4 dan or 6 dan? You decide!

[Uberdude]

How easy/hard is it to tell the difference between a 4 dan and a 6 dan player from their moves? Here are 5 games of European 4 or 5 or 6 dan players, can you work out their ranks? I will give 2 points per correct answer, 1 point if 1 rank wrong and 0 points if 2 ranks wrong. So a perfect score would be 20/20. Random guessing would give 10/20. What threshold should we take as demonstrating the truth of "It easy to tell the difference between a 4 dan and 6 dan"? (for sufficiently strong players, weaker players might like to play this game for fun/interest but them being bad at it doesn't show 6 or 7ds couldn't be good at it, I hope some strong players participate). I propose >=18 of 20, but am open to slightly changing that threshold before I mark any answers (it is important to choose your test and define the success criteria before doing the experiment). If you recognise any of the games please be honest and don't vote on it and we can give you the average score from the other answers for that one. Please use the hide tags for your answers; above the comment box there is a row of buttons, click "Hide" to insert the hide tags. Or make it yourself like this with "Hide" in square brackets and the slash in front of closing tag:

Code: Select all

[Hide]
Player 1: Xd
Player 2: Xd
Player 3: Xd
...
[/Hide]

EDIT: Apologies, I made a mistake with the first game (one of the players was outside the 4-6 dan range, can you tell?). So I'll hide that and add a replacement with Players 1a and 2a. Also I added 2 more bonus games to the end. So ignoring wrong game 1 if you do the 2 extra games as well then perfect score is 24/24 [Edit 2: bad arithmetic, 28/28] and let's say 21/24 [Edit 2: 24/28] is pass mark.

Original Game 1, ignore

Game 1a


Game 2


Game 3


Game 4


Game 5


Bonus games
Game 6


Game 7

[bugsti]

Game 1:

white 5d
black 5d

Game 2:

white 6d
black 6d

Game 3:

white 4d
black 4d

Game 4:

white 5d
black 4d

Game 5

white 4d
black 5d

[bernds]

Well, I'm not trying to guess exact ranks. If nothing else, because I don't know the time controls and high-dan blitz games can look really poor on occasion. But, I thought the following players looked quite strong:

2, 4

While I wasn't as impressed by:

1a, 2a. 5, 6, 10, 13, 14

I guess the others were kind of in the middle?

[Uberdude]

Well, I'm not trying to guess exact ranks. If nothing else, because I don't know the time controls and high-dan blitz games can look really poor on occasion.

These are games from old seasons of the Pandanet Go European Team Championship, so 1 hour main time with 25 stones in 10 minutes overtime (some games have move timing information). Online, but in a fairly serious tournament so not mindless blitz.

[Bill Spight]

In his commentaries, renowned bridge champion and editor of the The Bridge World, Edgar Kaplan, used to total up the number of points chucked by each player. If we did that using Leela Zero Elf or other top bot, would those totals, or the average loss per play, correlate with rank?

[Gobang]

These are games from old seasons of the Pandanet Go European Team Championship.

Really? So how can you be sure that the game was in fact played by the player who says he or she is the player and not some other player or a computer?

[Uberdude]

Really? So how can you be sure that the game was in fact played by the player who says he or she is the player and not some other player or a computer?

I cannot, just as when you play me at a real life tournament you cannot be sure you aren't actually playing Tom Cruise with a face mask.

Except I'm taller

or a computer?

That would not have helped as computers were weaker than 4d at the time.

[Gobang]

So all this is is a waste of time, just like 99% of the babble around the topic of detecting online cheats.

[Uberdude]

So all this is is a waste of time, just like 99% of the babble around the topic of detecting online cheats.

I disagree, though I think your constant repetition of negativity is a waste of time.

[Gobang]

So all this is is a waste of time, just like 99% of the babble around the topic of detecting online cheats.

I disagree, though I think your constant repetition of negativity is a waste of time.

And I think your negative comment is a waste of space. I am just trying to inject some sanity into a topic that seems to lack it.

[HermanHiddema]

So all this is is a waste of time, just like 99% of the babble around the topic of detecting online cheats.

If babble around the topic of detecting online cheats is a waste of time, then what does that make your babble about the babble?

[gamesorry]

Player 3: 4d
Player 4: 6d
Player 5: 6d
Player 6: 6d
Player 7: 4d
Player 8: 6d
Player 9: 5d
Player 10: 6d
Player 11: 6d
Player 12: 5d
Player 13: 6d
Player 14: 4d

Edit:
Player 1a: 4d
Player 2a: 6d

[Schachus]

I'm way too weak to distinguish playing strength at that level (and also I know from chess I know that even for players at your level or slightly stronger/weaker its very very hard to guess level of play from 1 game).
Hence I optimize my expected score by guessing 5d on all players. I bet it is gonna give me a better score than most others.

[Bill Spight]

Vanitas, vanitas, omnes vanitas!

[Uberdude]

I won't give the answers yet in case more people want to try figuring out the ranks, but will score those given. So if you like solving simultaneous equations maybe you can deduce something.

Schachus: 16/28

gamesorry: 14/28

bernds (taking "looked strong" as 6d, "not impressed" as 4d, and "middle" as 5d): 16/32

bugsti: 10/20

Also my 1d wife played the guess the ranks games, but I made it a bit easier/fun by telling her the answer after each game, which allowed her to calibrate her answers as she doesn't have much idea what a 5d or 6d European's play looks like, just me as a 4d. Also she correctly guessed the age group of one of the players .

Her answers:

1: 3d
2: 4d (or maybe 5)
3: 6d
4: 5d
5: 4d
6: 4d
7: 4d but weaker than previous pair
8: 4d ditto
9: 5d
10: 4d but used to be 5d (old person, can't fight with new generation)
11: 6d
12: 4d
13: 3d, turned into 1d after reviewing with LeelaZero which I did suggest would make everyone look weak
14: 3d, ditto

Her score: 18/28