Best Time Controls For Improving

[amnal]

Other possibilities are:
- even if you learn less per game, you can play enough extra games that you ultimately learn more
- fast games teach different things, such as improving the ability to recognise shapes automatically

OK. What is the mechanism? How can a fast game teach that? Here is how I improve my ability to recognize shapes automatically:

Step 1: I don't know a shape.

Step 2: I play a game and I think hard about a move and I pick the wrong shape, because I have no experience with the right shape, and I can't see deeply enough to know that it's right.

Step 3: When analyzing the game, I identify the fact that I wasn't satisfied with my move, or I find that it led to a bad result. I either dig in deeply and discover that the right shape would have worked better, or a stronger player points it out to me.

Step 4: I think about it for a few minutes until I think I understand why the right shape is right. I try to think of other times when this shape would be useful.

Step 5: Because of all this thinking and the context of my game, I stand a very good chance of noticing this shape the next time I have an opportunity to play it. Occasionally I remind myself of what happened in my game to reinforce my memory of it.

Step 6: Now I know the shape.

I think we can agree that if you follow those steps, you will wind up seeing lots of good shapes! There's no mystery step where a miracle happens.

If I were playing a fast game, step 2 doesn't work. I make mistakes because of time pressure, so it is more difficult to identify what errors were the result of time, and which errors were because I was genuinely confused. Steps 3 and 4 only work if you spend a lot of time in the review. If your future games are fast games, step 5 doesn't work very well, since you often won't spend enough time on your moves to say "Oh, two weeks ago my game looked like this. I remember the right idea was X. Why was it the right idea again? Does it apply here?" If you notice it at all, you'll just have to guess about that stuff.

So this process seems like it doesn't work well if you play fast games and it cannot work if you do not review.

What is your alternative process?

You're clearly a methodical, logical and dedicated learner. I'm sure your method works extremely well for you, but I maintain that it's short sighted to say it's the one true way. Specifically, for everyone and in all situations.

Also, when I say 'fast', I don't mean 'you must play every move quickly' so much as 'not spending ages thinking about every move'. The problem isn't that the game must be finished quickly, but that (again, in my opinion) the extra thought is not so beneficial as just playing something and seeing what happens.

For example, I have recently been teaching some beginners. Beginners have almost no clue what to do, because a lot of the way experienced players choose moves is by automatically pruning the search tree significantly. Or, more simply, we know what sorts of moves might work well. Often the beginner thinks extremely hard about an atari-connect situation in the middle of the board, plays it, then is surprised when I simply tenuki and take a bigger point on the side. Yes, the beginner learnt a little about reading there, but far more important is the surprise that the opponent might play somewhere else entirely. There are many such surprises for a beginner, and I think that simply seeing them is more important than necessarily reading every one of them out themselves.

I use the example of simple beginner moves because it's easy to describe and something I've been thinking about recently so I can remember it, but I've always felt that similar effects are important right up into sdk and higher. For instance, my feelings about what will happen in a fight are significantly influenced by simply having seen hundreds of similar situations and having a feel for what kind of things will work or not. This is part of what allows me to prune the decision tree and look more closely at the few important lines, but I don't think I could gain that knowledge without having tried many different things in many different games.

I do feel that this is less true as experience increases, and deep understanding can more readily be gained by deep thought about the situation. In that case, the methodical approach is the best way.

[hyperpape]

I remember being bad enough that I often didn't see that in this position, a was a threat to capture, even though the White stones had two remaining liberties and two stones that would prevent a ladder. I was only reading b as a follow-up.

$$
$$ . . . . . . .
$$ . . X X X . .
$$ . X O O O X .
$$ . O a b a O .

Click Here To Show Diagram Code

[go]$$
$$ . . . . . . .
$$ . . X X X . .
$$ . X O O O X .
$$ . O a b a O .[/go]

Isn't it obvious that you'll overcome mistakes like that just from seeing them over and over again? And then keep playing, and seeing the position, and it eventually becomes hardwired, second nature to see that, which is how you progress at Go--things stop being work and start being obvious. Who knows how long that goes on--surely you don't get to be 9 dan that way, but it goes a long way.

