@Cedur216 said in #13:
> Guess this is successful enough to stay established for the future?
They should make it an annual event.
> Guess this is successful enough to stay established for the future?
They should make it an annual event.
Guys, this is important.
I didn't think I would agree with the complaints that all stages should be swiss, but now I understand why.
Arenas are just an immense factor of luck, depending on if you get people who can cooperate. And these are quite few. Anyone else:
-> won't berserk back
-> won't agree for early draws if both don't have a streak
-> won't resign in losing positions (stalling at worst)
-> won't accept draw in dead endgames wasting your time to no end, at worst killing your nerves so you even lose
-> won't draw in the last seconds of the arena or even refuse to resign when they clearly should, denying your points in mischief.
They have no sense of strategy and are just egoistically greedy for rating.
Also when you have two wins, the next opponent becomes an immense factor of luck. In my case, it once was an IM with 2300 who didn't even zerk back (why did you even play, buddy?)
Of course, in swiss there are many byes, but at least these are compensated by reducing your Buchholz score. However compensation for uncooperative opponents doesn't exist so much.
So, arenas are not suitable to determine the top 50 or whatever.
I didn't think I would agree with the complaints that all stages should be swiss, but now I understand why.
Arenas are just an immense factor of luck, depending on if you get people who can cooperate. And these are quite few. Anyone else:
-> won't berserk back
-> won't agree for early draws if both don't have a streak
-> won't resign in losing positions (stalling at worst)
-> won't accept draw in dead endgames wasting your time to no end, at worst killing your nerves so you even lose
-> won't draw in the last seconds of the arena or even refuse to resign when they clearly should, denying your points in mischief.
They have no sense of strategy and are just egoistically greedy for rating.
Also when you have two wins, the next opponent becomes an immense factor of luck. In my case, it once was an IM with 2300 who didn't even zerk back (why did you even play, buddy?)
Of course, in swiss there are many byes, but at least these are compensated by reducing your Buchholz score. However compensation for uncooperative opponents doesn't exist so much.
So, arenas are not suitable to determine the top 50 or whatever.
Any links to invitational arenas?
@Serezha139 as the announcement says, you will get it in time
@Cedur216
I disagree. Arenas are fine. I only stand by the opinion that the berserk option should be banned in such arenas, then tournament results would be much more likely correlate with how well, in terms of _chess_, players have played all games.
I'm not saying that the Swiss system is not suitable for a similar purpose, but:
1. In this case there should be many rounds, I would say at least 20, and preferably 30 - so that there are many games, since we need to determine not 1 winner, but 50 "winners".
2. There should be many of these Swiss tournaments, not less than these arenas at Lichess this year, so that people have more chances.
But the Swiss system has its disadvantages, for example:
1. In such tournaments, especially with a large number of participants, strong players have a lot of games with players who are weaker. In arenas, the higher you go up, the more likely you are to play each next game with an equal or stronger opponent (according to the standings) - I find the chess results of this approach to be more reliable.
2. By default, there are no repeated parings with the same opponent. In arenas, you may play the same player more than once, especially if you perform similarly in the tournament. Thus, you will have a mini-match to determine who performs better.
3. Winners often don't play each other. In arenas, winners are much more likely to play each other, often more than once. This applies not only to tournament winners, but also to players on adjacent places.
4. When we try to determine not 1 winner, but 50 "winners", if you are already in the top 50, in the Swiss system it becomes easier to use "negotiated" draws with equal opponents to get points and keep your current place in the top 50, and play only with those who are formally weaker. In arena it’s more difficult to do this, because in certain cases you can get 0 points, as well as lose a streak (or delay its possibility for one more game), if we use arena streaks option.
5. I personally don’t want to determine the top 50 by extra indicators (Buchholz, etc.). This situation may arise, if you don’t try to avoid it. Look how dense the distribution of points is even in arenas - with the Swiss system, I assume, it will be even denser.
Although the question of the "ideal" system is an open question.
Personally, I'm quite happy with the arena option in the early stages so far. The main thing I don't like is the using of the berserk option in such tournaments, because as a result many players who played worse games take higher tournament places. By banning berserk, we minimize the probability of that happening.
I disagree. Arenas are fine. I only stand by the opinion that the berserk option should be banned in such arenas, then tournament results would be much more likely correlate with how well, in terms of _chess_, players have played all games.
I'm not saying that the Swiss system is not suitable for a similar purpose, but:
1. In this case there should be many rounds, I would say at least 20, and preferably 30 - so that there are many games, since we need to determine not 1 winner, but 50 "winners".
2. There should be many of these Swiss tournaments, not less than these arenas at Lichess this year, so that people have more chances.
But the Swiss system has its disadvantages, for example:
1. In such tournaments, especially with a large number of participants, strong players have a lot of games with players who are weaker. In arenas, the higher you go up, the more likely you are to play each next game with an equal or stronger opponent (according to the standings) - I find the chess results of this approach to be more reliable.
2. By default, there are no repeated parings with the same opponent. In arenas, you may play the same player more than once, especially if you perform similarly in the tournament. Thus, you will have a mini-match to determine who performs better.
3. Winners often don't play each other. In arenas, winners are much more likely to play each other, often more than once. This applies not only to tournament winners, but also to players on adjacent places.
4. When we try to determine not 1 winner, but 50 "winners", if you are already in the top 50, in the Swiss system it becomes easier to use "negotiated" draws with equal opponents to get points and keep your current place in the top 50, and play only with those who are formally weaker. In arena it’s more difficult to do this, because in certain cases you can get 0 points, as well as lose a streak (or delay its possibility for one more game), if we use arena streaks option.
5. I personally don’t want to determine the top 50 by extra indicators (Buchholz, etc.). This situation may arise, if you don’t try to avoid it. Look how dense the distribution of points is even in arenas - with the Swiss system, I assume, it will be even denser.
Although the question of the "ideal" system is an open question.
Personally, I'm quite happy with the arena option in the early stages so far. The main thing I don't like is the using of the berserk option in such tournaments, because as a result many players who played worse games take higher tournament places. By banning berserk, we minimize the probability of that happening.
Removing berserk would indeed remove some of the inconsistencies I mentioned.
(btw expecting 1000 participants, 10 or 15 games would be fine for a swiss)
(btw expecting 1000 participants, 10 or 15 games would be fine for a swiss)
Isn't there a weighting not to play the same player in an arena? I've noticed if I play the same player twice, it's with opposite colors, and I don't tend to play the same player more than twice.
@umhia because the arena is large enough to make a threefold pairing quite unlikely.
But in smaller arenas, or with many games, the same pairing can be arbitrarily frequently
But in smaller arenas, or with many games, the same pairing can be arbitrarily frequently
This topic has been archived and can no longer be replied to.