Self review

[Martin1974]

Hi everyone!

I'm quite a beginner and ony have limited access to having my games reviewed by stronger players. But I nevertheless try to learn something of every game I played. So, here is my question: How to do a decent "self review"? For which issues do I have to look? Is there for example a standard set of questions I should ask myself while clicking through my games? What kind of notes do you suggest to take? If any, that is.

Surely, everyone of you has developed his own methods. So, please feel free to tell my every trick you have!

Thanks everyone!

[Boidhre]

The simplest way is to ask yourself the some questions about the move you just played:

1) What happens if I don't play there, is it really that bad. (if it was a reply to your opponent's move)
2) What is the biggest point on the board right now? (It doesn't matter if you're wrong here, spotting these will improve in time, the important thing is to be looking both in-game and when reviewing)
3) What alternative moves have I, are the ones that look like they need to be replied really ones that need to be replied to immediately?

There are more but that's a start. The other thing is to try and find between one and three big mistakes if you can. This is helpful.

[Unusedname]

How many times does your opponent play a move that makes you wish you moved there instead of the move you made?
How many times did i think my opponent have to respond to me but he didn't?

Is a good thing you can keep track of as you play

[jts]

These were questions for self-review I posted once for another beginner who was having trouble figuring out how to gauge how well he was playing. It's possible some of them are slightly too advanced or too basic for your level, but I hope you can choose the questions that feel right and use them to enrich your own reviews.

Hi EdLee,
Certainly agree with your view: “Work on your own moves, don't worry about your opponent's level (much less their rank)”. But perhaps the qualifier is that the opponent's level (in the broad sense) is the only widely-used measure beginners have to gauge progress.

I think this is precisely what Ed denies. Here are some ways you can gauge your progress:

(i) Are your moves consistent? Do you come up with a plan and follow-through, or do you have to abandon it after the next move?
(ii) Do you play moves that attack or defend groups that are already completely alive? Do you neglect to play moves which attack or defend groups that are nearly dead?
(iii) Are you playing thin formations with tons of cutting points? Who is getting cut to ribbons, you or your opponent?
(iv) Are you playing urgent moves before big moves, and big moves before small moves?
(v) Are you looking for, and finding, multi-purpose moves that answer your opponent's threats but do a little bit of extra work, as well?
(vi) Do you play in a way that shows an understanding of sente, gote, and miai?
(vii) Are you deciding whether to invade or reduce based on an appropriate understanding of the whole-board situation?
(viii) Are you spotting ladders, nets, tesuji, and other techniques that you've been studying?

All of these questions will help you gauge whether your games were good or bad, and over time whether you're improving. Your opponent's rank is, for now, completely meaningless.

And I would add another question (which would have been implicit in -i-, -vi-, and -viii- for Tezza, but depending on your level you might want to consider explicitly): "Are you noticing stones in atari (1 liberty), stones with uncomfortably few liberties (2-4), and sequences which aim to exploit the liberty problems of multiple stones at the same time?"

Really, several of these questions can be broken down into more basic sub-questions, so if any of them seem interesting but a little vague I can try to rephrase them.

[Bill Spight]

First, play on small boards. Games are quicker and review is easier.

Second, try to discover the game-losing play.

Third, during a game form goals and sub-goals. When you review, if you failed to reach any of them, try to find out why. Perhaps a goal or sub-goal was unrealistic. If you met your goals but lost the game, ask yourself what was wrong with those goals.

Fourth, remember that often the real solution to a problem is not to get into it in the first place. Do not just ask what to do in a certain position, ask about how you got into it.

Good luck!

[Martin1974]

Thank you very much, folks! Splendid informations in just 1 day!

[Splatted]

I don't know if it's just me, but I always used to get frustrated trying to categorise moves as right or wrong since usually I just ended up deciding that I don't know what's best. Instead I like to focus on exploring the effects of different moves. You should definitely try and evaluate the results, but I don't think there's any need to find that one move that's better than all the others.

