How to deal with 5-4 openings: Tips and tricks

Recently I saw a question about a particular 5-4 opening asked on the baduk community page on reddit.

A problem position involving the 5-4 point. How should white play in this opening?

Dealing with 5-4 openings in Go

The person who asked the question is also a reader of Go Game Guru, so I’ve decided to post my response here. In this article I’ll deal mainly with the possible continuations, pitfalls and techniques for handling 5-4 openings. I won’t spend much time discussing the differences between the 5-4 point and the (4-4) star point – which were also mentioned in the original question. I think that once you see some of the variations discussed in the article, some of the differences between 4-4 and 5-4 will become apparent anyway.

This particular opening emphasizes many of the different strategies involved in using the 5-4 point nicely, but I’ll also touch on some different situations briefly.

Improve your competence and your confidence

Many Go players I know are not confident in their ability to deal with openings using the 5-4 point. On the other hand, some players understand the 5-4 point well and use it as a weapon in their games.

Ultimately, opening on the 5-4 point is just another move and sometimes the difficulty in dealing with it is largely psychological. In this article we’ll cover some effective ways of dealing with your opponent’s 5-4 play and also show you some tricks that you or your opponent might try. We would like to improve both your confidence and your skill in using or playing against 5-4 openings. Just remember, after reading this you’ll probably know more about the 5-4 point than most of your opponents do.

Be a part of our monthly question and answer series

People who subscribe to our newsletter already know about this, but I’ve heard talk of a monthly question and answer series starting soon for our newsletter subscribers. If you’d like to get our newsletter and be a part of the monthly questions and answers, you can do so for free using the form in the top right of this page, or on our homepage.

Playing a flexible opening

Before we get caught up in all variations of the problem position above, I’d like to show you another opening briefly. Black’s moves in this game are a good demonstration of flexibility in the opening, as well as another nice technique for managing the 5-4 point.

An interesting and flexible way of simplifying the opening for black

The moves up to 6, in the diagram above, are all normal and reasonable. Both players are taking the big points in the opening. When white plays the knight’s move press at 6, it becomes clear that white wants black to take territory in the corner in exchange for influence over the center and top side of the board.

It would be common for black to continue playing in the top left, going along with white’s plan. However, in this game black switches to 7, taking control of the top side before white can. White’s continuation at 8 is natural and black trades the corner for the top. Looking at the whole board, black has quite a nice opening and the top right star point is working well…

Black's aji in the corner: Later black might play 12-16, leading to a complicated fight after white continues with A

… And white still hasn’t taken complete control of the corner, which is a little bit unsettling. Play won’t continue in the top left right now, but black has potential (aji) in the corner starting with B in the opening diagram. It’s worth noting that black shouldn’t exchange A for B here, because that would help white fix the weakness at B. Also, if white uses sente to play at A now, black can play C and be satisfied with a fast opening.

This novel opening idea is discussed by Go Seigen in his challenging, but worthwhile book A Way of Play for the 21st Century.

The reason I’m showing it here is because it demonstrates that there are always alternative ways to play in Go, especially in the opening. Often we convince ourselves that there are only one or two options for the next move. There are tactical situations where this is sometimes true, but generally you have more freedom than you think. Remember that one of the great things about Go is that you are free to play wherever you want. Don’t straight-jacket yourself with joseki or opening moves.

Stating the conclusion first

Now let’s return to the original problem diagram. I’ll state my conclusions first – to make it easier for you to follow along – but it will take a fair bit of analysis to explain them so please bear with me.

White C and D are good moves in this opening

According to classical Go theory, approaching the lower right corner is the biggest move for white now, because it stops black from finishing a position there by playing A. White’s most common move in the corner is also A, though B is often played by people who want to keep the game simple. In my opinion, both A and B are ineffective in this opening and will make the game uncomfortable for white.

White C, on the other hand, is a special technique for approaching the 5-4 corner which is particularly effective when the opponent builds a formation extending from the 5-4 stone, as black has done in this game. White D is also a strong move, which calmly gives white a good position while keeping the game simple for now. White E and F are also interesting moves to experiment with, but we don’t have space to discuss them today (this article is already long enough!).

Why shouldn’t white play the 3-4 point?

If white enters the corner at the 3-4 point, black will press with 9

So you’re probably wondering why white shouldn’t play at the 3-4 point, right? Let’s cover that first.

If white enters the corner at 3-4, black can respond with a nice knight’s move press (at 9 in the diagram). As we saw earlier, this move attempts to enclose white in the corner and is quite natural.

