Weekly Go problems: Week 75

Here are the weekly Go problems for week 75.

Black plays first in all problems and all solutions are labeled ‘correct’. Have fun!

Easy Go problem

Making two eyes is basically about separating your eyespace into two distinct pieces. If you’re not careful though, white will stop you from doing that.


Download SGF File (Go Game Record)


Download the solutions to the easy problem as an SGF or PDF file.


Intermediate Go problem

This looks like a textbook style tesuji problem. However, in actual play you need to read a bit further ahead and think about what white wants too. There’s nearly always room to negotiate in Go.


Download SGF File (Go Game Record)


Download the solutions to the intermediate problem as an SGF or PDF file.


Hard Go problem

The ABCs of capturing races are ‘Always Be Counting’. Black has a cute tesuji and, as soon as you spot it, the problem becomes fairly easy.


Download SGF File (Go Game Record)


Download the solutions to the hard problem as an SGF or PDF file.


Still want more Go problems?

You can find Go books packed full of life and death problems, tesuji problems and other valuable Go knowledge at the Go Game Shop.

Discuss other possible moves

If you have any questions or want to discuss any of these problems, please leave a comment below at any time. You can use the coordinates on the problem images to discuss a move or sequence of moves.

You can also download the solutions as a PDF or SGF file by clicking the links below each problem.

About David Ormerod

David is a Go enthusiast who’s played the game for more than a decade. He likes learning, teaching, playing and writing about the game Go. He's taught thousands of people to play Go, both online and in person at schools, public Go demonstrations and Go clubs. David is a 5 dan amateur Go player who competed in the World Amateur Go Championships prior to starting Go Game Guru. He's also the editor of Go Game Guru.

You can follow Go Game Guru on Facebook, Twitter, Google+ and Youtube.


  1. Thank You David! πŸ™‚
    I don’t absolutely want to appear conceited, but I would like to ask You one question: it seems to me that, in the intermediate problem,Variations stemming from white E3 instead of E2 seem to me slightly better for white. For example:
    1) D4 E3 H4 G5 J4 F6 J2 D5 G6 H5 E6 F7 which builds up a wall for white on the outside (w can save his group on the lower side with B3)
    2) D4 E3 H4 G5 J4 F6 J2 D5 E5 D6 G6 H5 E6 E4
    3) D4 E3 H4 G5 J4 F6 G5 G6 H5 H6 J5 K4 E6 D5(or E4) J6 J7 H7 G7 K7 J8 K6 K3 B3 and now the shoulder hit at D8 doesn’t seem to work because of the cut at C7
    There are so many other variations I can’t write here (one in which white connects under G1 H1 for example) but, anyway, although I am convinced that they must not be good as they seem to me since you haven’t inserted in your pdf (and you always are so complete) I would like to you to point me out what I am missing (it will probably be something obvious, in which case I apologize!).
    What do you think about it?

    • Ops I haven’t said that my point was that, by playing on E3 instead of E2, white can hane at F6 instead of extending on G6 (if I wasn’t wrong of course!)…

    • David Ormerod says:

      That’s a really good questions Braket and, in some similar shapes, connecting solidly (as with E3) can be the better move because it leaves more potential for cutting later. It’s not always obvious which approach is better.

      The variations here can become complicated, but I think defending with E2 is better here, because it gives white much better eye shape in the corner. White’s group is almost alive in the corner after E2, and black’s only way to try to capture it is to play B2 later and start a risky ko.

      After white plays at E3 instead (bD4 wE3 H4 G5 J4 F6 K3), as you’ve suggested, white’s corner isn’t alive and has bad aji. Black can kill it at that point by simply playing B3 (but it’s white’s move).

      If white plays at E6 (or D5) next, black will patiently connect at E4 – let’s assume wE6 bE4. Taking away one liberty from white turns black B2 or D2 into a strong attack (it’s no longer a ko) and also creates potential for another tesuji at E1 later.

      If I were playing white, I’d atari at B3 after black E4. Black will connect, because there’s still a weakness at D2/B2, so white should defend at B2 (so that B4 isn’t immediately sente).

      Black has to move out now. It might also be a good time to exchange E1 for D2 first (because later white could resist with G1, but not now because the potential ladder at G6 makes bJ2 sente). That’s a minor detail anyway. Black can move out with D7.

      The question really is whether white can sustain the attack on black after bD7. There are many variations, but according to my reading neither enclosing with wE8 or pushing and cutting with wD6 bC6 wC7 works for white, so black will escape.

      For example: bD4 wE3 bH4 wG5 bJ4 wF6 bK3 wE6 bE4 wB3 bC4 wB2 bE1 wD2 bD7 wD6 bC6 wC7 bD8 wC8 (wB6 doesn’t work either here) bG6 wH5 bJ5 wH6 bG7 and then F7 and H7 (ladder) are miai for black.

      Obviously, there are many other possible sequences, but I hope this gives you some ideas that you can investigate further yourself.

      • Many thank you for your answer David! May I ask you what happens in your opinion if in the sequence you suggested, white plays E8 after bD7, in order to try to seal him in? I think black is going to live unconditionally (for example using the usual weakness of the keima shape C9-E8, i.e. cutting at D9) but in compensation white ends up with a huge influence on the outside and his group on the lower corner is alive. Thanks πŸ™‚

        • David Ormerod says:

          It still requires careful reading and move order, but since white has cutting points around G6, E7 and D8, black can still aim to cut and do something similar to when white cuts.

          Black also has a weakness around D6, so in this case I’d play an asking move at B9 and see how white responds.

