I want to know my rank

[andy12514]

Can anyone tell my approximate rank (online) according to this game? I played games with my friends offline usually.

http://gokifu.com/share/3b6j

Edit: Sorry I forgot to mention. This is a game between me and my friend. I was black.

[EdLee]

Did you self-play as both and ?
Many super-human bots are now readily available; e.g. on KGS.

[dfan]

I am pretty bad at judging games at this level but I will guess 15k; 20k feels too low but 10k feels too high. Of course the way to actually find out your rank is to play ranked games.

[CDavis7M]

I'm no expert on at Go or guessing rank, but some moves stand out to me: B23, B39, B49, B55, B57, B79, B145, B147, B149, B151. I'd guess (purely a guess) OGS 18-14kyu. I have a friend on OGS that is 12kyu and that player has better technique. You can get a ballpark rating by playing against the computer bots on OGS. I recommend it. https://online-go.com/play

[Knotwilg]

No, this is not ddk level. Some choices are unconventional but the flow is very smooth. The timing of playing elsewhere shows high level. Possibly high amateur or historical or AI self play. I would not be comfortable playing these players and I'm 2d.

This question doesn't come from a beginner who wants to know his rank. I'm sorry but it comes across as a trick question.
My best guess is you took KataGo and picked not the blue move but a lower ranked candidate each time.

[CDavis7M]

No, this is not ddk level. Some choices are unconventional but the flow is very smooth. The timing of playing elsewhere shows high level. Possibly high amateur or historical or AI self play. I would not be comfortable playing these players and I'm 2d.

This question doesn't come from a beginner who wants to know his rank. I'm sorry but it comes across as a trick question.
My best guess is you took KataGo and picked not the blue move but a lower ranked candidate each time.

Taking a look with AI, I think you are right. Some of the moves look weird, like if a DDK said "ill just play here," but the move is actually a lower-tier AI suggestions. And now going through the entire game, the end of game suggests you are right as well.

So I'll stand by my original statement. Looks like 14kyu game even it's actually a 14kyu picking wonky AI moves.

[EdLee]

The plot thickens.

[andy12514]

Here is the answer: This is a game on IGS (pandanet). I (black) am 3.5d, and my opponent (white) is 4.5d. My other accounts: fox:5-6d, OGS:1d (had not played on OGS for a while, maybe I progressed a bit.).

Sorry if this seems a troll post, but I do not post this to troll you guys. Rather, when I am reviewing my game, I have a kind of weird feeling that why are my moves so much "like" a DDK. So, I want to see how people view my games.

@dfan, @CDavis7M, thanks for verifying my "weird feeling". Guessing a player's rank is not easy. I have similar experiences that I underestimate people's rank.

@Knotwilg, thanks for complimenting my moves as "high amateur, historical, AI self play", even guessing me playing AI suboptimal moves to trick, I am stones from that. Though in this game, I think I played pretty good, only making minor mistakes. That said, some weird or suboptimal or unconventional moves are actually learnt from AI, while others are just choices I made during the game.

[Knotwilg]

Hi again

I won't dwell on it but I just couldn't fathom how a player who's clearly better than the majority of this forum would be unaware of his own rank. A 4-5 dan would not do that unless if he wanted to troll the forum and 4-5 dans are beyond that, so I could only think of a newbie feeding a historical or AI self play game to us. Two things made me doubt that: the sequence 85-87 where White ignored the leaning maneuver and Black didn't take advantage of it, and the decisive capture in the top right, where White overplayed at 96. Those were signs of amateur play.

Your question "why do my own moves feel DDK like" I can't really answer because to me they don't. What I can say, as a weaker player than you are, is that your moves look like you have a firm grip on fundamentals but are not influenced by any fashion of the day. Also, your moves are solid, rather than sharp, although you did punish that overplay at 96.

Already 7 is unconventional. All of us would stretch. But then the triple crawl before playing elsewhere is sound play.

Sacrificing 2 stones on the lower left and extending at 37, while superficially close to strength, is again pretty solid.

The corner extension at 39 is unconventional but it limits White's expansion already, so it's not bad at all.

45 indicates you haven't studied modern joseki but it's only very slightly suboptimal and up to 49 you can't say Black is worse off than White

The extension at 51 is more of the same: superficially close to strength, fundamentally ok.

