What level should one reach before starting to teach TPKs?

Many who benefited from teaching as TPKs would want to pay it forward, but nobody wants to give bad advice.

(In this context, teaching refers to game reviews and teaching games, not teaching the rules or what a ladder/eye is.)

What is TPK?

I feel only D players have good enough understanding of fundamentals to teach. But I did a teach game to a 11kyu and I advised not to trust everything I said. :joy:

Whether it works or not is really not up to me to say. I have no clue.

“Twenty-plus kyu”, to differentiate them from other DDK’s.

I think you’re ready to teach anything that works for you, as long as it’s bringing the other party closer to being a better player. If you want to learn something, it’s okay to do it in intermediate phases where sometimes “wrong” or “misleading” information is taught.

For example, you would not go about teaching general relativity to middle school students in a class about gravity. You’d probably start with Newton, even though the theory is not accurate.

1 Like

To add, I think everything I “taught” or stated in game review is very basic stuff from beginning books. Not like personal experiences and readings from d players. One is probably better off just reading those books themselves, much like listening to 9p players directly instead of my relay/interpretation, which can go terribly wrong.

Subject matter expertise is not teaching expertise. Haven’t we all had, at one time or another, a high-end teacher who was simply horrible as a teacher, for one reason or another?


Very true.

However, a good Go teacher has to be a stronger player. This is different than some great sports coaches who were never good former players.

I’d say you should be around 19k before teaching TPKs.

TPKs start as people that don’t even know the rules. You can teach them what you know, and that’ll be substantially better than nothing. What an atari is, corner-side-center for openings, the basic approach joseki, whatever you’ve managed to pick up. There’s a big gap between “played once” and 19k, and you can start teaching someone how to cross that gap as soon as you’ve crossed it yourself.


I would say that as soon as you have a solid basis for how to improve, which worked for you and is repeatable and clear, you can pass it on.

I can’t see why a 22k could not help a “25k+” get out of the <900 glicko point doldrums with some very basic advice.


If you feel not strong enough to teach, just play together and learn both from the experience.

Sometimes good motivation and a friend to play may be way more useful than some tips from a higher teaching position :wink:


I agree with the posts above me. In my own words:

For teaching TPKs, who are just starting out with Go, to fill them with motivation and inspiration is much more important than to have great playing strength as a teacher.

The following pieces of knowledge will easily get a player to DDK:

  • Rules of Go
  • Life, Death and Seki
  • Connect and Separate
  • Where to find opponents and Go resources

These are very factual bits. It involves little reading and experience, so anyone can teach them.

On the other hand, I find it very difficult to teach DDKs and SDKs. At any moment, if you do not use careful wording in every explanation, you might pass on your own misunderstandings and biases in judgement. The student forms bad habits as a result.

I like to “give back“ and occasionally review games. So I try to stick to the teachings and examples that I got from stronger players and books written by professionals.


My rule of thumb is 9 stones. If you’re within 9 stones, you have little to add. Chances are that it’s just your reading that’s better. This is based largely on experience; I just don’t pretend to teach strong SDK because I feel that all I’ve got to add is different opinions. :slight_smile:

It’s also more important for the teacher to know what he doesn’t know - and where to look to find out.

Both players should be aware that anything said by the ‘teacher’ may be more or less helpful, but hardly ‘true’. The more you know how little you understand about the game (i.e. the stronger you are), the more general your advice should be. Sometimes you can say for sure that a move is better than another (when the other is a plain ol’ mistake), but it is usually more useful to give strategic advice.

When I only have a few minutes to review a game, I’ll aim to provide minimal but strategically successful variations (maybe not the best move, but clearly accomplishes goal X). I’ve been playing with the idea for a framework for DDK to simplify decision making and it focuses largely on strategic aspects. Tactical decisions (which move specifically) are a function of the current strategic goal.


There are just too many tecnhiques and basic ideas to miss, that even low level players can notice easy mistakes in each other’s games and help each other fix.
I am 15 kyu and i try to teach basics to 20k+s, review when i win and tell them which of their moves made me glad. There was nothing but very positive feedback.
For myself, I can say that I wouldn’t mind getting reviewed or even taught by equal level players, if they think they have something to contribute based on what they saw.
When you’re a newbie, it can be hard to set a study method, find material and find someone to bug with your questions. I bet most people would welcome some tutoring, even if you’re not the best player around.


Mutual learning experiences (which have been shown to be largely ineffective, if my memory of the educational research I’ve come across serves me right) are something quite different from what the OP asked about - “teaching”.

The most notable benefit a teacher can provide is scaffolding (cf. Piaget) - structuring familiar information in a way that helps you ‘grab a hold’ of the key concepts you need to understand to proceed “to the next level”.

If the ‘teacher’ barely knows more than you do, that’s something they can’t provide.

From my personal experience, there is no required rank before teaching Go. (My first teacher was 10k.) In fact, in order to teach the basics of Go, you don’t even have to be a Go player. (See, e.g., school programs for chess.)

Here’s some tips for inexperienced teachers:

  • Stay humble. Realize that your own rank is more or less the limit of how far you can train someone, so try to avoid speaking in absolute terms.
  • The most helpful thing an inexperienced player can do is convey their own love for the game, which can motivate the trainee’s own self-learning process.

