I came across this reddit thread discussing the main difference between SDK and dan level players.
The top comment mentioned this:
This will vary quite a lot from player to player, but in my experience, the main difference between an average kgs 4k and an average kgs 1d is what they are thinking about as they play.
At 4k most players will frequently worry about territory a lot, and play moves to make territory for themselves or to reduce or destroy opponent’s territory.
A 1d will spend much more time worrying about what groups are strong and weak, and often choose strategies based almost entirely on this issue, refusing to worry too much about the immediate territorial impact of these choices unless said impact really drastic.
A large part of what got me to dan level was to really make my peace with the notion that weak groups trump pretty much every other strategic consideration. When I review games with single digit kyu players, I frequently point out to them that when I talk about go I use the words “strong”, “weak”, “sente” and “gote” very often, but “territory” almost never, whereas for them it is usually backwards.
I find this very interesting. As an SDK myself, I do focus a lot on territory in the game, and frequently worry if my opponent is starting to make more territory than I do.
I also focus on what groups are strong and weak, because I can find ways to benefit from attacking my opponent’s weak groups. (And not get attacked and let my opponent benefit). “Benefit” here means primarily making/securing my own territory.
Basically, territory is the main thing what I think about when forming in-game strategies. I didn’t see anything wrong with that since it’s a game about territory.
This comment now makes me wonder if I have been thinking about it the wrong way. How do Dan players think about this comment?
But it’s also a game about making more territory than your opponent.
So you can make territory, make territory, make territory, all the time, every move, but if it’s not large enough it’ll either make the game very difficult or impossible depending on the situations.
It might require living in areas where the opponent is strong with a lot of influence, or playing more risky reductions than you might want.
Territorial play seems to be in style though, as Katago and other modern ai have been very effective at destroying moyos.
Totally. When I say I focus a lot on territory, that includes both making my own territory and reducing my opponent’s, so I often try to not allow my opponent make a big moyo, unless I could also make a similar sized or larger moyo.
In my ～5k level games, my opponent seems to focus on the same, so we usually play an opening where neither player could make a big moyo and territory is rather separated across a few different groups.
I am not really sure if there’s something wrong with that.
I agree with it, which is why my comment, the 2nd most upvoted on that thread after the one you quoted, was basically saying the same thing in fewer words plus a book recommendation:
Around that level, improving your understanding of how strength and weakness of groups impacts the game flow is important. Attack and Defense book from the elementary go series is good for this. Being better at this is one of the key reasons dan players beat mid kyus barely thinking.
Something else to add, think about reductions more, and invasions less (ponder how that’s related to strength of groups).
I would add though that around 4-5d level, people start thinking about territory more again
This all makes a lot of sense. I guess my mistake is thinking too short term here. I always try to look for immediate benefits (in territory most of the time) when I attack my opponent’s weak groups.
But if I just keep my group strong and my opponent’s group weak, at some point I will be able to benefit. I don’t always have to think about how that converts immediately to territory benefits.
Thanks for the insights! I have a feel for what you mean, but don’t understand it well at my level.
When I played a 5 dan before, I don’t feel like making a lot of bad moves, but in the middle game suddenly I find 2-3 groups of mine are vulnerable and all his groups are strong. Then everything collapsed once he started attacking.
I think I know what you mean - Whenever I invade and try to live inside of opponent’s territory, I inevitably make his surrounding groups strong and form great influence.
What’s missing for me is a better understanding of how those thick walls/strong groups affect “game flow” like you said, or how one should benefit the most from them.
Once I asked a 5-dan player pointing to one of his thick walls: “This group looks like it has a lot of influence, but I don’t see how it could turn into territory later.” He laughed and said “influence always turns into territory.” I didn’t fully understand what he meant.
I will check out Attack and Defense, but would really appreciate it if you have a dumbed down explanation for an SDK level player on this
Go is essentially about territory so there is nothing wrong in thinking about territory. However, taking territory is not the only way to end up with territory. Using strong/weak groups is one of them; in a way, it’s better because if you do well you will naturally end up with territory instead of taking it deliberately.