[jts]

What is the actual source of the "lose 100 games quickly" proverb? The rhythm of the proverb, at least in English, owes something to the wisdom of Silenus, so when I first heard it seemed blindingly obvious that the "quickly" meant "as soon as possible". Like that, the proverb actually makes lots of sense, particularly in a time and place where there weren't many beginners at any one time. The beginner wants to see quick improvement, wants to stop feeling bewildered, and the answer is that improvement comes gradually, one game at a time. (The bewilderment only deepens, but that's a different issue.) --- Now that most beginners prefer to play other beginners, it's not quite so certain that the games will be losses, but patience is still valuable. --- Most people, though, seem to interpret the 100 games/losses as presenting a concrete obstacle to smashed through as quickly as possibly, possibly by playing blitz games or games on small boards.

[cata]

I remember being bad enough that I often didn't see that in this position, a was a threat to capture, even though the White stones had two remaining liberties and two stones that would prevent a ladder. I was only reading b as a follow-up.

$$
$$ . . . . . . .
$$ . . X X X . .
$$ . X O O O X .
$$ . O a b a O .

Click Here To Show Diagram Code

[go]$$
$$ . . . . . . .
$$ . . X X X . .
$$ . X O O O X .
$$ . O a b a O .[/go]

Isn't it obvious that you'll overcome mistakes like that just from seeing them over and over again? And then keep playing, and seeing the position, and it eventually becomes hardwired, second nature to see that, which is how you progress at Go--things stop being work and start being obvious. Who knows how long that goes on--surely you don't get to be 9 dan that way, but it goes a long way.

I wouldn't say it's obvious, but yeah, some simple things you will just pick up subconsciously no matter what you do. It's not obvious that you will pick up those things faster by spending time playing blitz rather than playing slow games, though. And I don't see what the advantage is in sitting around waiting to try to pick them up instead of learning them on purpose!

[yoyoma]

A dramatic reenactment:

Step 1: I don't know a shape.

Ahhhh how did I get myself into this mess?! Whyyyyy??

Step 2: I play a game and I think hard about a move and I pick the wrong shape, because I have no experience with the right shape, and I can't see deeply enough to know that it's right.

What do I do? What do I do! Ummmmmm, here! Bonzai!

Step 3: When analyzing the game, I identify the fact that I wasn't satisfied with my move, or I find that it led to a bad result.

Waaaaahhhhh all my stones died! What happened?

Step 3b: I either dig in deeply and discover that the right shape would have worked better, or a stronger player points it out to me.

Stronger player: Dude everyone knows you should just play here man.

Step 4: I think about it for a few minutes until I think I understand why the right shape is right. I try to think of other times when this shape would be useful.

What? Why would you pla.... oooooooh!

Step 5: Because of all this thinking and the context of my game, I stand a very good chance of noticing this shape the next time I have an opportunity to play it. Occasionally I remind myself of what happened in my game to reinforce my memory of it.

Oh man that game will haunt me forever. Never again!

Step 6: Now I know the shape.

[hyperpape]

It's not obvious that you will pick up those things faster by spending time playing blitz rather than playing slow games, though. And I don't see what the advantage is in sitting around waiting to try to pick them up instead of learning them on purpose!

At the time I made that mistake, I was definitely able to see that 'a' captured (and I was missing it in non-blitz games). It wasn't something I could not literally read. It was something I was overlooking. So I'm not sure how I'd study it.

Maybe the difference between you and many other players is that the things you see, you remember a great deal more easily. For many of us, it seems to take more work to get something through our thick skulls.

[Mef]

I'm a chess expert (USCF ~2150) but only 2k Go player (I've been playing just for a year and a bit) so take this with an appropriately-sized grain of salt.