In this case white can’t easily trade the corner for the outside as I showed earlier, because black has already made a position on the right side. So if white wants to respond locally the best options are A, B or C. I’ll discuss why A isn’t a good move below.

White C is sometimes playable, but is usually a bit slack because it seems like white has been forced into a low position (on the second line). If white plays C, black could follow up later with R3 or N2. Black will choose between these depending on what black wants to achieve.

White B is a better move. In fact it’s the best continuation once white plays at the 3-4 point. However, the final result isn’t particularly good in this opening and it leads to a position that is favorable for black. There’s a lot to discuss in this article, so the discussion of white B got edited out, but I’ve provided some variations in the sgf file at the end of this article for those who want to know more.

Falling for black’s trick

If white does attach under at A, he will have fallen for black’s ‘trick’ in this opening. As shown in this diagram, if white plays at 3…

White tries to manage the situation after black's severe move at 4

…black promptly cuts at 4, splitting white’s stones and challenging white to a fight inside black’s sphere of influence. This is what black was hoping for all along. White’s moves at 7 (this is the key move) and 9 are a clever technique for managing the emergency and are worth remembering. The continuation through to 15 is a sort of joseki, however I don’t think it’s good for white in this game.

Looking at the result, even though white has managed to capture a stone and get out of immediate danger, the position still favors black. Black A is well positioned to exert a subtle pressure on white’s group and erase its influence. Even though white looks thick now, the group could later become heavy.

But why does black let white capture a stone in a ladder?

That’s a good question. Here are two diagrams that show possible continuations, but there are many more possible moves. If you are interested, you should look this up in a joseki dictionary.

White wins the capturing race - up to white, 10 white is one liberty ahead in the capturing race

This black 1 looks clever, taking a liberty while blocking the ladder, but it also fails after white creates a ladder with 14-18












Can’t white just play the normal joseki?

In this case, no. The ‘normal’ joseki (or at least the variation often played by amateurs) backfires superbly. Once again, this exactly is what black is hoping for.

Black A is the joseki continuation (and the shape move) locally

The snare is complete - black crushes white with 10











Because of the nature of this position, black can go all out attacking white’s three stones on the right hand side. This makes things incredibly difficult for white and the best option for white now would be to leave the three stones for now and defend the bottom.

You can see that black’s right-center start point stone (A in the diagram) is working beautifully to make this attack possible.

Trying to keep the game simple

Often when players are unfamiliar with the 5-4 point, or when they want to avoid immediate fighting, they like to approach from the opposite 3-5 point, as shown in the diagram below.

White's group feels constricted

This move (white 8 ) is often good for simplifying the game, because white makes miai of  extending along the bottom or into the corner. Yet, in this opening I don’t find the result to be that appealing for white.

The problem is that when black extends to 11 it works well with the lower left star point stone and white’s group feels constricted. It helps if you try to develop an intuitive feel for white’s discomfort in positions like this because it can sharpen your intuition and improve your judgement.

Black can attack white’s group at A next, so white should continue with A or B to prevent any immediate complications.

Even though black’s frameworks in the lower left and on the right side look thin, it’s difficult to invade successfully when your own group isn’t that strong. Black would be happy with white’s immediate invasion, because then black could start a double attack against white’s invading stones and the two stones along the bottom. Double attacks are one of the joys of the middle game of Go.

The three space extension of white 1 is also possible

White could improve this situation slightly by playing a three-space extension to the fourth line, as shown here. Black would still check white with 2, but white has a bit more flexibility now and it’s harder to surround white’s group in one move. If I were playing white I would tenuki (play elsewhere) after black 2 and adopt a scorched earth policy on the bottom.

However playing this way does depend a bit on your style and I still don’t think the position is particularly good for white overall.

Playing this shape (a three-space extension from the third line to the fourth) is usually a good way to create a flexible group, but in this case white’s flexibility is reduced slightly because black is anchored so securely to the lower right corner at the 3-3 point.

A better way to keep the game simple

Here’s a better way to keep the game simple. Compare the diagrams below to the ones we’ve just looked at. I’m sure you’ll be able to feel how much more comfortable white’s group is.

Splitting the bottom with 1 and then taking the key point at 5 gives white a good opening

Yes, I know it’s another two space group. In some ways this does look very similar to the diagram above where I said that white was uncomfortable. So what’s the difference? There are two key points to take note of here:

  1. White’s group has moved two points to left, which reduces the value of black 2 and restricts the scope of blacks lower left star point significantly.
  2. There are two spaces separating white 3 and black 4. This doesn’t look like much, but it does a great deal in limiting black’s ability to attack white severely. Again, try to get a feel for how much more ‘room to breathe’ white has here.

