I’m downloading details for completed games.
About 500 games downloaded so far. It’ll take a while…
Presumably it’s these three everyone will be waiting for?
Likely, thought there are also a few with 8 still in progress, and not sure that’s enough of a difference to be important yet.
Unusual 5-4 sequence in keskinep–seranhelo.
(4) is a bit passive, as is (12), which I’d expect as an extension to the north, allowing (2) to be captured.
For those reasons, Black’s result ends up looking quite nice.
It took a couple days but eventually I have a procedure for downloading only new finished games.
I also downloaded all games that were finished yesterday.
So now I am all set for next weeks. 
The whole picture so far:
Some analysis on timeouts.
Tournament started on june 30th 2020.
On the fourth day there was a timeout hit: 480 games.
Then all went quietly for a month.
On august 7th timeouts happened again, but not in such disruptive way.
Timeouts happen on a weekly routine (not on weekends) but sometimes my time zone makes a little trouble.
I grouped games by end date’s month, so to compare timeouts over different periods.
Overall completed games decrease each month and so do timeouts…
…but as a percentage of total completed games, Timeouts are somehow increasing.
Up to yesterday we had 99 finished games in the month of May, of which 59 ended by timeout (60%).
So I’d say that @S_Alexander 's impression is confirmed.
Looks like we have 53 games paused right now… 
That’s counting stone removal too, if there’s any. 
I’m testing the Api about that, but the syntax for paused games isn’t very straightforward
More interesting 5-4 play spotted in hokaw–Ran-Ran:
I wonder, is that the most efficient way to kill a 3-4 invasion beneath a 5-4?
Is it dead?
And is it even an invasion? 
Moves 3, 5 and 7 tell a different story. Don’t they?
Why would it be more likely for someone with all games outstanding to end their last game later than a person who has only a few, or even one, game left?
You might be right - but I don’t immediately see firm logic which justifies that. Some could just be taking their time on all games equally, while others may focus on just a few at a time. Completely different approaches, and no telling which ends last.
So… it will be interesting to see at the end if the last few games that finish this round indeed came from people who currently have lots of games left.
Since time-outs are forbidden, only one game to report:
Also only one game
If both plays had tried to play the strongest moves in the top left, it could have really kicked off.
Instead White just played (1) at (2) and play shifted elsewhere.
Well, the fight looks bad for White so I can’t blame him.
The lower right “sente-taking” noseki is something I’ve been seeing lately.
White doesn’t seem to realise that exchanging (5)–(8) doesn’t mean he can happily omit X.
How long do you think it’ll be before we reach the final 500?
In Snowonthetop–NoxTheJester, X can’t really be described as a pincer.
It appears that Black can continue very comfily:
Well, in a real game (7) should really fix in the top left.
Hmm, or should it? Perhaps the top left is not quite as urgent as locking down the right side.
