EGF vs AGA pros win-and-continue match

[k0n0]

I’ll agree with that definition, Uberdude. By this definition, I don’t think that Eric was unsportsmanlike in his behavior - do you?

It is questionable whether conditions where Eric could decide between "i will be sportmanlike" / "I will be neutral" were met.

I mean, they were met only if Eric knew that Mateusz had so serious connection problems that there was a danger to lose on time.

[Kirby]

You should stop fallacious argument like this one:
last 3 pages on go4go win by 10 points or more:1/90
win by resign:73/90.

So let me clarify what you mean: it is not unsportsmanlike to go to the counting phase, but it IS unsportsmanlike to go to counting phase if you are 10 behind on the board.

Is 7 points ok? At what move shall we be required to resign? Is it a function of points?

Please enlighten me...

And what you say is not "playing until the counting phase is a totally sportsmanlike way to lose" but "playing until the counting phase hoping a win by time/win by lag is a totally sportsmanlike way to lose".(I don't say it's what Eric did)

Um, no. Why do you want to change the content of what’s i said? I believe this is called a straw man argument.

Second logic error: you assimilate loss by time and by lag as the same.
But in the first case ,its Mateus responsibility, second case it's not.(you can't ask to play 30s byo-yomi agaisnt 1 min)

This is debatable. There were no special rules for lag as far as I know. If I were playing under those conditions, I’d avoid playing on a poor connection. I’m sympathetic to his situation, so I wouldn’t say a lag is his fault. But it’s not totally resistant to mitigation, either.

If we accept loss by lag as a classic loss, you legitimate DDOS.
We are not even sure Mateus was not a victim of DDOS by an angry american kid who want to avoid a sweep.

I do not advocate DDOS in any way.

But i expect more than only neutral behavior from someone who play for his country.

I’d hope that you’d hold Mateusz to the same standard.

[Kirby]

I mean, they were met only if Eric knew that Mateusz had so serious connection problems that there was a danger to lose on time.

We need to stop treating Mateusz as the victim of connection issues. Do we have data about his network latency?

Some point to the city he was in - obviously he had a bad connection. This is problematic:

1.) If his choice of Internet connection does have worse latency, then this was a poor strategic choice on his part to play an Internet game.

2.) It has not been shown that Mateusz even had worse network latency. Some folks argue this since he lost on time. But how do we know Eric’s connection wasn’t worse? Maybe Eric lost a byoyomi period from lag. Maybe he lost thinking time. All what-ifs. We have no data on latency issues from either player, except that one player lost on time.

FWIW, my parents live in rural Michigan, in the USA. I wouldn’t think of playing a tournament game there, because their internet connection sucks.

[Tryss]

We need to stop treating Mateusz as the victim of connection issues. Do we have data about his network latency?

If there was no connection issue, there would be no debate : Mateusz would have lost.

The only factual question is : Did he makes his move on his computer in due time?

If the answer is yes, he's a victim of connection issues. Note that it doesn't say what to do in this case, you could still award the game to Eric, void the game, or whatever the organizators decide.


(by the way, if the answer is no, he's a liar, and that could be considered cheating)

[dfan]

The only factual question is : Did he makes his move on his computer in due time?

(by the way, if the answer is no, he's a liar, and that could be considered cheating)

I believe there are plenty of ways for the answer to be no and for him not to be a liar (e.g., he thought he played his move but the KGS misclick-prevention system denied it, and he didn't notice).

[Kirby]

We need to stop treating Mateusz as the victim of connection issues. Do we have data about his network latency?

If there was no connection issue, there would be no debate : Mateusz would have lost.

The only factual question is : Did he makes his move on his computer in due time?

If the answer is yes, he's a victim of connection issues. Note that it doesn't say what to do in this case, you could still award the game to Eric, void the game, or whatever the organizators decide.