It totally baffles me that someone can claim that fast games are better for improving than slow games. *snip some details*

The "lose 100 games" advice is unbelievably weird. I can't imagine how this advice can really be how someone learned Go; more likely you play 100 games and you're still 20k. When you take a college course, do you try to learn the material by quickly skimming 10 textbooks but not doing any exercises?

I think you are misunderstanding the nature of this advice. This is not meant a a learning strategy for someone who is familiar with the game aiming for targeted practice at their weaknesses. This is advice for someone who has just learned the rules, and is based on the fact that the fundamental strategy and tactics in go are not readily deducible from the rules. When you are first starting out the vast majority of your moves will be nonsense. This is not meant to detract from any beginner of the game's intelligence, it's a simple fact. When I first started out most of my moves were nonsense too...spending a long time on these moves means you are spending a lot of time on nonsense. As you play you will get start to get a feel for what's really important in the game (cuts, connections, basic life and death, basic capturing techniques, very basic push/block sequences, etc), much moreso than you would from just reading books at this level. Much of the lessons you are learning are not things that require deep analysis, they are things like "Oh, I need to watch out when my stones are in atari" or "That first line block needs a protective move or they can still cut"

In fact your statement "you play 100 games and you're still 20k" encompasses it perfectly, only you made a different value judgement than I would have. I would say "If you play and lose 100 games you're already 20k! So you might as well just be playing games quickly until you hit 20k." With the benefit of online play it's not hard to play 100 fast games in a week. Going from complete beginner to 20k in 1 week would be reasonable progress in my eyes.


The comparison to a college course is not a good one, a college course assumes you already have a reasonable level of proficiency in whatever you are learning. A better example is children learning their multiplication tables -- often they start with simple repetition, often at a rapid pace. Or perhaps riding bicycle....you could spend hours and days reading about the physics of bicycles and whatnot....or you could just get on it and try to ride. Which of these two people will be riding their bike sooner?

[flOvermind]

I wouldn't say it's obvious, but yeah, some simple things you will just pick up subconsciously no matter what you do. It's not obvious that you will pick up those things faster by spending time playing blitz rather than playing slow games, though. And I don't see what the advantage is in sitting around waiting to try to pick them up instead of learning them on purpose!

I think the real question is: Do you really learn these things better by deep analysis of the position, compared to just seeing the position happening a few times?

Especially for beginners, it often happens that they try to analyze a position, then play a move, and the opponent plays a reply that they didn't even consider. From that, you learn just one thing: Next time, I should consider that response. Or, on a more basic level: Next time, play a different move.
But you could have learned the same thing by just playing a some move quickly, after briefly asking yourself "Do I remember a similar position? What did I play then? What happened?". With enough repetition, these questions become automatic, with no real effort involved. The advantage of this method is that it takes significantly less time, while the learning effect is almost the same.

Of course, as you get better, your ability to analyze positions improves, so the above gets less applicable. Hence the proverb "Lose your first 100 games quickly", and not "Lose games 10000 to 10100 quickly"

[cata]

Especially for beginners, it often happens that they try to analyze a position, then play a move, and the opponent plays a reply that they didn't even consider. From that, you learn just one thing: Next time, I should consider that response. Or, on a more basic level: Next time, play a different move.

Maybe, but even for beginners, it seems to me like that you just don't learn very quickly that way. For example, if a new player looked at hyperpape's position from last page, studied it for a minute, and didn't realize that A was a threat to capture, then on one level, they might learn: "Next time, I should consider A!" That's fine as far as it goes...

...but if they are surprised they didn't see A, and they think about why they didn't see A, they might learn: "When I have a couple stones clumped up that are surrounded on three sides, they might get in trouble, even if it looks fine at a glance. I should double-check potential moves there to make sure they are really OK." That is the kind of thing to learn that really pays rent.

But you could have learned the same thing by just playing a some move quickly, after briefly asking yourself "Do I remember a similar position? What did I play then? What happened?"