The invasion at 57 is natural. The continued leaning at 69 feels slow.

The running fight into the center is fundamentally ok.

...

And then we come to the points mentioned earlier.

I'm curious about your path. How did you get to this point, where your moves are fundamentally ok, solid to the point of slow and void of any fashion of the day?

[kvasir]

It is very much an exaggeration that it looks like a DDK game but it is certainly a bit weird. Neither player seems to value sente, making territory efficiently or the chance to attack. So it is missing some elements that you see in strong players games. It is just one game and we have all played weird games.

Then if it is Pandanet white would be giving -5.5 komi but white play is very conservative, maybe white wrongly assumed that a few odd moves in the opening meant bad moves and he was just waiting for more of that. The overplay around move 90 could just be a way to test black in what white thinks is a hopeless reverse komi game. That brings me to the other strong player trait that this game was completely missing: an intense middle game.

What was really supposed to be the answer to "what is my rank?"? That you are π / 2 dan in French Polynesia or something like that, I mean if you play a lot online then you kind of know your rank in this situation. It is a bit trollish to not give the context that you play a lot and what ranks you are interested in.

[andy12514]

Replying Knotwilg:

I started to seriously learn go after Alphago came out, and got to around IGS 4k. Then, I got stuck, but after leela zero came out, I started to improve again by learning with AI. About your thoughts for my moves:

Move 7: I know the conventional move is to crawl, but that is too boring. I want to play new variations. I recall first seeing that in a pro game, think it is so cool, and learnt it. In fact, in the Samsung Cup this Wednesday (Park Jungwan vs. Chen Yu Chao), this move was played by Park Jungwan.

About the left down corner: Move 23 is a move I learnt from AI again, and I had played out some of the variations, but do not know which is optimal. AI suggests that at move 31, I should just play at the 2-2 point. However, after white's move 32, I know that saving the 2 stones is not worth it (2-2 itself is much bigger than the 2 stones), and want to effectively throw them away.

Move 39: Actually, in this case, the two space extension is not optimal(by AI), but there are cases that a 2-space jump is (similar to this case, where white has an influence down there).

Move 45: I know the conventional move is to play 3-5 point. But I don't want to play that, since it is boring. My move is actually a modern joseki point (also played by pros), in fact the blue point by Katago in this situation. Actually, white's move 46 is problematic. Locally, white should play the 4-6 point instead of 2-6 point. After move 47, my 4-4 stone group got 1 more liberty compared to the conventional joseki.

Move 51: Yep, AI still thinks the left upper corner is bigger. But I still think chances are I can attack the white stones down there, so I value it bigger than the top left corner.

Move 85-87: I am not sure what you mean, but I don't think black has the time to do that. (Maybe I am being conservative here.) Move 87 is still worth a move, directionally okay, but AI suggests that black plays 1 or 2 space wider.

Move 96: Yes, it is an overplay by white. The first game-deciding mistake in this game. It took me some time to calculate that move on move 91.

Replying kvasir:

I think I should make it a guess rank post. I know the middle game is not intense, but I just follow my logic to play.

[dust]

Rather, when I am reviewing my game, I have a kind of weird feeling that why are my moves so much "like" a DDK. So, I want to see how people view my games.

I think you've more or less answered why in your commentary. You game looks unusual because you tend to reject moves that are 'boring' and like to try unconventional or more obscure moves that are likely to surprise your opponent.

That doesn't quite accord with your description that you "just follow my logic to play".

[dfan]

My guess of 15k wasn't based on any reading; I just skimmed through the game in under a minute and saw that Black seemed to play lots of moves that players "learn not to play" by the time they're 10k. Also I admit that I was influenced by the "can anyone tell my approximate rank" question, which is generally asked by people who do not know their approximate rank and are therefore relative beginners, so the weird moves reinforced my prior. Mostly what I learned from this discussion is that my usual instinct to not answer these sorts of questions is the correct one.

[CDavis7M]

Also I admit that I was influenced by the "can anyone tell my approximate rank" question, which is generally asked by people who do not know their approximate rank and are therefore relative beginners, so the weird moves reinforced my prior.

I too thought this was just a Reddit shitpost. And turns out it was.

There's a better place for this kind of thing: https://www.reddit.com/r/badukshitposting/