For advanced players who enjoy thinking about tewari analysis during the opening, I’ve provided some additional comments about this position in the sgf file at the end of this article. In the diagram above, white could also play 3 at 4. Now that that white has exchanged 1 for 2 the situation has changed and white can enter the corner at the 3-4 point.

Here’s another interesting way of playing which Younggil pointed out to me when I showed him the draft of this article.

White makes a comfortable position on the bottom and can invade later at A, B or C

White creates an even stronger position at the bottom (but loses sente). Playing this way is nice because it turns white’s bottom group into a kind of launchpad from which invasions at A, B and especially C can easily be supported. This is an example of storing power which can be released during later attacks.

I really like this approach of splitting the bottom. It is simple, but it is powerful. Depending on your style, either of the last two diagrams are good for white. There’s still one more technique I’d like to cover though, which is invading at the 3-3 point.

Invading at the 3-3 point is a good technique in this type of opening

Entering the corner at the 3-3 point is an aggressive strategy, which aims to swiftly take territory while making black’s right side position inefficient. Here’s how it works.

Black's only territory is at the bottom

The moves up to 16 are a joseki, and though there are other ways to play here, it turns out to be hard for black to make efficient use of 5 and 7 no matter what black does. When white snakes out from the corner with 16 it undermines and ruins black’s right side position. This makes black’s earlier exchanges of 5, 6 and 7 look questionable.

In compensation black does get some influence across the bottom side and in the moves through to 22 black forms a solid territory there. On the other hand, white currently controls three corners and has built a framework on the left side of the Go board.

When you think about it for a bit, black’s only territory is at the bottom, with nearly all of black’s moves being concentrated there. This isn’t particularly efficient, so white needn’t be jealous of this territory and needn’t worry about trying to invade it.

In relation to white 18 and 20, 22 could also be at A. In this game, black is very strong at the bottom, so I’d prefer to play a bit conservatively and make it harder for black to stir up trouble on the left side.

You may be wondering about white 20. Isn’t that a weird move? It gives black territory with 21, but maybe black is entitled to that territory?

Don’t forget about the junction between two large frameworks (moyos)

Black creates a large framework extending from the bottom

If white rushes to extend on the left side with a move like 1, black will seize the opportunity to create a large framework with 2 and 4. It’s easier to prevent this from the outset, which is why white played 20 in the previous diagram.

White would have trouble invading the bottom anyway, so it’s better to let black have certain territory, which is limited to an acceptable size, than a huge uncontrollable framework which gives white a headache.

Try these techniques in your own games

Next time you get a chance, why not try some of these techniques in your own games? We’d love to hear how it goes.

Go game record with bonus variations

There were a few things I had to gloss over in this article to keep this article to a manageable length. If you’re still interested you can have a look at the additional variations and explanations in this sgf file. There are also some variations which Younggil added when I asked him to review my comments. You can either browse it on this site or download it using the link below the player.


Download SGF File (Go Game Record)


Related Articles

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. scwizard says:

    Thank you!

    As you can imagine I’ve been looking forward to this article greatly.

  2. elementc says:

    Why were you so worried about the article being “too long” if you already wrote the extra stuff? The longer the better I say. :>

    Usually when I see the 5-4 in actual games it always seems to be paired with a 3-5.

    • David Ormerod says:

      Thanks for the feedback elementc. I draft these sorts of articles as sgf files with comments, because it helps me structure my thoughts. For this article, I also asked Younggil to proof read the draft, to see if he agreed with my whole board assessments.

      In this case the omitted parts were long diversions, which I felt broke up the flow and distracted from the key points I’m trying to teach. I provided the sgf for readers who wanted more detail, so you’re not missing out on anything :). Just download it and you can read all the comments from the draft.

      As for 3-5 with 5-4, that would be the subject of another article, though you can still apply much of what you’ve learned here. The main opening in this article was one I played often as a low kyu/dan player. It was very effective against my opponents at that level, but I don’t really play this way anymore.

      • elementc says:

        Ah, I didn’t realize you meant it was fully intact. Thanks.

        As for the 5-3 combo I mentioned, I have an SGF where I asked younggil about it before. If you’d like to see it feel free to contact me on KGS.

  3. Hi David,

    It is a greate article. I have a question, in the diagram you named “White makes a comfortable position on the bottom and can invade later at A, B or C”, what happen if black do m3 or n3 instead of move 2 in the diagram. Also, I think it will be better if you could number the diagrams.


