Why not?
meanwhile, the opponents gets to play… I mean, if you need to pass let’s say 10 times, to get extra 30s… your opponent got 10 handicap stones in her favor… which is worth more than 30s of thinking time.
3 Likes
Hmm ok yeah that’s a reasonable argument…
So, it seems that if we introduced:
3s passing time credit; and
Minimum absolute time (8mins?)
Then there would only be “playing nonsense” left unresolved.
Progress? Can it be done?
There was another thread about adjusting the warnings for unusually short games I think. Should there be minima introduced in other time settings? (1s simple or 0+1s Fischer or 1s byo yomi anyone?)
2 Likes
3s is too short for check if you answer ot not.
I feel unfair to timeout because of pressing the clock, it’s different as putting stones.
How? If you are immune to timout when passing. If you have to play/snswer it’s more relevant to time management choice (like solidity vs thiness for ex).
1 Like
3s plus any time you had left on your clock though
The “passing time credit” idea is unjustified IMHO. First, people don’t need to restart repeatedly to gain significant thinking time. They can just take 30 sec or a minute or whatever the first time they pass. Second, the idea that there is a significant problem with people passing to gain thinking time is dubious. In most cases, repeated restarting occurs after the game is irretrievably lost. The purpose is to frustrate the winner into abandoning the game. Indeed, in almost all of those cases, the restart-pass-restart is done immediately—no thinking time is gained. This sends the message to the winner, “Hey, I’m going to do this forever.” Anther category of restart cases is where a beginner, or even an experienced player who doesn’t know how to count, uses the scoring phase to get an accurate score. If they are losing by just a point, for example, they restart in the hope that the winer will make a blunder; this is extremely common.
1 Like
I thought the problem being addressed is that the winner passes with a few seconds left on the clock but the opponent plays some nonsense so it is the passing players turn again. They have to pass immediately again (but it might take a second to do so, register, internet lag, whatever. Another nonsense move and after a few cycles the winner had timed out.
3 Likes
Yeah the problem is not the repeated passes but the repeated breaking the pass. But if the offender see the clock of the passer reset at 10 while his own clock continues to go down, he may become quickly discouraged.
This just seems to be a normal attitude to check if something is solid or not.
Like score cheating, time cheating (via stalling in absolute games or restarting in super-blitz games) is much more common than the number of reports show (at least when I was moderating). The worst case I handled was a guy using the restart trick in games of 4 sec per move IIRC. He had won dozens of games doing this without ever being reported. I warned him, and he disappeared, presumably after creating a new account with a VPN. Similarly, when warning someone for stalling in absolute games, I usually found that they had been doing it for some time without getting reported.
3 Likes
That is just garden-variety stalling. They should be warned, and then banned if they do it habitually.
1 Like
I guess but what if this 3s pass period means they cannot do it successfully at all. That’s better isn’t it?
1 Like
No, I don’t think that is better. Putting the onus on someone to pass just because their opponent has passed is fundamentally unfair. It wrongly assumes that a person playing after a pass is doing something wrong. A pass, like any other move, should be a free choice. A cheat can leverage that rule. A player short on time, ahead in score, but with a complicated and flawed position could pass as a tactic. They win if the opponent passes and they gain thinking time for no good reason if the opponent decides to play the position out.
Also, other legitimate reasons for the opponent not to pass would be to continue a winning ko or to close boundaries.
Generally speaking, I think it is a bad idea to create work-arounds as a substitute for moderating. Stalling (self-atari moves and other useless moves, infilling, and repeated restarts) is often a precursor to score cheating or a sign of someone who may in the future score cheat. (Of course, I am not speaking of the beginners who stall because they don’t know better.) Warning and ultimately banning people for stalling also removes some present or future score cheats.
2 Likes
I favor both of these things. 8 min was the minimum ranked time when I played at KGS, and that seemed fair. Frankly, I only play with “live” setting where the minimum is 10 min, and didn’t know that you could play faster ranked games under blitz. The max time there is 5 min, which is too short for most purposes, and shorter - which gets crazy for 19x19. 8 minutes isn’t even currently an option for absolute time games. If changes are made - I think it would be sensible to fill that gap of 5-10 minutes where you can’t set it. Maybe make blitz be up to 7 minutes and unranked, live be 8 min + and ranked?
Its somewhat situational when absolute time gets too fast to be “unfair”
Current minimum setting for ranked absolute is now 30 seconds, maybe some fast player could manage that 9x9 board, but for 19x19 it just wont be enough to reach scoring. (prove me wrong ;))
For 19x19 having 10 minutes for each player should be enough to reach the scoring even if the game lasts 300 moves. At least for those who understand that timesettings… I’m not sure what the sweet spot is, but i have a gut-feeling that its closer to 10 minutes than 30 seconds.
Makes me wonder about another question:
Should the ‘minimum time allowed’ for ranked games be increased based on the board size?
2 Likes
Certainly, that would be the most sensible. And I agree that there are time settings that are just impossible. I’m sure I’ve finished many a 19x19 game with 5 minutes left on the clock, most for 2+ minutes. But I still wouldn’t say you should have ranked games at 5 min.
I don’t suppose you would be able to fill out this table would you?:
| Games With Time Abuse | Games Without Time Abuse | |
|---|---|---|
| Blitz | a | b |
| Non-Blitz | c | d |
I think you know what I wanna do lol. (Er maybe it wasn’t you that does stats? I can’t say “you know what I wanna do” :S )
1 Like
Sadly, no, we don’t log this kind of data…
I think I know the thing of which you’re thinking that I know that thing, but I don’t know if you know stats well enough to pick the correct test, and I know that I don’t know which test to pick, therefore I can’t say I know the thing of which you’re thinking that I know it.
( Most of my stats knowledge stems from teaching it to pharmacy students, where I was practically one week ahead of the students at all times 
Stats is my nightmare, I’m glad I only have to work with measure theory in abstract settings )
(( incidentally and relevantly I also taught epistemology, with which I feel substantially more comfortable than with stats ))
4 Likes
The data collection method I proposed does suggest a single sample with two categorical variables, so it’s a chi-square test for independence.
It would be nice if OGS started tracking things because a lot of claims are made that impact OGS. Bunburyist could be totally right that blitz is perfectly fine as is, or you could be right. It would nice to know !
1 Like
Yay, I remember statistics correctly, then.
It’s hard to track these things for several reasons: there’s no moderator tools available with which we can categorise reports, most games featuring cheating aren’t reported, and for the things that are tracked we have no way to search or filter them (e.g. reported users are tracked, and for any individual user we can see previous reports, but we can’t produce a list of users that have been reported for a certain thing, or even a list of all users that have been reported)
1 Like
Stats is overrated.
Every single event has a 50% chance of happening because it either happens or it doesn’t!
1 Like