Losing by abandonment unfairly penalizes people with a poor connection

Sort:
cyhollander

Although my connection is usually fine, I've lost a couple of games when it dropped for a short time and I couldn't reconnect in time, although I would have had plenty of time on my clock after reconnecting. This isn't fair to players whose connection drops through no fault of their own. If the connection hadn't dropped, they would be entitled to use that time thinking; they should likewise be entitled to use that time to reconnect. There's no need for an abandonment time limit, when the games already have a time limit built in.

 

I hope someone from chess.com sees this post and takes this feedback under advisement (of course, anyone else can also feel free to comment if they agree or disagree).

Edit: I'm sorry for accidentally posting this thread in the General Chess Discussion forum instead of a website feedback one. If any administrator is able and willing to move it to a more appropriate forum, I would appreciate that.

cyhollander

My connection is good for the most part, while I think few people's connections are so good as to never drop for a minute or two. Furthermore, not everyone in the world has access to or can afford a good internet connection.

Anyhow, I don't see how you're addressing my main point. I agree with you that one's opponent should not have to suffer on account of one's bad connection, but, again, the time limit is already in place to limit one's opponent's suffering. A player is entitled to take a certain, mutually-agreed-upon amount of time over his moves, beyond which, he must forfeit the game. If he is entitled to take this time for thinking, by what rationale should he not be entitled to take the same amount of time for reconnection?

This site's implementation of the abandonment timer is further problematic in that, unlike the game's main time limits, which are flexible, clearly specified, and agreed on by both players, the abandonment timer is 1) not explicitly agreed to by either player, 2) not specified in detail in any place easy to find, let alone on the Live Game page, so that players cannot even be said to have agreed to the abandonment timer implicitly, and, most importantly, 3) not flexible, so that if someone wishes to play a live game at chess.com at all, he must agree to chess.com's abandonment timer willy-nilly, even if he would prefer to play without it—even if both players would prefer to play without it, in fact.

This last point is the most important, in my opinion. If the existence and length of the abandonment timer were an option that could be selected and/or customized when one searches for a live game, in the same way as the main game clocks, this would ideally address all concerns: people like me, who would prefer a longer abandonment timer or none at all could play games with those settings, while people like you, who like the timer, could continue to play games with those settings. In the absence of this ideal solution, however, I continue to feel that the game clocks, being the only time limits that the players choose, know the details of, and agree upon, should be the only time limits imposed on them.

asilverman77

Just had this problem with a vengeance in a kids' tournament.  Dropped connections resulted in six of ten scheduled games decided by "abandonment".  Other mysterious problems of pairing and registration appear to be also related to dropped connections.   Rural area with many using DSL.  

Da-Vere

A well reasoned and sound logical thread cyhollander! The abandonment can bite both ways too. I have a pretty good and reliable internet service but it can and does bug out on me occasionally. I feel bad for my opponent more than myself as a purposeful abandonment on my part would be poor sportsmanship. However, this just happened to me. It was a blitz game and it was my turn. It was to be a checkmate in one move, if i hadn’t lost the connection. I don’t know if it was the loss of internet or app crashing as my iPad was locked up and had to delete/reinstall the app to return to chess.com. My point is that the abandonment rule can slice you both ways. Lose one you should win and win one you should lose. 

hoodoothere

I have a very good fiber optic cable fast connection and this still happens to me but only on this site.....and usually when I am in a winning position....wondering if it is possible for certain other software savvy players to crash your connection on purpose????

llama44

I don't think the point is to punish people with bad connections. Those people are relatively few.

I think it's to punish people who would fake a bad connection to harass and stall the game. Those people greatly outnumber those with a real bad connection.

As for making abandonment time or setting an additional option, that's reasonable, but it also comes at a cost (there is no decision like this with only good points). Additional search elements makes wait times longer because you're fracturing your player pool into players who want X and players who want Y etc.

Maybe a better option would be to have what in the old days on ICC was called "no escape" i.e. the game was an automatic loss if you disconnected. Then set it so that this will be the rule even if only 1 player has that option turned on in their settings (in other words not a seek element so it doesn't split the pool).

That would make finding a game that allowed disconnections difficult... you'd probably have to join a group where people play each other like that, but at least there'd be the option.

llama44
hoodoothere wrote:

I have a very good fiber optic cable fast connection and this still happens to me but only on this site.....and usually when I am in a winning position....wondering if it is possible for certain other software savvy players to crash your connection on purpose????

That's just a psychological trick. When you lose connection in a bad position you won't care, but when you lose connection in a good position you remember it... so searching your memory will give you a false impression. Also the emotional effect of losing a good game like that causes those games to be weighted more in any case.