(by the way, if the answer is no, he's a liar, and that could be considered cheating)

Yes, I agree that being a victim of network connectivity issues is possible - my complaint is more that it is a slippery slope to label Mateusz as a network connectivity victim without more data.

I work in the field of network analytics, and we make graphs of latencies, netflow coming from routers, and so on. One thing that's clear is that the streaming data is quite volatile. Even for a 10 second window of streaming data, you'll have jumps and drops in network activity on a regular basis. Customers are often interested in anomolies, and in order to detect anomolies, we need to analyze the trends of historical network data. For example, sometimes a router might get a lot of traffic at 10pm UTC on Tuesday, but less traffic other days.

Coming back to the problem at hand, let's assume that we say that, "yes, there was network lag at the 10 second window of time in which Mateusz lost on time". What can we do with only that data to be fair? Not much.

Let's say you make a concession to Mateusz for his network issue at time [<game-end> - X] for an X-second window of time. Then to be fair, we need to backpedal the game and see what network issues occurred at earlier windows of time. Maybe Eric had more network lag than Mateusz for the entire game. Or maybe it was up and down. Or maybe Eric had worse latency except for at the very end of the game. Who knows?

It's a big oversimplification to give an exception for a lag at one moment near the end of the game, when no concessions have been made for the other couple of hundred moves.

Unless the rules stated what to do for lag beforehand - which they didn't.

[Uberdude]

Referee (in)decision: Mateusz lost on time

Information about the 4th game of the AGA pros vs EGF pros match - I played a game with Eric Lui, which was almost finished. I was winning a lot and just very small endgame moves left on the board, so that a 20 kyu player wouldn't lose this game. I played a move before the 10s countdown. The Go server on which we played (KGS) didn't place a move after my clicking. My time fell. There were two proctors in my flat who saw that I played that move way before the end of the time.

First decision of the referee (referee of the match is Kim Myungwan) was to continue the game from the position, in which the KGS stopped to work.

Then, three players from the AGA pros team (William Gansheng Shi, Calvin Sun and Eric Lui) made an official protest - in their opinion I should lose by time. They wrote that the proctor's job is only to check if the player doesn't cheat.

Referee changed the decision - we should play a new game.

Then I spoke that it's not my mistake that the KGS doesn't work well. I spoke that in my 2nd game from this match (against Calvin Sun) I had a similar situation - I clicked when there were 3s left of the penultimate time, the move appeared on the board after several seconds, so that I lost this time. I had one more time left and decided to play (in this game and in all the next games when I had the last time left) every move before the last 10s countdown.

After reading this information, referee changed the decision again - I lost by time. This is the final decision.

[Fenring]

So let me clarify what you mean: it is not unsportsmanlike to go to the counting phase, but it IS unsportsmanlike to go to counting phase if you are 10 behind on the board.

Is 7 points ok? At what move shall we be required to resign? Is it a function of points?

Please enlighten me...

its not a question of points, its a question of intention.

And what you say is not "playing until the counting phase is a totally sportsmanlike way to lose" but "playing until the counting phase hoping a win by time/win by lag is a totally sportsmanlike way to lose".(I don't say it's what Eric did)

Um, no. Why do you want to change the content of what’s i said? I believe this is called a straw man argument.

this what you say:
"And that's why it's wrong to claim that Eric should be resigning - after all, there exist systems (like the KGS ranked system) for which a loss from lag is indistinguishable from a loss by time."
For me it means:You can't expect someone to resign if he can have a win by lag/time.
For me both part of the sentence are true,but the second one to justify the first, it is for me unsportsmanlike behavior.

I do not advocate DDOS in any way.

But if a player is DDOS, its a free-win for his opponent, and no-contest possible, because if your connection is your responsibility protection against DDOS too.

[Kirby]

Fenring, ddos attacks, trying to get your opponent to lose on time by playing inside their territory, etc. are all unsportsmanlike. None of that happened here as far as we can tell, but if it did happen, the person doing it would be cheating, Id think.