          For example, if white simply responds with C8, bC7, wB8, then black’s weakness around D6 is fixed in sente and black can just push and cut with bD8, D9, E7, F7, F8, E9(!), G6, H5, J5, H6, G7 and white doesn’t have a good move (some white stones will be captured).

          So white will probably fall back after B9 and black can live fairly comfortably. B4 will be sente against white’s corner (when played at the right time), because of the E1 for D2 exchange earlier.

          I don’t think white’s influence will be that great because there will still be many cutting points for white to worry about.

  2. Question about the intermediate problem. I understand and agree with moves 1 – 8, but move 9 was difficult to find. I kept wanting to play E7 here as opposed to D8. It seems to me that this still gives a good result and does not provoke white into starting another fight right away. Thoughts?

    • David Ormerod says:

      That’s definitely a question that’s worth asking yourself before playing the move Philip.

      D8 is a fairly common idea here (not just for attacking, but also for moving black’s group somewhat weak group into the center), but E7 could be better in some circumstances – for example, if there’s a lot of potential on the left side and you really don’t want white to push up at D9.

      If you play E7, white will probably jump at G8, then bE9, wD11 and black has to play something like F10 or E11 next.

      My feeling is that D8 is a slightly lighter way to play, because after that, only the two cutting stones and getting into the center are really important. Anyway, play whatever you think is better and see how it goes :).

  3. The intermediate is very useful to me, I think. Understand when one has to make a negotiation in baduk is one of the thing that one must try to do in orden to get stronger. This is a good example, thanks to you David.

  4. @Philip, it depends on your playing style, but what you call fighting i see (local) pressure moves which are in general good. If White tenuki you can attack the stone at C9. If White answers, the natural direction of such (local) conflict would tend to go towards F8 or G8, which is good for attacking White 3 stone group. Playing E7 is only pressuring the White 3 stone group. D8 is multipurpose. But that’s only my opinion

  5. @Gertjan Thanks, makes sense. I get the whole leaning attack concept, I guess sometimes it just feels risky to me because you might make white strong on one side and then fail in the attack. Sometimes I feel like it’s better to wait and see how things to develop. I wasn’t sure if this was one of those times or not.

    • David Ormerod says:

      Before you play D8, you do need to figure out how you’ll play if white tries various moves to cut, so from that perspective I understand that D8 can be a difficult move to play. Remember that you can sacrifice stones when they’re not important. One thing that crossed my mind when looking at D8, is that it sort of makes miai of F6 and D6 for black. After that, there are still many possible moves.

      If you can’t figure out a good way to deal with a potential cut in a game, then playing more tightly is good. The Korean concept ‘haengma’ is basically about developing as rapidly as you safely can, without being forced to go back and patch up in an inefficient way.

      As players get stronger, they generally stretch their stones out a bit more and reduce the number of exchanges they play up-front to a minimum, so that’s something to aim for.

  6. I could not find move 9 in the intermediate one. The hard one was indeed very easy once you sit down and think.

    I used to be good in intuitive moves and bad at reading. It seems like concentrating on the problems here at GGG has reversed the issue.

    • David Ormerod says:

      Hopefully not reversed Dieter, rather rebalanced? πŸ™‚

      It seems that move 9 wasn’t the first instinct for a few people.

  7. Guillaume C. says:

    It would be nice to have a possibility to try the problem with a timeline or something to measure our speed to solve.
    Like in insei training: “100 problems: 30 minutes, 3 moves each”
    We all did most (or all) of the previous problems, so we should be able to find the solution within a correct time.
    For instance, if you solve the easy problem in more than 1 minute: your average rank is under 15k, between 30 sec and 1 min: 10k etc..
    So that we can also speed train ourselves.
    The point of these problems is to show us the good way to approach a tsumego and to give us a good intuition, right?
    I hope some other people will find this idea interesting. πŸ™‚

    • David Ormerod says:

      It’s an interesting idea. How do you think it could be implemented though? It sounds like it might require a fair bit of coding.

      There’s other software and websites like goproblems.com, gogrinder.sourceforge.net and gochild2009.appspot.com which do similar things I think. Also, the problems from our site are included with Gobandroid, which is an Android based Go app.

      I’ve also found through doing these weekly problems that I’m not particularly good at judging the difficulty of a problem. Possibly it’s because something that seems hard to one player seems easy to another, based on experience and everything else – even amongst players who play at a similar level. But maybe it’s just me.

  8. Guillaume C. says:

    It sure ain’t easy πŸ™‚
    And of course the difficulty of a problem might change from a player to another so it must be a “global” difficutly so it might need more than one people to find the right estimated rank.
    So maybe on the other hand, if there’s a way to record the time, like maybe, the user has to register his rank then can access to the problems with timer and compare his time with the score of the other.. Like “I’m 1d and my time is around the time of most 3d” or “around the time of most 5k”. It doesn’t mean anything more than a local estimation, but might be nice.
    Again, that’s if other user find it interresting..
    It sure doesn’t seem easy to implement so maybe, as we say in French “C’est un chΓ’teau en Espagne” (I think the proper translation is “It’s a castle in the air”).

  9. Guillaume C. says:

    And by the way. When I was 5k I was so into tsumego that I often was doing 3d-5d problems.
    And 3d friends of mine were havig difficulty over 1k-1d problems because they were really strong in fuseki and in dealing with influence over the board..
    So the level of a tsumego is an abstract concept, I think.

  10. Hi. I agree with David’s comments about D8 – a natural leaning attack, but also strengthens white here.

    I feel like suggesting E9 (even d9!?) here too… but I shall restrain myself…

    …. oops! :-).

    By the way I love these open problems, and some of the discussions that ensue are like a well organized post-game analysis! πŸ™‚