Yes. I’ve been playing relatively slowly, but I’ve won a few games by resignation because I think my opponents become tired of continuing.
1 Like
missing the 5 times maximum in this copy/paste which is quite relevant info too.
How can we missed it? Because many don’t care to read that sort of thing. Easy.
I read the Documentations & FAQ from A to Z, but I think I am an exception 
There is a lot of interesting info there.
So, if you have questions, feel free to ask. I like to do research, so I won’t mind if you ask me. 
2 Likes
Looks like linear trend line doesn’t fit anymore:
Second degree polynomial fits way better:
According to this model, we should reach 10038 games… well… never!
Second degree polynomial isn’t a good choice in the long run. 
Since decided games number can only increase and we expect that it should increase slower and slower, maybe a logarithmic curve could fit… well… not really:
Looks like we are increasing faster than logarithm. According to this model, we should reach 10038 games before day 500! Not bad!!!
There’s hope for the elderly people like me!
7 Likes
I’m thinking the curve should be asymptotic with the 10038 line.
Like an asymptotic regression?
with the curve meeting the line at eternity.
2 Likes
We could share some of the more interesting games. We’d have to be careful not to discuss any ongoing ones, though (but just posting them should be fine).
Data update:
Total number of participants: 2233
Total number of players in this round: 2233 (100.00%)
-----------------------------------------------------------
The following data is about the current round only.
-----------------------------------------------------------
Disqualified or dropped out: 334 (14.96%)
Still competing: 1899 (85.04%)
Number of groups: 224
Progress: 6412 games decided out of 10038 (63.88%)
-----------------------------------------------------------
Max. games decided by a single player: 9
Min. games decided by a single player: 0
Max. games still open for a single player: 9
Min. games still open for a single player: 0
Max. fraction of games done by a single player: 100.00%
Min. fraction of games done by a single player: 0.00%
Number of players having finished all games: 381 (17.06%)
Number of players having finished no games: 85 (3.81%)
-----------------------------------------------------------
Groups with all games finished: 3 (1.34%)
Groups with no games finished: 0 (0.00%)
7 Likes
Cool!
I’ve done some tests. The curve changes a lot depending on parameters.
I tried to eyeball them, but I don’t know a proper way to optimize them.
These tests brought me to a conclusion, though: probably we should expect three distinct curves:
- the first month / 40 days with many games decided because of timeouts, this isn’t very representative and should be treaten as starting turbulence
- a curve for fast players
- a curve for slow players
The overall curve should be the sum of them. So I expect three distinct trends in the final curve.
First trend is gone. Sadly I missed half of it, having data only since july 16th.
I believe that in the next two/three months we should reach about 8k games, continuing the current almost linear trend. Then I expect a knee in the curve and the slowest trend will prevail.
We’ll see. 
3 Likes
If people really are deliberately trying to slow the tournament down, then it becomes as much a test of their patience as anyone else’s. Can they really be bothered to spend the next 4 years of their life visiting this site just to annoy people? I think (hope) we’ll see them drop out over time.
5 Likes
I’m not sure you would need more than one curve. Agreed the first 40 days should be excluded as noise, but surely there is a relatively even spread in between fast and slow players, it’s not boolean.
2 Likes
It’s not really noise if you happen to understand it. Like didn’t people exactly predict some of the big jumps in games ending due to timeouts etc?
I have added two more dashboards about Games and Points:
We are approaching 2/3 completed games and black is slightly ahead: 33% VS 32.3%
400 players completed all of their games while 71 completed none.
Zbingu (2d) joined ShioRamen (8d) at the top of the leaderboard.
7 Likes
One more dashboard with details about disqualified and resigned players.
You can click to select Disqualified or Resigned and then have some more information hovering on charts
3 Likes
Data update:
Total number of participants: 2233
Total number of players in this round: 2233 (100.00%)
-----------------------------------------------------------
The following data is about the current round only.
-----------------------------------------------------------
Disqualified or dropped out: 355 (15.90%)
Still competing: 1878 (84.10%)
Number of groups: 224
Progress: 6667 games decided out of 10038 (66.42%)
-----------------------------------------------------------
Max. games decided by a single player: 9
Min. games decided by a single player: 0
Max. games still open for a single player: 9
Min. games still open for a single player: 0
Max. fraction of games done by a single player: 100.00%
Min. fraction of games done by a single player: 0.00%
Number of players having finished all games: 438 (19.61%)
Number of players having finished no games: 68 (3.05%)
-----------------------------------------------------------
Groups with all games finished: 3 (1.34%)
Groups with no games finished: 0 (0.00%)
5 Likes
So 2/3 games are gone so far.
Still with a slight advantage for black.
One player has 9 points and takes the lead of the ranking.
He’s a player from the top bar. The only one with 9 wins so far.
and here he is! (applause)
So far 12 players managed to win 9 games, but only one in the top bar.
I also discovered that two players are playing black in all of their games.
41 players are playing 8 games as black and the same number are playing 8 games as white.
Nobody plays 9 games as white.
4 Likes
Looking at points diagram, just to make sure I have understood how things work for the second and following rounds. Will the groups be based only on points acquired in the first round, or is the current general OGS rank taken into account? I wonder because starting at -4 and having won 7 games I’ve now 3 points. Good chances that the last game will be a win because the last opponent seems to let all his games time out and is already disqualified. So, with 4 points I would have chances to be in the second round in a group of stronger players, right?
1 Like
I think so.
Groups for second round are based on tournament points.
Since we have a lot of players, “subsequent pairing method” should be meaningful. Since it’s “strength”, I expect to see groups made of players with same points and similar strength
You should be in a group with 4 points players with a rank similar to yours.
1 Like
There are 2230 players that have have 9 games. For those players, the distribution of number of games as white is given by:
[0.00089686, 0.01838565, 0.0690583 , 0.17085202, 0.23497758, 0.24573991, 0.17668161, 0.06502242, 0.01838565, 0. ]
Compare this to the theoretical binomial (9, 0.5) distribution:
[0.00195312, 0.01757812, 0.0703125 , 0.1640625 , 0.24609375, 0.24609375, 0.1640625 , 0.0703125 , 0.01757812, 0.00195312]
We can also compare in terms of the absolute numbers and expectation:
actual = [ 2, 41, 154, 381, 524, 548, 394, 145, 41, 0]
theory = [ 4, 39, 157, 366, 549, 549, 366, 157, 39, 4]
actual denotes the observed values, while theory denotes the expectation rounded to the nearest integer.
1 Like
That’s fascinating, but I don’t understand what that means. 
1 Like