There should be a policy on what happens in the case of lag for future tournaments.

[Javaness2]

Referee (in)decision: Mateusz lost on time

Information about the 4th game of the AGA pros vs EGF pros match - I played a game with Eric Lui, which was almost finished. I was winning a lot and just very small endgame moves left on the board, so that a 20 kyu player wouldn't lose this game. I played a move before the 10s countdown. The Go server on which we played (KGS) didn't place a move after my clicking. My time fell. There were two proctors in my flat who saw that I played that move way before the end of the time.

First decision of the referee (referee of the match is Kim Myungwan) was to continue the game from the position, in which the KGS stopped to work.

Then, three players from the AGA pros team (William Gansheng Shi, Calvin Sun and Eric Lui) made an official protest - in their opinion I should lose by time. They wrote that the proctor's job is only to check if the player doesn't cheat.

Referee changed the decision - we should play a new game.

Then I spoke that it's not my mistake that the KGS doesn't work well. I spoke that in my 2nd game from this match (against Calvin Sun) I had a similar situation - I clicked when there were 3s left of the penultimate time, the move appeared on the board after several seconds, so that I lost this time. I had one more time left and decided to play (in this game and in all the next games when I had the last time left) every move before the last 10s countdown.

After reading this information, referee changed the decision again - I lost by time. This is the final decision.

Very interesting. I think we had all guessed what the original decision from Kim Myungwan was.

The reason for the protest sounds a bit petty to me. You are after all trusting the proctors to make sure that no cheating happens. It would be interesting to see the exact wording of the protest.

All in all, I'm pretty disappointed that the referee's original decision was reversed. I really expected a rematch to take place.
<edit>
It is particularly surprising to go from Restart from the clock malfunction > Rematch > Mateusz loses by time.

[Uberdude]

The second change of decision is hard to understand for me: Mateusz says he made some allowance for his poor connection (but that ended up insufficient to prevent time loss), and this makes the referee go from rematch to time loss. Presumably Mateusz is making this point as an argument that rematch is unfair to him (he worked hard in the game to make a 99.9% chance to win on the board position vs Eric, which becomes only 50% for a rematch) and resume is better, and the referee then goes the other way in a "If you don't like my idea, tough, have a worse result!" reaction. Why not apply the same reverse psychology to the first protest: "I say resume -> AGA players protest, fine if you don't like my resume idea -> Mateusz wins by fiat because that's obviously what would happen on the board!".

[Kirby]

The description here is a bit interesting. A rematch seemed to be planned after the initial protest, but then after hearing Mateusz's explanation of time troubles in the second match, the decision was changed to a loss.

[Kirby]

I guess the argument could be made, if Mateusz were having time troubles on earlier matches, then maybe he should have done more to address the issue for this game. He said that he played moves early, but that wasn’t enough if he really lost due to lag.

[Javaness2]

The second change of decision is hard to understand for me: Mateusz says he made some allowance for his poor connection (but that ended up insufficient to prevent time loss), and this makes the referee go from rematch to time loss. Presumably Mateusz is making this point as an argument that rematch is unfair to him (he worked hard in the game to make a 99.9% chance to win on the board position vs Eric, which becomes only 50% for a rematch) and resume is better, and the referee then goes the other way in a "If you don't like my idea, tough, have a worse result!" reaction. Why not apply the same reverse psychology to the first protest: "I say resume -> AGA players protest, fine if you don't like my resume idea -> Mateusz wins by fiat because that's obviously what would happen on the board!".

Indeed. If a rematch was not acceptable, you would expect a double default to be the next step. We do not know if Mateusz said that he did not accept a rematch. Given this was in a face to face meeting, the sequence sounds pretty cuckoo. Still, maybe the scandal can generate some newspaper headlines.

[Uberdude]

Well, let's hope Eric plays a stupid misclick in a huge middle-game ko fight next game