[Boidhre]

I don't know if it's just me, but I always used to get frustrated trying to categorise moves as right or wrong since usually I just ended up deciding that I don't know what's best. Instead I like to focus on exploring the effects of different moves. You should definitely try and evaluate the results, but I don't think there's any need to find that one move that's better than all the others.

I try not to think in terms of right or wrong but better or worse in terms of some quality, e.g. territory, thickness, influence, whatever. I've an imperfect grasp of all of these so obviously my analysis is going to be somewhat suspect in almost all instances. It's still useful though I imagine to try and evaluate positions since this is effectively what you're doing when you're considering any move when playing. Practice makes perfect etc.

[Splatted]

I try not to think in terms of right or wrong but better or worse in terms of some quality, e.g. territory, thickness, influence, whatever.

Trying to evaluate positions is definitely worthwhile, but it can be frustrating if that's your goal in reviewing a game. Sometimes I can spend a week studying one specific point in a game and still be unsure which weak group I should strengthen, but I can gain an understanding of how my opponent can take advantage of those weaknesses, and what kind of game is likely to result from each choice. Regardless of what I'm aiming to achieve I'll probably spend my time quite similarly, but what goal I'm emphasising makes all the difference in whether that's a week of curiosity driven exploratory fun, or a slow torturous grind resulting in failure.

P.s. I really like the idea of focusing on a specific quality; I imagine it encourages you to decide what what you want from a situation and then play in the way that best achieves that aim.


Edit @OP: I made a thread on this subject a little while ago. You might find it useful:

http://www.lifein19x19.com/forum/viewto ... =11&t=8142

[Boidhre]

I try not to think in terms of right or wrong but better or worse in terms of some quality, e.g. territory, thickness, influence, whatever.

Trying to evaluate positions is definitely worthwhile, but it can be frustrating if that's your goal in reviewing a game. Sometimes I can spend a week studying one specific point in a game and still be unsure which weak group I should strengthen, but I can gain an understanding of how my opponent can take advantage of those weaknesses, and what kind of game is likely to result from each choice. Regardless of what I'm aiming to achieve I'll probably spend my time quite similarly, but what goal I'm emphasising makes all the difference in whether that's a week of curiosity driven exploratory fun, or a slow torturous grind resulting in failure.

Sure. But that's why we talk to other players about positions. I genuinely enjoy taking a position I'm really unsure about and getting a few others and chatting about it either in real life, on a go server or on here. The others don't have to be massively stronger than you, we all have different strengths and weaknesses. Evaluating positions is perhaps best done as a dialogue.

[Splatted]

Yeah that's great fun, but I was talking specifically about when you're studying alone.

[Boidhre]

Personally, I review what I can when alone, reserve the rest for when I've others to chat to and spend the rest of my time with tsumego or pro-game commentaries. But, yeah, I do know what you mean.

[jts]

Third, during a game form goals and sub-goals. When you review, if you failed to reach any of them, try to find out why. Perhaps a goal or sub-goal was unrealistic. If you met your goals but lost the game, ask yourself what was wrong with those goals.

This is quite important. Ideally you want to remember what moves you made, and why you made those moves, and what reply you expected from your opponent, and what course of action you would follow in response to each reasonable reply. If you remember these things, you'll have a lot to think about, scry from a different perspective, and analyze during the review. If you don't remember them, then you won't have much to think about. And of course if you never had any plan to begin with, there's little chance of you remembering one afterwards! :p

This goes for any question you could ask yourself in the review. If you were already thinking about that topic in the game, you'll be able to see a lot more when you revisit it afterwards.

[Martin1974]

Thank you all very much again! Your tips have been and will be most helpfull.

[happysocks]

Rarely do this and even then will usually only go over the portions of games where big mistakes are made (typically mine) to see if any possible improvements are apparent. Thank you all who posted there are some interesting questions for consideration listed here.