Which ranking would be practically impossible to beat (not reach but beat) for a beginner?

A "beginner" is defined here as having played less than 200 games.

Also, let's say rankings are based on the KGS ranking system.

Also, if your answer to the question is "it depends", try to come up with a rank which would be absolutely impossible to beat. Would really a 9 or 8 dan player be possible to beat within 200 games? Probably not.

As a beginner, I am very interested in this, since this ranking, whatever it is, would be nice to reach.

Cheers!

Are you talking about KGS? Then there are ways of winning and losing games that have nothing to do with rank, so the answer is there is no level.

I am just basing the ranking on KGS.

I exclude all kinds of cheating or weird wins and losses that don't have to do anything with skill. There are so many different situations. Hard to define exactly, of course.

Well, there are people who come from other games that involve some similar skills, such as chess, who can be SDK almost immediately, certainly within 200 serious games. I have no idea how strong a player who grew up in the one of the classic go houses in Japan would have been after 200 games, considering the kinds of games they would have been playing and---more importantly---who'd they be playing.

So basically what I'm saying is there's a problem with your definition of beginner.

I don't think anyone will be able to give you a definitive cut off point, just a range of ranks over which it becomes increasingly unlikely a beginner will win.

_________________
Regards,

Peter

Well, first you have to clarify whether you mean in an even game or in a handicap game. Second, you have to say whether you mean regularly or rarely. In a game that is mis-handicapped by one stone, B wins 1/3 of games, and then fewer and fewer for each additional handi you remove. You can choose the probability that you consider "possible", and then calculate the rank gap you need based on that.

Do you also exclude the situation in which the 'beginner' plays 200 games, say, over the course of four years - one per week - and enlists the help of a teacher to review the games, and studies many books full of go problems and so on? (This is a rather exaggerated situation, perhaps, but not as exaggerated as you might think.) It wouldn't surprise me if such a person could reach mid-dan ranks.

When I say that it's not as exaggerated as you might think, what I mean is that I know people who've reached 15-12 kyu within a matter of a few days, and I know people who have played for years and never reached that level. Number of games played is one metric, but it's not the only one or the most useful.

For me, it's not the question that bothers me so much as the attitude that's reflected by the above statement. What's your motivation?

I agree, I can't understand why you're looking to maximize your rank such that your total number of games remains below 200. I can understand wanting to place a restriction on the time it takes, but number of games?

_________________
We don't know who we are; we don't know where we are.
Each of us woke up one moment and here we were in the darkness.
We're nameless things with no memory; no knowledge of what went before,
No understanding of what is now, no knowledge of what will be.

I read it that the OP wants to know what rank he needs to reach so that he never gets beaten by someone who has played less than 200 games.

_________________
Regards,

Peter

You're right, I misunderstood.

I really don't know what the answer is. One of my friends, who I introduced to the game, is at around 170 games on KGS and has played at least 30 games offline, and is just below 12k. I would think that would mean he would have an extremely difficult time beating a 9k. Of course, someone who for some reason set out with the goal to become as strong as possible while playing less than 200 games would be stronger than him at the 200 game mark.

_________________
We don't know who we are; we don't know where we are.
Each of us woke up one moment and here we were in the darkness.
We're nameless things with no memory; no knowledge of what went before,
No understanding of what is now, no knowledge of what will be.

Exactly! I think having that ranking is a definition of being good at go. That's my motivation.

Number of games isn't a great way of looking at it. Some people can improve very quickly based off fewer games played and plenty of study on the side (tsumego etc), or simply better quality games, 200 blitz as a beginner versus 200 slow games against far stronger players. Define as someone playing 3 months or whatever and you'll get a better answer, though even then go prodigies will mess things up for you.

I disagree that being able to always beat beginners at any game, not just Go, defines being good at the game. Good players can always beat players who can always beat players who can always beat players who can always beat players who can always beat beginners at a game.

_________________
We don't know who we are; we don't know where we are.
Each of us woke up one moment and here we were in the darkness.
We're nameless things with no memory; no knowledge of what went before,
No understanding of what is now, no knowledge of what will be.

It may be the OP wants to avoid the situation of sitting down across from someone, getting thrashed, and asking how long they've been playing and getting "3 months" back as an answer after they themselves have been playing two years. This is frustrating in almost any activity. Rather than actually thinking 5k or whatever is being good at go.

Then, as all the rest of us said, it hugely depends on what kind of sub-200-game beginners you're thinking about.

Then don't ask.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

Very interesting and unusual question. Interesting.

Still, if I had to answer this, I see two aspects.

1. What rank can one reach within 200 games, and
2. What rank will always beat that rank.

Taking into account that ranks indicate only statistical likelihood of beating a person, there might never be a 100% (or 0% chance.) Which brings up a third question:

3. What winning likelihood would you accept? 10%? 1%? 0.1%? 0.01%? Or what?

Question #1 is crucial here, and I have no clue how to answer. Not sure if anybody does - it just depends on too many factors.

In more practical terms, as a very inexact approximation, I'd say this:
1. Lets assume average person can reach up to 10k if they are really trying hard, and
2. Lets assume 1d has enough winning chance against 10k to satisfy you.

So just get to 1d and you're golden. Heh...

_________________
- Bantari
______________________________________________
WARNING: This post might contain Opinions!!

I would say 0.1 %, one in a thousand.

I think this was a very interesting answer.

Since you have defined the expected win percentage, which is 0.001, I can answer the problem from, say my side, using ELO rating system and EGF ranks.

I am pretty sure I haven't played 200 19x19 games yet, but I might be close or over if you count the smaller board games. My rating is 5-6kyu ish KGS. Let's say, I am the guy and I am 6kyu.

According to ELO formula the expected win percentage is magnified 10 times with each rating difference of 400. So for 0.001 chances, you need a rating difference of 1200.

So with EGF ranks: 6kyu = 1500 ELO rating.

That means we need a guy with 2700 rating. Which again with EGF ratings is equal to 7 dan amateur or 1d professional.

Note: This is a quick calculation and is probably wrong, since according to reference, EGF modifies the normal ELO formula somewhat. But 0.001 is a mighty difference.


Ref:
http://en.wikipedia.org/wiki/Elo_rating_system
http://en.wikipedia.org/wiki/Go_ranks_and_ratings