And here’s some more tips, after putting on my lawyer hat:

  • Avoid being “conclusory.” In U.S. law schools, it means stating a conclusion without providing any support. The word “because” is your friend here. For example, “I would have descended here because enclosing the space does nothing if your opponent can just walk through this gap.”
  • Tie your reasoning to a “rule” when possible. U.S. law schools teach a method of legal writing called IRAC (issue, rule, analysis, conclusion). In a Go game, the issue is the bad move. To do the “R,” try to identify a rule (e.g., a proverb) that supports your preferred move over theirs. For example, “a Go proverb says ‘respond to attachment with hane,’ so I think you should have played here instead of tenuki.”

I’d +1 this. My chess coach was an awful player. He knew the basics enough to get someone in the door, but beyond that, he’d lose to almost anyone in the chess club. However, he knew a bit about how to improve, so he could, for example, tell us to study a specific opening, or assign us to do game review with someone else. We were regularly competitive at a state level, despite his almost complete inability to actually play.


I think this sounds totally sound for SDK level, just from what I see looking up. There appears to be little in the way of “framework” that you can teach or learn at that level. I say this because I’ve never seen anyone teach one.

That is not the case for DDK in particular.

There is a framework of teaching, which for me came via Dwyrin’s basics, which can be grasped and passed on to substantial benefit. I can even write down what it is (and I do that, in a gradual way, for each person I try to help with teaching).

It surely would not be perfect, because it is just a DDK passing on to other DDKs what was learned. But the students appreciate it a lot, and it advances them, so that is proof enough :slight_smile:

I think that if you have something like this: something concrete to share that helped you, then you can pass it on. That’s the main criteria for me.


1 Like

I get why everyone is posting feelgood messages here, but again, unless I gravely misunderstood OP, that’s not what he asked.

He didn’t ask “what rank should you reach before starting a club” (a topic I’ve actually addressed in another thread) “what rank should you be before promoting the game” or “what rank should you be before passing on someone else’s teachings”.

His question is quite literally

What level should one reach, before… [giving] reviews and [playing] teaching games.

Now I’ve seen some 20k+ - 20k+ “teaching games” and it was nonsensical variations upon nonsensical variations. That’s why I’m saying variations only make sense if you know what you’re doing. If you’re not at least at the cusp of reaching SDK, your strategic ideas (i.e. anything beyond reciting proverbs) are probably not going to be very useful, either. However, if you’ve got at least 9 stones on your opponent, merely by virtue of playing someone stronger, they will be able to benefit from the game.

However, I do think there’s some distinct advantage to everyone discussing their intentions for any given move or sequence, as it might help develop strategic thinking.

I do see that point.

I didn’t feel motivated to share what I have learned till I was about 12k, and what I had learned had kind of been proven to be useful to me at least in a systematic way.

But the ‘9 stone’ thing appears a bit extreme - maybe that’s one point. It may be delusion, but I feel that I can convey these basics to people around 17-18k, which is only 5 stones.

I also think that someone at 20k really could teach someone at 25k how they got there - and even though the lessons might ultimately be “non-ideal”, they might give the student something to hang on to, plus enthusiasm with that, which is worth it. A 20k can tell a 25k “dude, don’t start trying to make 2 eyes with all your stones right at the beginning” for example, which is a mistake I’ve seen TPKs making.

Maybe the number is ‘5’ not ‘9’ :smiley:


1 Like

Now we’re getting somewhere.

I don’t recall who said it, but I think it was a pro: “Anything beyond 3 stones is a teaching game.”

Now we all know that proficiency/time is best approximated by a power curve, so the stronger you get with respect to your peers, the more time and effort you will, on average, have spent, translating to experience you may very well be able to convey in a teaching game. Even if you’re “only 3 stones” stronger. Perhaps this is not the case on OGS, considering most people’s rank fluctuates by ±2 stones.

A 5-stone difference usually means the recipient wouldn’t stand a chance in an even game, which gives the “teacher” leeway with respect to the playstyle they use. This, I think, is immensely helpful as the stronger player can focus on playing “proper” moves. Whether they actually can (play proper moves) depends on individual (adapt-)ability moreso than mere rank.

A 9-stone difference gives the stronger player carte blanche. They can play entirely passively and let the weaker player experiment. They can tell you what’s going to happen ahead of time, switch colors and whatnot, they are free to choose any teaching method they like. They will always be in control of the game.


I think another worthwhile question might be: how TPK are we talking? There’s a big difference between a 20k and true, 25k+ plays-self-atari beginner. A 19k teaching TPKs won’t have a ton to add for someone close to the threshold, but definitely possesses the skills necessary to teach true beginners.

Maybe in terms of stones: 3 stones is a worthwhile teaching game, but the stronger player might probably won’t be able to control the game. At 6 stones, the stronger player can probably control the game, and tell the weaker player why they’re losing. At 9+, the stronger player probably knows what weaknesses to expect ahead of time, and can steer the game to push the weaker player to play in a way that’ll help them see and improve their flaws, as well as giving fairly accurate descriptions of why certain moves are good or bad.

Beyond that, I don’t think it matters. Having a 10k teach a TPK is probably just as good as having a pro do so. The skill of the teacher at teaching will matter far more than the difference in strength, as the weaknesses of a TPK compared to either a 10k or a pro are so rudimentary that the vastly increased strength of the pro, and the additional lessons they could teach, are irrelevant, given the lack of understanding of the student for basic fundamentals.