Definitely -- but that doesn't sound to me like the kind of thing you are thinking very hard about if you are only taking ten seconds per move!

Let me give an example from my games, which is a pretty basic example but not a beginner example: A few weeks ago I played a game and I did one of those 3-4 approach josekis where you jump out and the other guy jumps out along the side. I had the opportunity to play A:

$$
$$ . . . . . . .
$$ . . . . . . .
$$ . . . . a . .
$$ . . O . . . .
$$ . . . . X . .
$$ . . O . . . .
$$ . . . . X . .
$$ . . O . . . .

Click Here To Show Diagram Code

[go]$$
$$ . . . . . . .
$$ . . . . . . .
$$ . . . . a . .
$$ . . O . . . .
$$ . . . . X . .
$$ . . O . . . .
$$ . . . . X . .
$$ . . O . . . .[/go]

in a situation where it would have sealed black into the corner fairly securely. A stronger player pointed out to me afterward that I didn't play it, although it was very big. Since I had taken some time on the game, I knew that I had considered it, but didn't play it because I thought another move on a different side was bigger. When I looked at it, I didn't see why it was big in this case, but after playing out some variations in review, I could appreciate that white got a lot of power by sealing black into the corner. So I learned that I might be intuitively underestimating the value of sealing in a group like that, and I should think twice about it next time when I see a similar opportunity.

A few games later I encountered this position as white after starting on 3-3:

$$
$$ . . . . . .
$$ . . . X . .
$$ . . . . a .
$$ . . O . b .
$$ . . . . . .
$$ . . . . . .

Click Here To Show Diagram Code

[go]$$
$$ . . . . . .
$$ . . . X . .
$$ . . . . a .
$$ . . O . b .
$$ . . . . . .
$$ . . . . . .[/go]

I don't usually play 3-3, so I had not seen this move before. My initial instinct was to tenuki, because there were some other moves I wanted to get, and it looked like my corner stone was still safe. But then I remembered my previous experience about seeing the value of sealing stones in, so I gave a second thought to analyze what would happen after black followed up with A or B. After looking at it, I decided it was important to not allow that, so I played A, which I think was a much better choice in that situation.

That is the sort of thing which I could not learn very easily if either my first game or my second game was a blitz game. It would take a lot of experience to teach me the same lesson if I hadn't stopped to think about it.

[mlund]

Losing 100 games is great for some things, worse for others. It depends on the nature of the losing experience.

Losing 100 games against other beginners without any more experienced players or teachers to give you analysis is a terrible, terrible waste of time. You can still learn in that situation, but you learn so much less than if you had a teacher. Respect the idea of Handicap Go through every stage of your learning.

My own learning experience is very far from complete, but I feel like first 10 concepts to grasp ought to be something like this:

1.) Rules

2.) Surrounded!

3.) False Eyes and Double Atari

4.) Ladders and Nets

5.) Basic Shapes (Nobi, Tobi, Keima, Tiger's Mouth, Ponnuki, Bamboo Joint, Empty Triangle, Broken Shape)

6.) Corners, Sides, Center

7.) Properties of Board Edges

8.) 3-3 Invasion Joseki ("the hoshi does not face the corner")

9.) Basic Shapes 2: Life, Death, and Miai

10.) Walking

With an beginner, Handicap Go is a way for the more experienced player to create examples the student can learn from. All these concepts can be illustrated in handicap games. Introduce the concept before the game. Play the game and pay deliberate attention to the opportunity to make one or more of these concepts relevant. Record the game and then do an analysis, emphasizing where that concept came into play. Show variations.

For these purposes shorter time-limits are probably best. These are fundamentals that need to become instinctive to a player for him to progress. More advanced players are going to use clock time to read larger problems to distill them down into these fundamentals at the conclusion of the sequence.

Once a beginner has gotten over those first hurdles I think they're ready to start doing more independent reading to start digesting core Fuseki, two or three 4-4 corner joseki, and practicing Tsumego for Life-and-Death and Tesugi problems.