It's the same sort of psychological trick for people who think their dreams come true, or that there is more crime or busier hospitals on days with a full moon.

Anyway, no, your opponent can't make you lose this way. This has been asked a million times

Recently some people have been pointing out that if your opponent hacked into chess.com they'd have access to, for example, thousands of people's credit card information. They're not going to bother doing that to win chess games.

llama44
ChessAntics2020 wrote:

I have noticed it where it shows my opponent has a weak connection and they keep reconnecting and winning. Not on time, but playing better moves. 

This is also a trick of the mind. Cheating with an engine, whether it's manually or some automated program, wont have an effect on the connection. What's probably happening is their bad connection somehow is causing you anxiety because you're hoping it will be to their disadvantage, maybe they'll even lose right away by never making a move again. You lose focus, play worse, and their moves seem better.

hoodoothere
llama44 wrote:
hoodoothere wrote:

I have a very good fiber optic cable fast connection and this still happens to me but only on this site.....and usually when I am in a winning position....wondering if it is possible for certain other software savvy players to crash your connection on purpose????

That's just a psychological trick. When you lose connection in a bad position you won't care, but when you lose connection in a good position you remember it... so searching your memory will give you a false impression. Also the emotional effect of losing a good game like that causes those games to be weighted more in any case.

It's the same sort of psychological trick for people who think their dreams come true, or that there is more crime or busier hospitals on days with a full moon.

Anyway, no, your opponent can't make you lose this way. This has been asked a million times

Recently some people have been pointing out that if your opponent hacked into chess.com they'd have access to, for example, thousands of people's credit card information. They're not going to bother doing that to win chess games.

Yeah, probably just random chaos bolstered by selective memory, it doesn't happen that often anyways.

llama44

It doesn't matter how poor your connection is, or if you close your browser, your clock never pauses, (at least not like that).

llama44

But it helps to think of it as there are 3 records of the time. Your opponent sees one, you see another, and the chess.com server is keeping track of the real time.

So when you move your clock stops right away, as soon as the server sees your move, your real clock is stopped, and last of all your opponent will see your clock stop.

So if your opponent moves, and it takes 1 second for the information to go to the server and then to you, then their clock will appear to gain 1 second.

llama44

And for example, if a game is 5 | 0 then it should be impossible for it to last 10 minutes (because that would mean both players reached zero).

But try timing your games. In reality, if both players get close to zero, a game like this might last 10 minutes and 5 seconds of real time.

cyhollander
llama44 wrote:

I don't think the point is to punish people with bad connections. Those people are relatively few.

I think it's to punish people who would fake a bad connection to harass and stall the game. Those people greatly outnumber those with a real bad connection.

 

How do you know this is true? All you can see is whether your opponent loses his connection. You have no way to tell whether he disconnected deliberately or he lost his connection. At best, you can take a guess based on the circumstances of the game, but this is far from proof positive (I can tell you, for instance, that several times I have lost my connection in a game that I was losing but wasn't yet ready to resign), and without a way to confirm your guesses, you really can't place much stock in them. I don't think that site policies should be based on speculative and untestable assumptions like your assumption that most who disconnect do so in bad faith.

But even if we granted that assumption, the abandonment timer still would not accomplish much, because an opponent who wishes to harass you by dragging out the game can do that under the existing conditions by simply not making a move but not disconnecting. It's mildly irritating when an opponent does that, but, again, that's what the clocks are for. You wait for his time to run out and collect your win.

TL;DR: Given that an opponent who wants to harass you can do so without disconnecting, I 1) don't see that an abandonment timer prevents people from stalling the game if they want to and 2) therefore don't see a good reason to assume that most people who disconnect are trying to stall the game, as opposed to legitimately losing their connection. For both reasons, I don't think the possibility of bad-faith disconnections is a strong reason to institute an abandonment timer.

asilverman77

"TL;DR: Given that an opponent who wants to harass you can do so without disconnecting, I 1) don't see that an abandonment timer prevents people from stalling the game if they want to and 2) therefore don't see a good reason to assume that most people who disconnect are trying to stall the game, as opposed to legitimately losing their connection. For both reasons, I don't think the possibility of bad-faith disconnections is a strong reason to institute an abandonment timer."

This is the reasoning I'm trying to promote on abandoning the abandonment timer for kids' tournaments in chesskid.   One of these fora has the observation that "the server knows" who is being a bad actor.  If so, the chesskids' coaches should be informed, since there is a human layer of management not available at chess.com, and where the discipline for bad sportsmanship can be best applied, so there is no longer any reason for an abandonment timer.  So far, no traction at chesskid on that idea .... 

llama44
cyhollander wrote:
llama44 wrote:

I don't think the point is to punish people with bad connections. Those people are relatively few.