However, this is actually a very difficult concept that not even Dan players can master. Many pros also take too much territory and leave too many weak groups and suffer because of that, but many can also settle their weak groups very easily. It’s all part of the strategy, so it’s good to try out different things and see which one works better.
Nick Sibicky often said it was better to chase a weak group across the board than to kill directly, because killing only gives you points in one area but chasing a weak group makes you strong all over the board.
Re: invasions vs. reductions. Reductions seem to remove fewer points but are more difficult to attack compared to invasions so are often equivalent or better than invasions. On the other hand, invasions are often good when they aim at capturing stones, or at least at making profit while attacking stones. In this game for instance, this move
is not merely an invasion but it is a ladder breaker (see O9) (actually O9 can be captured in a net, but J3 did contribute later to attack when Black saved O9), and in addition aims at attacking the three black stones.
Interesting example. I thought “reduction” means reducing territory without causing a fight.
I know some version of another Go proverb is when you get attacked, make sure your opponent is also attacked. Being the only weak group around opponent’s strong groups just gives them a lot of free sente moves to build territory / become stronger elsewhere.
I feel like this is a really good example of that.
Yes: the stronger you get the more obsessive you are about weaknesses. Not just weak groups, but weaknesses in general. But there’s more to say.
A 4k player doesn’t come out of nowhere - it generally took a lot of work to get there. As a DDK they learned about big moves (corner → side → center), but they lost games because they didn’t know how to defend their stones. So then they learned how to make strong groups in the opening, but they got overly defensive and started playing puppy go. So then they learned the value of sente, and they defend their groups just enough so that they can’t obviously be killed in one or two moves, and then go off and play a big move or attack something. That got them to 4k or so.
To reach 1d, the player needs to understand the difference between “strong” and “can’t be killed”. If your opponent can pressure your group so that it’s forced to scratch out life with 4 points in gote in the corner, that group isn’t strong. It isn’t dead, but your opponent gets to whack it like a point piñata. Likewise if you build a territorial framework but leave behind shape defects that allow your opponent to invade easily, then once again your stones aren’t really strong. This is tough to learn because it’s about timing the global position rather than just evaluating a local exchange - if you fix a weakness too early then it’s slow, but if you wait too long then your position collapses.
The next step after that is learning the difference between “efficient” and “strong”. In games between strong amateurs or between professionals, you’ll often see one player leave a clump of stones with little eye shape right next to an enemy fortress. The stones definitely aren’t alive, but they aren’t being defended, and mysteriously they aren’t being attacked either. What’s going on? A group can be “de facto strong” if there is no efficient way to attack them. Sure you can chase them into the center, but you leave behind a bunch of weaknesses and your position doesn’t allow you to profit from the attack. And if you try to too hard to capture them then your opponent will let you spend 5 gote moves to secure 25 points, instead making 30 points by taking the rest of the big moves on the board.
This is actually very accurate. It sounds a lot like what’s going on in my games lol. Do you think there’s something fundamentally wrong about this 4k player’s way of thinking?
At my level, I don’t see how that’s wrong yet.
I mean, I understand the difference between “strong” and “can’t be killed,” and I try to make certain groups stronger to help with attacks / forming moyos, but like you said, I usually think about the big moves / attacking first, and just hope to make my groups stronger naturally, as this feels like more efficient.
No, it’s not wrong! The progression I’m proposing is something like:
Big moves (DDK) → Don’t die (SDK) → Take sente (5k-ish) → Play solid (1d-ish) → Play efficiently (5d-ish)
Each step in the progression is a refinement of the previous step, not a refutation. To improve, the typical 4k - 5k player should stay hungry for sente and big moves but start to pay closer attention to the weaknesses on the board. This could mean playing a solid defensive move early, not as a reflexive response to the opponent’s move but because it permanently diffuses counterplay in a key part of the board. Or it could mean ignoring the opponent’s threat because the damage can by mitigated by exploiting a key shape weakness.
(Of course every player develops differently - a given 4k player could have a dan-level understanding of attack and defense but a 9k understanding of endgame, for instance.)