- Marty Lund

[yoyoma]

Ok a more serious post from me:

I think it's important to put some concrete numbers when they talk about fast/slow/blitz games. I think we can all agree that 1s per move is too fast to be optimal. And on the other end thinking for 1 year on each move can't be optimal either!

For me, I interpret the advice to lose 100 games quickly to mean:

1) Play some games on 9x9 boards, if you use time controls at all use 5+5x:30 or similar.

2) When you move to 19x19 boards again use something like 5+5x:30 or similar for the first several games until you reach some decent level where longer times will be more productive, say 15kyu or so?

Also I view the primary purpose of the advice is to counter the tendency to want to not play at all as a beginner out of fear or shame, rather than some sort of warning against the dangers of playing too slow. I reckon many people confused by the advice aren't the type to avoid playing games because of this, they think it's obvious that you should play to get better!

Instead of playing 5+5x:30, playing 30+5x:30 is fine too. The only bad thing is a lot of people have a tendency to think in circles when faced with a situation they don't know what to do. That's the primary target of those advising to move more quickly -- the act of just waiting for inspiration to strike.

On the other end of playing faster, I would argue that you can learn by playing many 1+5x:10 games and then reviewing them carefully afterwards, but it probably won't help as much as playing 5+5x:30.

[Mef]

Instead of playing 5+5x:30, playing 30+5x:30 is fine too.

From my experience, in most casual (read: non-tournament) games these are the same time setting.

Just for fun, to see if my suspicions were warranted - I just did a quick check of 10 non-blitz games from the EGR on KGS in rank ranges 5k to 20k (time controls ranged from 55:00 :30(1) to 5:00 :30(5) but mostly somewhere in between). I found the average time per move prior to reaching byo-yomi ranged from 4.5-8.5 seconds per move, and in all games combined the number of moves I saw that took more than 30 seconds to play prior to byo-yomi was less than 10.

I then decided to look inward....checked out my last 5 games (played at KGS automatch medium timesettings which is 25:00 + :30(5)). I found the games still averaged around 7-9s/move, but there were on average about 5 moves total each game that would be greater than 30s, and of that usually 1 or 2 moves that was more than a minute. Curiously though, those moves that took longer were not near the opening, but instead were at late midgame and the time was taken counting the board, not analyzing variations (at least, that's what I used it for, I can't speak for my opponent, but judging by their moves that's what they used it for too.). I'm fairly certain that these games would have been unaffected by being played at 0:00 + :30(5), and certain they wouldn't have been hindered if it was 0:00+ :30 (10).

Overall perhaps not enough to qualify as a significant dataset, but I'd say it's reasonable to suggest that when you are a beginning player as long as you have :30 (5) the main time is inconsequential.

[Josh Hatch]

I'm kind of a slow player I guess. When playing with 30min + 5 x :30 time settings I'm usually in byo-yomi somewhere between move 60 and move 100. That's an average of 35 - 60 seconds per move (approximately). Of course I also use whatever time my opponent uses to think as well.

I definitely agree that beginners don't need more than 20 - 30 seconds per move though. When I first started, and I see it when I watch weaker players games as well, I would spend a while thinking and still play the wrong move. It'll probably continue to happen to me more often than I'd like as long as I continue to play Go no matter how strong I get.

[xed_over]

I think some people are taking proverbs just a bit too literally -- proverbs aren't meant to be taken literally.

It doesn't matter if a beginner's first number of games are won or lost, played fast or slow. I think the point is to get a good number of games under one's belt as soon as possible in order to gain some practical experience, because most of your initial learning is going to come from playing rather than from studying or being taught.

(and when I first heard this proverb, it was only 50 games)

Taking any proverb too literally, you'll end up with this Gieco commercial...

"A bird in the hand is worth two in the bush"
http://www.youtube.com/watch?v=vWj6fh-IJ5Q