    • David Ormerod says:

      Hi Daniel,

      Good question. If black splits white at n3 (or m3), white can take control of the corner. Before doing that white might want to sacrifice k3 to get a foothold in the center.

      For example, this might be white’s plan:

      Black could play 10 at A, but then white’s corner is stronger (because white can start a capturing race at 10). After this if black B, white C. If black C, white D. Earlier on, black might consider resisting with 6 at 7. This is what white wants, but black should try to resist.

      I’ve been thinking I need to number the diagrams too. I’ve been trying to work out the best way to do it.

  4. LucNoSensei says:

    Hey David, cool article for all those 5*4 phobics out there 🙂 I like that 3*3 move too, but I guess I should since we discussed this stuff a few years ago and you pretty much taught me exactly that 😉

    The article subject matter reminded me of something Fan Hui said a while back, about 5*4 and “strange” opening moves in general. What he said could be summed up a bit like this:

    “Don’t worry when your opponent plays moves you don’t know very well, often it means they’re afraid to play normal moves because they are less knowledgeable than you are about opening and joseki. Just be careful of tricks and you’ll be fine!”

    Of course, it’s as much fact as it is jest (anyone who knows Fan Hui can guess that much) but it holds some reassuring truth for those who panic everytime they see something they’re less comfortable with. The path I’ve chosen for myself is a bit different. I tell myself “a double ni ren sei game could confuse me to no end within 70 moves if we happen to explore obscure variations so in the grand scheme of things, I really know nothing more about 4*4 than 5*4. Maybe I know twice as much, but that’s still just 2% compared to 1%”.

    Maybe one of those mantras can help someone approach the board not with fear, but with zen confusion heheh.

  5. Hello David,

    thanks for this fine article which is an enjoying read and from which I (hopefully) learnt a lot. And while I’m at it, thanks for this whole web site, this is a great addition to the Go web!

    Looking forward to read a LOT more here …

    Greetings from Germany, Tom

    • David Ormerod says:

      Thanks Tom, I’m glad you’re enjoying our site and I hope you get to use these ideas in your games :). Good luck!

  6. Anonymous says:

    Can black play P4 instead of P3 against the san-san invasion ?

    • David Ormerod says:

      Yes, that’s possible. White would probably slide to S5 then. If black plays S6 next, white could consider R6 and give up the 3-3 stone as part of a trade, if necessary.

  7. wow nice sight love this post 🙂 like the 3-3 too the 5-3, 3-5 move was played on me in a few games back an as a new play only been play on and off for a year i got hit with it bad lol i play at the 3-4 point i hate the kgs ranking system thouhg got me listed as 2nd dan 🙁 i dont like ranked games any tips would be helpful thx

    • David Ormerod says:

      Thanks goldd, I’m glad you enjoyed the article. You could always make a new account on KGS if you’re having a problem with your rank, or you can turn your rank off in your user profile and it will become (-).

  8. Thanks for this article, really helpful 🙂

  9. Anthony says:

    Thanks David

    I have only recently “discovered this site” and I find it to be the best teaching/go news/forum site I have encountered. I feel the targeted focus of the articles with clear and concise examples and commentary lends itself to more efficie

  10. Anthony says:

    Sorry about the truncated post.

    I was simply praising your streamlined approach to some of these problems that have often been dealt with in a rather tedious and heavy handed manner. While this information dense approach supplies nearly endless analysis and variations I feel it actually robs some of the potential benefit to practical knowledge useful in live games, whereas your teaching style improves this practical knowledge because the focus allowsa a player like me to more easily apply these tactics. Anyway just my opinion. Love this site. Thanks

  11. David

    I would love to use a couple of the sequences here in an article I am writing for the British Go Journal.

    They would, of course, be properly attributed, and the article would include a link to this page.

    Would that be OK with you?


    • David Ormerod says:

      Hi Chris,

      Thanks for asking and you’re welcome to use anything on our site as long as it’s attributed to us. If you scroll to the very bottom of this page, you’ll find Creative Commons license information. Hover your mouse over the link for more details and then click on it for the full details.

      And don’t to forget to let us know when your article is published! 🙂

  12. That’s great David. Thanks so much for the quick reply.

    For the purposes of full disclosure: the journal is technically made available for a price (although in reality I would be surprised if anybody buys it, as it is distributed free to British Go Association members).

    I wouldn’t be confident in saying that that didn’t violate the “non-commercial” aspect of the licence…

    The journal should be published in the next few weeks – but the online version will be behind a members-only wall for 12 months. However, I think that the SGF will be available for download, and should contain all the article text.

    I’ll let you know.