Feature Request: Don't change rating points when terminating abandoned match

OP request seems logical to me. If player A has a higher rating than player B and player B disconnects in an even position, player A has two options:
(The exact numbers depend on the rating difference, I choose arbitrary values to show the point)
* Claim victory --> +5
* Call a draw --> -1

So if +5 or -1 are both valid options, then I don't see any reason why +/-0 should be a problem for the rating system because zero is in between. However, I don't have a strong opinion on whether it's worth implementing this option, since -1 and 0 are almost the same anyway:)

@Clarkey said in #3:
> Yes, but this is online chess - I could also claim the win. I'm being punished for not taking cheap points.

The thing is lichess follows fide rules
and those rules are for chess regardless of platform
and as lichess has said in the pass the dont pick and choose which fide rules they follow

Too many words...
OP has a point.
There should be an abort option.-

I also think an extra option "abort game", in addition to the existing options claim a win, call a draw, resign, or wait, would be a good addition for when the opponent has left the game. Of course the player who remains in the game must have an entirely free choice from those possibilities.

I agree that @Clarkey has a point.

@Cedur216 said in #10:
> How do you define "very far"? And if I blundered very early or am not satisfied with the opening, can I disconnect intentionally?

I don't think that was the point. In the current scenario, players can also not claim the win. If someone wants to abuse the system by intentionally disconnecting, it is still possible.

@Cedur216 said in #10:
> Also nobody's a dick for claiming win.

Of course not. But the OP never claimed that. From what I could tell, it is simply their personal preference and that's fine.

@Cedur216 said in #10:
> you do, except you're wrong in a practical way.

No, they are not, not entirely at least :) But I see your point here @Cedur216. The addition of this feature to abort games in case of disconnection will likely increase the possibility of rage quitting. The current system can also be abused, but it doesn't encourage users to rage quit.

However, this issue can be solved automatically by Lichess intervening if the user intentionally rage quits as opposed to having connection troubles. This shouldn't be too hard, the mechanism should be similar to detecting whether the user decides to let the time run out intentionally.

Whether it will be worth the investment considering the pros, cons and effort is another discussion though :)

I don't think that the OP#S suggestion would increase rage quitting, @sgtlaugh. It is just for those cases when we see our opponent disappeared without having a reason in the game situation, when the odds are completely open or even at their favour.
I'd like to be able to say: ok, pity we can't sontinue this game, but be it so. Let's cancel it without consequences.

Here's a general problem: How do you let codes and programs make that human judgement of "without having a reason"? Besides, getting a disfavored opening (e.g., the London system for some people) might also become a reason for systemic disconnecting.

@Sybotes said in #16:
> I don't think that the OP#S suggestion would increase rage quitting, @sgtlaugh. It is just for those cases when we see our opponent disappeared without having a reason in the game situation, when the odds are completely open or even at their favour.
> I'd like to be able to say: ok, pity we can't sontinue this game, but be it so. Let's cancel it without consequences.

Of course, it would be for specific cases. But if this is implemented, then some people will try to abuse it by quitting/disconnecting intentionally in lost or unfavorable positions. Although, I believe this will be a minority and should not pose a problem in general.

@Cedur216 said in #17:
> Here's a general problem: How do you let codes and programs make that human judgement of "without having a reason"? Besides, getting a disfavored opening (e.g., the London system for some people) might also become a reason for systemic disconnecting.

It shouldn't be too hard. The key should be finding patterns or deviations from usual patterns. A very simple metric could be to consider whether the user who got disconnected was in a favorable or unfavorable position when they got disconnected. If this is highly deviant than their usual win/lose ratio, then there is a good probability that the user is intentionally disconnecting.

In any case, I don't believe it will cause any problems even if we just implement the feature without this automation. Lichess already has the feature to claim a draw in case one of the players gets disconnected. Some users could try to take advantage, but I believe it will be a minority and the overall effect will be negligible.

@Cedur216 said in #17:
> Here's a general problem: How do you let codes and programs make that human judgement of "without having a reason"?
No need to code that. Let it be up to me. When I think "doesn't look like they volunteerly left the game", I just want to say "cancel".