If any individual game is holding up a tournament, the mods can simply end that game after it is reported. If nobody is paying enough attention to realise it’s the last game of a round, they probably aren’t too upset by the delay either.
2 Likes
I think this is the root of the problem: holiday time has a very high cap and accrues very quickly, and can easily double the expected length of a fast correspondence game. I would reduce the max to something like 2wk non-supporter 4wk supporter, and recharge at the rate of say 1 day every 2 weeks. That means a site supporter can fill up their bank in a little over a year
If Vacation time wasn’t so overpowered, I would be a lot happier with auto-vacation. I think I remember chess.com having auto-vacation
2 Likes
This I think is fine, though. If they’re dead or so severely ill that they can’t play moves online, they’ve got bigger problems than their opponents having to wait longer for them to time out. I would still want less OP vacation time, though
That premise, which I agree with and was a longstanding principle regarding serial timeouts, has just been denied by @teapoweredrobot. It appears that the new standard is anything counts, which is one reason why there is an explosion of escaping from correspondence games. In that light, this thread seems pointless. I thought progress had been made here, but that apparent progress seems to have been turned upside down now.
I think the reason for this is mainly practical. There is not much to differentiate someone who got hit by a bus from someone who simply had a change of heart.
At any rate, I don’t think there is much harm in giving compassion to a broad array of life events (baby, work, depression etc.)
This is speculation. Quantify the causal relationship and it will be a worthy discussion point.
I think the best way to proceed would be to focus on finding and closing loopholes in the current system, given @GreenAsJade’s original question, and @teapoweredrobot’s summary of the current crossroads:
1 Like
When anything goes, the whole subject becomes speculation. People can cry “busy” and get a free pass to escape as many games as they want. They can have five long strings of serial timeouts in four months, and it doesn’t matter. “Busy, don’t you know.”
1 Like
I think I’m ok with that. They’re just games at the end of the day. It’s not like they’re not paying rent or taxes or something.
I don’t think we should need to investigate or have people justify why they stopped playing some games.
That said if they timeout of games regularly with the intention of exploiting some loophole, like the one with mass annulling games that timeout, then it seems fair to either ask them not to do it, or just fix the code in some way - whichever’s easier 
4 Likes
First, how would you know? Timing out of a correspondence game is no longer a reportable violation.
Second, “intention of exploiting” is speculation, which is now very ill-founded since the all-purpose rejoinder is “I was busy.”
At the very least, the suggestion of a “trigger” or rather a stop-gap mechanism based on someone logging in should be implemented to eliminate the bug whereby a live game fails to break the string. The great thing about that idea is that it encompasses all activity. We have already seen the example above where someone created new games in the midst of a timeout string. And if someone is not too busy to be in a chat channel or the Forums, then they are not too busy to play their games.
You are correct that it would be difficult or impossible to verify one’s intention. However, it is possible to differentiate a serial-timeout from a serial-serial-timeout (e.g. “five long strings of serial timeouts in four months”). I think it’s reasonable for a moderator to use discretion to treat the two cases differently - it is not an all-or-nothing thing.
Completely agree, and I think this is the type of suggestion we should focus on - closing glaring loopholes in the existing rule.
3 Likes
We never know anything for certain. People can always feign innocence or stupidity for any kind of cheating or other misbehaviour. At the end of the day if someone doesn’t admit it you have to weigh it on the balance of probabilities. If someone hasn’t lost a correspondence game in many many months because all the ones they would’ve lost timed out, but they have many wins over the same period, then something seems to be up.
You can make up any excuse you want, but it’s hard to believe something like “I was too busy only when I was losing”.
1 Like
Ah, someone finally understands my point, perhaps. (You are not the only one who can condescend.) When the probabilities are based on a rare incapacitating event, then judging them has a strong foundation, but when they are based on an all-purpose ambiguity, they are ill-founded.
Not intending to condescend, just replying to what I was reading.
I don’t think the mass timeout rule is really there for a rare incapacitating event, I think it’s just an easy example where it seems harsh to blame someone.
If you look at the one of the original posts (I assume) of the introduction of such a rule, there’s no mention of why they timed out in order to qualify, they just did. It was more for the players who play a lot, than for a lot of players also.
Anyway the point is people stop playing for any number of reasons as mentioned
and I personally think that in the above 1 and 2 are fine, but when there’s evidence of 3 AND all the games are being annulled then there’s likely an exploit happening.
