Life In 19x19

two more points:

The first winner only plays with four players. That is kind of inadequate.
Two players might have to play with each other in six games. That is boring.

Erm. The only type of tournament that would, according to you, be neither "inadequate" nor "boring" is a round robin. But you just proposed they play swiss. Can't we just say that no tournament system is perfect and move on, rather than wrapping ourselves in antinomies?

Just because no system is perfect doesn't mean any fancy system can be applied in such an important tournament.

My post does not mean to criticize the AGA but rahter serves as a comment and suggestion for future tournaments. There will be future Pro-exam tournaments, right?

Talking about round-robin, it is by far the fairest system even though there are some disadvantages. The question is: is AGA ready for a 15-round tourny?

How about 12 candidates and a 11-round tourny?

I am not familiar with the Japanese system. According to Hikaro No GO, they use the round-robin. (Not a reliable reference, lol).
The Chinese system uses a 2-stage Swiss system, with plenty of rounds, like n+6.

I never heard a double elimination tournament described as "very complicated" or "fancy" before.

IMO I prefer double elimination to round robin since it avoids ties.

fentonaop wrote:The first winner only plays with four players. That is kind of inadequate.

fentonaop wrote:Just because no system is perfect doesn't mean any fancy system can be applied in such an important tournament.

Actually, if there's one thing a knockout tournament is good at, it is producing a justifiable first place. In this case it's a knockout after a best of 3, I think it is reasonable to say that the first place player will have been sufficiently vetted (they will have either beaten everyone, or beaten someone who beat them). The same could be said about the winner of the loser's bracket as far as earning second place. Given that the ultimate goal of this tournament is to separate the top two players from the rest of the field, I would say the format is well suited.

Mef wrote:
Actually, if there's one thing a knockout tournament is good at, it is producing a justifiable first place. In this case it's a knockout after a best of 3, I think it is reasonable to say that the first place player will have been sufficiently vetted (they will have either beaten everyone, or beaten someone who beat them). The same could be said about the winner of the loser's bracket as far as earning second place. Given that the ultimate goal of this tournament is to separate the top two players from the rest of the field, I would say the format is well suited.

Well said. We should also remember that only a handful of the sixteen players in the tournament have a realistic shot at winning. A player with a 7.0 rating may get into this tournament, but he isn't going to win. The number of players that actually have to be distinguished is few enough that the tournament format is entirely suitable.

Simple double-elimination is not a complicate or fancy system.

I mentioned 'fancy' because it applies three rounds of best-of-three matches in a total of 4-round tournament. That is very interesting because even most international championship tourneys only apply two rounds of best-of-three (or five) matches.

I mentioned 'inadequate' because the first winner only plays with four other players and the second winner only plays with five (or four) other players. They are not well exposed to many other players. This system equivalents to single elimination albeit the two winners have to defeat some of their opponents twice.

Again, I am not criticizing the AGA in this post. (that will be a new thread). It is just a technical disscussion on the format of this tournament.

pwaldron wrote:

Mef wrote:
Well said. We should also remember that only a handful of the sixteen players in the tournament have a realistic shot at winning. A player with a 7.0 rating may get into this tournament, but he isn't going to win. The number of players that actually have to be distinguished is few enough that the tournament format is entirely suitable.

I beg to differ. There are differences in their strength. However, they are supposed to be at the same level and treated equally in this tournament by organizers.

1. We should really have a sticky thread of arguments about the best tournament formats (or a senseis page, or a white paper summarizing scholarly research...).

2. In some sense, the fairest thing would be to give Andy a pro spot, without playing, and then fight it out for second. Not very sporting, though.

3. Like others point out, single/double elimination picks first and second really well. Any system has some probability of failing (that's probability and statistics for you) but it at least delivers a clear verdict.

fentonaop wrote:
I mentioned 'inadequate' because the first winner only plays with four other players and the second winner only plays with five (or four) other players. They are not well exposed to many other players. This system equivalents to single elimination albeit the two winners have to defeat some of their opponents twice.

The problem with your argument is this: If you failed to play either of the final qualifiers in direct competition, that means there were at least two other competitors in the field you have lost to. If you have already lost to two other competitors in the field before challenging the two players who have (thus far) proved to be strongest, it is very hard to make a legitimate claim you were one of the top two competitors in the tournament. One could perhaps reasonably argue that there is not enough information to distinguish the "1st place" qualifier (winner of the winners bracket] from the "2nd place" qualifier (winner of the loser's bracket), because there is a chance the 1 loss could occur to an opponent not faced by the champion...but that's not really the goal of this tournament. This tournament is meant to divide 1&2 from 3-16...and that's what it does.

Mef wrote:This tournament is meant to divide 1&2 from 3-16...and that's what it does.

Not only that, but it also cleanly selects 3&4 for seeds, as well as 5&6 for alternates.

you'll have to get the TD who designed this system to explain it... he'll do that better than me.

Mef wrote:

fentonaop wrote:
I mentioned 'inadequate' because the first winner only plays with four other players and the second winner only plays with five (or four) other players. They are not well exposed to many other players. This system equivalents to single elimination albeit the two winners have to defeat some of their opponents twice.

The problem with your argument is this: If you failed to play either of the final qualifiers in direct competition, that means there were at least two other competitors in the field you have lost to. If you have already lost to two other competitors in the field before challenging the two players who have (thus far) proved to be strongest, it is very hard to make a legitimate claim you were one of the top two competitors in the tournament. One could perhaps reasonably argue that there is not enough information to distinguish the "1st place" qualifier (winner of the winners bracket] from the "2nd place" qualifier (winner of the loser's bracket), because there is a chance the 1 loss could occur to an opponent not faced by the champion...but that's not really the goal of this tournament. This tournament is meant to divide 1&2 from 3-16...and that's what it does.

Here's a thought experiment:

You have 16 players of identical strength, On any given day, one of them will beat another one with a 50-50 shot.

The double elimination tournament MUST produce a first and second place (by design).

But that doesn't mean that it's a particularly good test of relative strength.

In this case, a proper test should find all of the players tied. And this test is not sensitive to that.

While we would want rating information to accurately capture a 50-50 situation, sometimes it's enough to know who played better in a particular game, or match, or tournament. The problem with, say, a ko isn't that the results aren't statistically robust, but rather that it gives no information about most pairs.

shapenaji wrote:Here's a thought experiment:

You have 16 players of identical strength, On any given day, one of them will beat another one with a 50-50 shot.

The double elimination tournament MUST produce a first and second place (by design).

But that doesn't mean that it's a particularly good test of relative strength.

In this case, a proper test should find all of the players tied. And this test is not sensitive to that.

I'm gonna guess that there's no tournament that you can feasibly hold in a week that gives you reliable information about relative strength in a case where players are extremely close to 50-50. That's just the nature of coin flips--someone will get lucky.

That's not to say a better tournament couldn't be selected, but your thought experiment is really the worst case scenario, and I think any tournament design will choke on it.

shapenaji wrote:Here's a thought experiment:

You have 16 players of identical strength, On any given day, one of them will beat another one with a 50-50 shot.

The thought experiment has a major flaw: With identical strength every result is justified. (And even the most elaborate tournament system can't produce a better pick than k.o., double k.o. or even a coin toss competition.)

Or in other words, the qualifier isn't intended to produce an evaluation of relative strength, but to select two players.