I think it's to punish people who would fake a bad connection to harass and stall the game. Those people greatly outnumber those with a real bad connection.

 

How do you know this is true? All you can see is whether your opponent loses his connection. You have no way to tell whether he disconnected deliberately or he lost his connection. At best, you can take a guess based on the circumstances of the game, but this is far from proof positive (I can tell you, for instance, that several times I have lost my connection in a game that I was losing but wasn't yet ready to resign), and without a way to confirm your guesses, you really can't place much stock in them. I don't think that site policies should be based on speculative and untestable assumptions like your assumption that most who disconnect do so in bad faith.

But even if we granted that assumption, the abandonment timer still would not accomplish much, because an opponent who wishes to harass you by dragging out the game can do that under the existing conditions by simply not making a move but not disconnecting. It's mildly irritating when an opponent does that, but, again, that's what the clocks are for. You wait for his time to run out and collect your win.

TL;DR: Given that an opponent who wants to harass you can do so without disconnecting, I 1) don't see that an abandonment timer prevents people from stalling the game if they want to and 2) therefore don't see a good reason to assume that most people who disconnect are trying to stall the game, as opposed to legitimately losing their connection. For both reasons, I don't think the possibility of bad-faith disconnections is a strong reason to institute an abandonment timer.

Hmm, yeah, that's a good point.

I guess it's out of some motivation to streamline user's experience. This is obviously speculative, but lets say they did their research and determined ___% of disconnects are permanent. In other words if someone loses connection for 5 seconds they're 90% likely to never come back to the game for whatever reason. So by auto forfeiting the player chess.com improves the overall experience of the site.

One problem with this is their data may be 10 years old at this point  So even with this speculation it makes sense for chess.com to revisit the rules.

cyhollander
llama44 wrote:

 

I guess it's out of some motivation to streamline user's experience. This is obviously speculative, but lets say they did their research and determined ___% of disconnects are permanent. In other words if someone loses connection for 5 seconds they're 90% likely to never come back to the game for whatever reason. So by auto forfeiting the player chess.com improves the overall experience of the site.

One problem with this is their data may be 10 years old at this point  So even with this speculation it makes sense for chess.com to revisit the rules.

That makes sense. I suspect you're more or less right about the reasoning behind the timer. It would be nice if a chess.com representative noticed this thread and were able to confirm or deny your guess or, better yet, give the detailed statistics behind the decision. While I'd still prefer a customizable timer or none at all, it would be reassuring to know at least that there was some solid thought behind the implementation of the current system.

llama44
cyhollander wrote:
llama44 wrote:

 

I guess it's out of some motivation to streamline user's experience. This is obviously speculative, but lets say they did their research and determined ___% of disconnects are permanent. In other words if someone loses connection for 5 seconds they're 90% likely to never come back to the game for whatever reason. So by auto forfeiting the player chess.com improves the overall experience of the site.

One problem with this is their data may be 10 years old at this point  So even with this speculation it makes sense for chess.com to revisit the rules.

That makes sense. I suspect you're more or less right about the reasoning behind the timer. It would be nice if a chess.com representative noticed this thread and were able to confirm or deny your guess or, better yet, give the detailed statistics behind the decision. While I'd still prefer a customizable timer or none at all, it would be reassuring to know at least that there was some solid thought behind the implementation of the current system.

In the old days on chess (long before chess.com) on a certain server if you lost connection the game was saved. And when both players were online again it was auto resumed by the server.

I had one funny experience like this where my opponent disconnected when we were both under 5 seconds and it was a complete mess of a position, we were both racing our pawns to queen and lots of other tactics going on.

Well I played some more games, then suddenly after one of my games ended that insane position immediately started. I somehow won on time by just premoving like crazy. That was fun but obviously there were downsides to that system (my opponent probably disconnected to review the position at his leisure, luckily I did the same).

On that site there was an option called "no escape" where any disconnect was an instant forfeit. So people had options.

Drawgood
It happens rarely enough that it’s not a problem. Besides there is no way to make it better than it is now. If there is no penalty for dropped connection then people will simply cheat and fake a lost connection when they’re losing.
llama44
Drawgood wrote:
It happens rarely enough that it’s not a problem. Besides there is no way to make it better than it is now. If there is no penalty for dropped connection then people will simply cheat and fake a lost connection when they’re losing.

But he's saying your clock should just keep ticking... so in the case of a losing position, how is faking a bad connection any different from refusing to move?