making a connection, protecting a cut or establishing a group. But it has a weakness: Keima always leaves the remote chance of a cut.

Let's see this situation:

As you can see, there is a black floating group, with only one eye. If black tries to escape at A, White will play at B, Black responds at C and White cuts at D.

Taking into account the superiority of White in this area, the cut will work nicely. If White didn't have that much superiority there would be a huge fight for the cutting stones, remember that cutting stones are very valuable, not just because they are 2 or 4 points; it is because they lead the huge black group to life or death.

So, from now on, if you have to escape, try single space jumps, and avoid keimas. The only way to play keimas while escaping is having read all the possibilities and knowing that you will succeed.

Counting territory

Saturday, March 3. 2007


One important thing to do during the game, as soon as possible, and where I'm very weak at, is calculating the territory of each player. Taking into account your situation you can endure drastic mesures or play defensively without risk, so this sometimes helps you to choose between two good movements.

There are several ways to count territory:
-count the pairs inside your territory.
-count groups of 10 points in your territory.
-compare territories at large scale.
This list is ordered from the most precise to the least as well as from the most time-consuming to the least.

I usually compare territories at the beginning of the middle game, to realise what my situation is. And keep doing this a couple of times till the end of the game.

More advanced users will tell you to do this, above all, after fights. After a fight is a good moment to stop, calm down, count your territory and see where is the biggest point to play next. This procedure, obviously, depends on the amount of time you have.

The method of counting territory by pairs or groups of ten is the most accurate mesure and can be played from the late middle-game till the end of the game. Personally, and this is just my point of view, I find "area scoring" better suited for me, as it is much faster and, once I got used to it, I can predict scoring with +/- 5 points, which is not bad for me.


You can find more about this issue at:


One situation I found myself today was the following one: I was losing inevitably the game, we had already played 200 moves and yose almost finished, my opponent suicides in a way that he loses around 20 points. Seconds later after this move, he leaves the match. Note that I said "leaves", not "resigns" (I'll make a post about this someday).

Try counting the points for each player on the next board, with white player capturing 11 stones and black 6, and a komi of 6,5 points.





White: 70 territory + 11 captures + 6,5 komi = 87,5
Black: 84 territory + 6 captures = 90,0

Black wins by 2,5. He shouldn't have left the game.

Now let's get to the important thing: estimating. This situations is from a sample game between two 10-kyuish players, I used this one because pro games are often too complicated for us to appreciate territory and life-or-dead status of stones.

As you can see, white territory is marked with crosses, and black's with circles.

When you do this, you must be aware of the undefined limits of the territories, as you can see on both sides from the m2 area; as well as the possible entry of black on white's moyo around J9.

Now we yet to count it by pairs, tenths or area estimation.

Black UL: 7 UR: 16 LL:47  LR:10. Plus 2 captures

White L:80  UR:18. Plus 10 captures.

My estimations, before counting were quite ok: Black had around 75 points, and White had something between 100 and 110 points.

The more you train this technique, the better it becomes, and it might help you sometimes to choose between playing defensively or ofensively.