I mean if it’s particularly annoying people that a certain person times out of losing games when they’re not being annulled we could say something to them, or somebody that dislikes them can block them etc. You could introduce some small punishment like the flag on their profile that used to exist, only that too was removed, I think maybe it was buggy.
2 Likes
Anoek’s 2014 post seems to address a problem in a very specific portion of the population, where there was little reason to suspect any cheating or habitual rude behavior (such as timing out of correspondence games).
Perhaps circumstances subsequently changed and a wider variety of players were triggering the rule, leading to a need to tighten the range of acceptable excuses. I don’t know. What I do know is that when I became a mod, the rule was explained, discussed, and acted upon in terms of rare events.
@jlt’s list of causes fails to mention cases of what I call unconscious abuse of the rule, whereby players routinely timeout the games they are losing. Ironically, I think this is by far the most common cause of serial timeouts. Timing out is now a widespread phenomenon in the correspondence culture. It is accepted because it is no longer a reportable violation. No, I cannot prove this statistically; indeed, it is probably impossible to prove due to a lack of easily accessible information, or, if the information does exist, it would require know-how and a wildly obsessive personality to get at it.
However, as I already noted, I have taken a tour of correspondence games and random players, which convinced me that routine timeouts are incredibly common. If one times out most lost games, it will create some serial timeouts as a matter of course. Most such timeout strings will be short.
This is why I asked “how would you know?” Timeout reports were the main source of leads that would bring about discovery of potential abuse of the rule. Has anyone, other than me, reported abuse of the rule in the last several years? Has anyone (not asking for names, you understand) even been warned for abusing the rule? Back in the day @Eugene and I handled most, if not all, of the reports. Has any current mod, aside from Eugene, ever handled a single report of abuse of the rule?
Yes, the flag was abolished because it was buggy—the automated removal process for the flag wasn’t working right. Restoration of an unbuggy flag would be one of the best things that could be done to ameliorate abuse of the rule.
1 Like
If we focuse a bit on the cheating pattern, it’s all easy. First mass timeout nothing to change (almost). Now if the fact is coming again (repeat detection) then we can consider the mass timeout differently. I would first involve any mod to clarify the situation with the players but this could go first with an automated warning message.
1 Like
I looked at my game history since January 2022:
- I played about 300 ranked correspondence games.
- Overall winrate 70%.
- I never timed out.
- 13 games (4%) ended by timeout. I’m not counting games in which my opponent never showed up.
- Among these 13 games, my opponent timed out while being ahead in 1 game, while being behind in 6 games and while the game was about even in 6 games.
- Two of these 13 games were part of a serial timeout. The games were about even.
Conclusion: serial timeouts are quite rare, and almost never used to cheat, at least against me.
2 Likes
Cheating isn’t the main issue, and my impression is that rude behavior, such as timing out, is much less common at your level.
Just so I can be clear. Are you saying that you main concern is not that there is cheating? So is it that timing out is disrespectful or some other kind of non-cheating thing that is the problem?
Full disclosure: I was driven to “do somefink” here because I had two “likely wins” on a site ladder annuled in fairly quick succession from mass timeout.
One looked a lot like a “cheat” to me, although I was peeved, so I may have been seeing what I “expected”.
Given how little I actually play (these were “all the games I had on the go at the time”) I can’t agree with that conclusion.
I think Conrad is right that there are likely more of these in the DDK area than SDK.
Conrad is certainly right that cheating isn’t the issue: I personally didn’t care whether it was a cheat or a “genuine RL event” …
… I just felt like not playing correspondence at OGS any more after two strikes in a row. (*)
So MY main concern is to stop the experience of “hey, this feature robbed me of a win”.
- : I suspect this is something that effects players who play less more than the lots of games people. For people who play less, each game is a little pearl that you polish over the months. For people who play more, maybe a mass timeout annul is just “meh, another grain of sand went that way”.
5 Likes
I’ve already stated it a couple times: the main issue is unconscious abuse of the rule, because lots of people escape from correspondence games simply as a habit. The consequence is that many games are unjustly annulled, with winners victimized, and probable winners victimized, and maybe winners victimized, and 50-50 players victimized (hey, I just spent X months on this game, and it gets thrown away?!). That’s not cheating, and it’s not mere rudeness; it’s rudeness with a horrible consequence.
Laying it on a bit thick there buddy, don’t you think?
1 Like