I've given her 13 stones and been able to beat her (this is usually close, and comes down to me getting a little lucky and saving a big group by killing another). She's been improving, though, and I don't think I'd want to give her more than 9 (and would really like to help get her past that point).

So what do you guys think? What are your big suggestions, tips, etc.?

Here are my "golden rules" for playing against DDK, beginner players.

1. Do not "let" them win, but do your best to keep the game within one or two points.
2. Try to maximize the presence of interesting "problems" that are at their level.
3. Don't talk during the game. Don't suggest why a move they made was bad, or explain why you're making moves yourself.
4. At the end of the game, give one piece of advice (a proverb, or a shape lesson, etc).
5. Play quickly, and encourage them to do the same (unless they're actually reading, which you may encourage them to do out loud, "if I move there, then you'll move there"-style).

3. Don't talk during the game. Don't suggest why a move they made was bad, or explain why you're making moves yourself.

I disagree slightly with Ross here. It depends on what kind of game you're playing, teaching or serious. I think talking, explaining, even allowing take-backs are ok in teaching games --- but even then it has to be a certain type of teaching game. You have to judge how well your student is catching on.

I disagree slightly with Ross here. It depends on what kind of game you're playing, teaching or serious. I think talking, explaining, even allowing take-backs are ok in teaching games --- but even then it has to be a certain type of teaching game. You have to judge how well your student is catching on. But as Araban says, don't overload. Keep it simple. Maybe one or two concepts at a time.

After a game with explaining, then have a game with no talking -- let the stones do the talking.

I am amazed at how appropriate the rules Ross expressed work for even up into the single digit kyu level.

Here are my "golden rules" for playing against DDK, beginner players.

1. Do not "let" them win, but do your best to keep the game within one or two points.
2. Try to maximize the presence of interesting "problems" that are at their level.
3. Don't talk during the game. Don't suggest why a move they made was bad, or explain why you're making moves yourself.
4. At the end of the game, give one piece of advice (a proverb, or a shape lesson, etc).
5. Play quickly, and encourage them to do the same (unless they're actually reading, which you may encourage them to do out loud, "if I move there, then you'll move there"-style).

The biggest lesson I've learned when it comes to teaching games is to make sure you don't give the beginner an overload of information. Keep it casual, simple, and interactive. I also think it's helpful that you go over the basics to make sure that you understand concepts well enough to explain - I don't mean this in a condescending way; on GD I made a thread for instance on how to explain the concept of 1 eye to a beginner in a clean and precise way, a concept that even a 30k knows! As Einstein used to say, "If you can't explain it to a six year old, you don't understand it yourself."

Another question, though. For number 2, you recommend maximizing the interesting "problems" at their level. Curious if you could expand on this a little bit (maybe others have input as well)?

It's not a matter of making purposeful mistakes--you should never do that even in a teaching game, in my opinion (although I might quibble about "slight inefficiencies"). I'll try to give an example of what I mean.

At the very beginning level, it's simply continuing patterns of play that at a higher level wouldn't be continued. For example, Straight Four In The Corner (or Twisted Four) may arise in the game naturally. If you take one of the points inside, your beginner opponent may have to think to determine the status of the group and where to play. For an opponent of your own level, this move would have been a waste of time and instantly answered, but it can provide valuable reading practice for a beginner who has not encountered this shape endless times. Yet your move was not a "mistake" (unless technically there were bigger moves elsewhere and your opponent ought to ignore your threat to kill the group), but at the same time it is "maximizing problems that are at their level".

At a more advanced level (both for yourself and your student), it becomes an issue of deciding which path to take where several will do (and in this case I grant that inefficiences may arise, but nothing a kyu player would call a "mistake"). For example, you may choose a local variation that includes a ladder rather than a different way of playing, because you want to test your opponent's ability to read the ladder (and suitably punish him if he falls short). Or you may choose an attack that is best defended by a hanging connection rather than a solid one, to ensure that your opponent is appropriately familiar with such connections (and punish him if he falls short). Or you may notice that they continually push from behind, and you make moves that take advantage of this tendency. If the same mistake is repeated again and again throughout the game without the student gaining the "aha!" moment on their own, that can be used as the #4 lesson at the end of the game (and hopefully they will remember seeing that pattern enough times in the game that they will connect it in their minds).

To do this well, I think one must not only be very skilled at Go, but also very skilled at this particular art. I know I can only do it at a very shallow level of what's possible, providing good fodder for 20kyus at best. Perhaps it requires a dan-level player to "lead" 15kyus into "interesting problems at their level"--and not just any player at that level, but one who is particularly good at this skill as well.

I will say that trying to practice it makes for much more interesting games against beginners than otherwise someone at a huge rank difference would otherwise get. If it benefits the beginner as well, then it's win-win all around.

This mostly applies online, but it can happen in person too:

The only recommendation I have is regarding something I've seen happen way too many times.. Please do not win by review if you ever happen to lose to anyone you are teaching. Accept the loss, review the main key points that looked awkward for your opponent(mainly review the idea behind their moves contradicting their moves played. Ask them why), and possibly review your biggest overplays(not every slight one). Honestly, though, do not overload your student. Nothing good comes from feeding people too much information at once.

There's also another one, when you start teaching people, you fall into a teacher role more and more often. If you see someone else conducting a review, don't back-seat teach or interrupt someone else's review of a game that you think is not good enough. If you feel it is extremely important then possibly point it out, otherwise enjoy. Everyone has different styles of teaching. There is absolutely no reason to try to upstage another reviewer.

I really like ross' list. +1

You never answered my question, though. What board size are you playing?
I've found that getting beginners to start on small boards helps them get more games(and therefore more patterns) under their belt.

I've been trying to get people I've just taught the rules to start on a 5x5, and play out several games till the end, only correcting them when they do illegal plays. The reaction seems good. It doesn't tire them as much as playing out a 9x9, and they seem to improve almost every game.

but I think a 5x5 at this point would be a little small.