It became a more serious problem when the new rank adjustment took effect (i.e., where provisional players have a 6k pseudo-rank, as I call it). This is because they get auto-matched against real 6k players, or accept open challenges by strong SDKs, who are much more likely than weaker players, I think, to ditch a game rather than waste time. I don’t recall this when the “humble rank” of 12k was being used.
I saw several cases where a beginner got a real SDK rank after “winning” multiple games that players had ditched (not due to sandbagging, as proven by the strong player’s record). Some strong SDKs didn’t want to play a “?” player, and so they would cancel (which is fine) or they would resign after a few moves if they belatedly noticed the player’s question mark. This is in addition to occasional cases where strong players resign or escape due to opponents not resigning or stalling. In the airbag cases with an established rank, we explained to the player that they should play weaker opponents, or create custom challenges with much lower rank ranges below them. The scale of the problem isn’t huge, but the airbag cases are difficult to deal with.
What happens here is when you add many small errors, the deviation for the sum gets smaller and smaller (because variances = squared deviations are summed, not deviations themselves), so the sum becomes less and less affected by luck and will reflect skill more and more. It is well known that when behind you seek complications (ie. try to enlarge variance and the luck factor, aiming for fewer but larger errors). There is nothing wrong with that oc - just be aware that the extent of the effect of luck is different.
Do you have a reference to support the idea that (a variant of) chess is less sensitive to late game blunders than go? (with comparable player level and time settings)
every player has the right to finish the game for any reason. if you decide to resign, i suggest that you count manually. you can’t trust the estimator to be accurate. partly, because the estimator uses chinese rules to estimate games with japanese rules.
i personally like to resign as soon as i am sure that i have no chance to win honorably, but i am always willing to play a game to the end without feeling abused.
No sorry, and I’m not even completely sure if it is true. Seems hard to actually verify.
Edit: what seems obvious though, is that Go has “temperature”, a gradual reduction effect for later moves (thus later mistakes). So a late big swing is less natural here. Also it can be enough if the other game is shorter - just to avoid the combo of a very long game AND a “who laughs at the end” character.
(to discussion above)
Saying that there is no randomness in Go is like saying there is no randomness in Dice
you maybe may learn to throw it so it never drops with 1, it is just object with physics that can be theoretically predicted. If you are robot, you can predict it perfectly. If you are human, you can’t, something always will be out of control. That’s real nature of RANDOMNESS.
Go is complex. Perfect information doesn’t mean you see everything
There is randomness in Go and rank is level of luck.
The latter does not follow from the former. Skill or strength does have an effect on your games. It IS possible to win by playing better (the 50 * 1 case with low deviation) - it’s just that it is also possible to win by luck (the 1 * 50 case).
Can’t we change the word luckyness by “right intuition”? Then we can debate if this is a skill that you can improve by practice.
I don’t see any luck in go (beside the cat jumping on the board) i only see differences in skills.
An old native American warrior proverb. It is foolish to keep fighting when one is loseing when one can fight again so thats how i decide when to resign if i notice im 20 points behind my opponent has the the majority of the board resign its a good sighn that your reading skills are improving if you can see defeat just dont resign to early or after a few moves it is a honorable choice to resign other then to fight a useless fight just dont do it because your opponent makes a few good moves unless they are devastating move next move they make takes 30 of your stones
Is there any luck in Go?
Sure: if your opponent makes a mistake, you’ve been lucky!
If you play slow moves, if you let your opponent save a dead group, if you miss that atari when filling dame, your opponent is very lucky.
Luck isn’t in game rules. It’s just in player’s choices. And it’s never for yourself: you can’t increase your winrate, you can only keep the winrate bits your opponent is giving you with his moves.
So, yes, someone could say: “my opponent misreaded a sequence and lost a group worth 50 points… I was very lucky!”
But luck has no place there: it’s just player’s mistake.
When that happens, the losing player can be upset or sad, rightfully, but he did it all by himself.
This is true expecially for players who are ahead in endgame: they are ahead, they only have to take care of not making blunders. If they do… so sad, but it’s their fault. They can’t get mad at their opponent for their own mistakes.
I’m not a strong chess player, but I think you can also have this “who laughs at the end” situation in chess.
For example, someone played very strongly in the opening and middle game and has a winning material advantage going into the endgame. But then he blunders the sure win to a draw by stalemate, perpetual check or some forced repitition, or even a loss by accidentally hanging a queen or something. I bet this is just as frustrating as a “silly” game losing blunder in a go endgame where you had a big lead.
Misclick, Internet dying, a phone call distracting you, being tired from a long day, stress from work or family, etc.
Anything that is not in the game but affects your mental state can cause you to not play at your best. Any of the above could easily cause you to miss something important, especially if the game drags on and your stress level increases from outside factors.
Since those things are not a factor in the board state at all, I would say I got lucky if I won a game because of those things happening to my opponent, even if they played well.
I read in the Go Memes thread one quote, something like this:; “Go is not about winning with briljant moves, its about losing by making mistakes”.
Well it very often feels to me like that; however wins is the one that makes less mistakes. Often when I win I think: well, when the opponent makes a mistake like that, of course I win! But then when I lose I just see a couple of moves I should have done differently.
No use calling that good / bad luck as far as I see. Just good moves and bad moves.
A. Mistakes due to lack of knowledge
B. Mistakes due to insufficient reading/analysis of the board
C. Stupid mistakes due to lack of attention, like missing an atari.
A, B and C are part of the game. Losing a game because of C is unpleasant but you can’t blame it on your opponent.
That said, when I play a game, I keep wondering how come my opponent makes so many bad moves and I still manage to lose…
“I think I could still learn from the game, or even learn about minimizing my losses after a big mistake.”
In my view the best way to think about this is opportunity cost. Time spent playing out a lost game is time not spent studying why that game was lost. It’s also time spent not playing a new game.
You should focus on the part of the game that is weakest. If you’re losing in the mid-game all the time, then practicing the endgame is not the best use of your time. Nor is it a good use of your opponent’s time.
