This is one of the unusual openings as @mark5000 guessed and @gennan found out - the new blackhole fuseki developed by Cho U and So Yokoku with four 5-7 in the center as openings. (the old blackhole opening is four 7-7 openings).
A lot of these very high starting positions influence-based openings had been tried before the AI-era. It is certainly playable, and sometimes with success, like black actually won this game
The key seems to be trapping the opponent into invading the center and sandwich them between corners and the sides.
I admit that I was always curious to see how would players perceived the board balance with these types of influence-heavy openings, which certainly don’t confer to any traditional fuseki and joseki. And I certainly feel the AI-era knowledge seemed to limit our imagination quite a bit.
If you were able to conceive this opening, it would be easier to understand the rest sequence - inverse the traditional idea of “hugging the corners” for territory, but extending and splitting the “edges” using the starting 5-7 stones.
Inspired by @bugcat’s wish for an easier round (and the fact that I myself am not very good at this game it turns out ^^), let’s try a round on a 9x9-board. I think it should be a little easier than the last rounds, but at the very least it’s less work to make a guess!
This game is between two GoQuest 8 dans. 20 moves have been played, 10 black stones are missing.
Click the board to make your guess. You can place any number of black stones, and I’ll give +1 point for each correctly placed and -1 for each incorrectly placed stone. I’ll wait around 2-3 days to end the round.
Small hint: the game is close to even right now (7.0 komi).
There’s two more black stones, which I don’t know where I should guess them to be. I considered E8, H3 and B5 (among others), but I couldn’t find a sequence for the moves that makes sense to me.
Wow, 7 guesses already I’ll wait until tomorrow to reveal the real board (since I said I was going to wait at least 2 days), but in the meantime, here’s a “heatmap” of all the guesses so far:
The game was played on 2020-10-12. The white player (who won by 2 points in the end) is right now the highest rated 9x9 player on GoQuest!
While checking their current rank I noticed that on their profile page it says
えっと…頑張ります…
Google translated:
Um … I’ll do my best …
Pretty humble for rank #1
Results
Amazingly, @mark5000 guessed all 10 stones correctly, and even got the order of moves right (except for the minor detail of move 11 and 13 switching places)
Maybe we should expect no less from rank #2 on the sitewide 9x9 ladder
Player | Score
Yes, one issue with this scoring system is that it doesn’t really measure how “reasonable” one’s guess is, it just measures how many stones one happened to get right (which could involve some luck, if you don’t happen to be familiar with a particular opening/corner shape).
So the scoring is not really “fair” in that way (at least for an isolated game, it averages out a bit over many games). One could think about having people vote on how likely other people’s guesses look, or even just scoring based on how many other’s guessed the same stone as you. But this would require some more work, so I think a simple scoring is “good enough” for our casual game, we just have to remember not to take the scores too seriously
I think our guesses also somewhat reflect our playing styles. I would favor splitting, contact, thickness and attack, and some would favor territory or influence, etc.
It is fascinating to overlay all our guesses together, somewhat reflecting the diversity of our styles I feel.
BTW, How did you draw that heatmap though? Maybe later rounds we could all have those heatmaps, and I wonder what they look like in our previous rounds.
This was actually an absolutely amazing concept to look at and see different answers from different people. Who cares what the original position was really like when there are so many creative guesses as to how it would have looked for them.
@mark5000 16
@gennan 5
)
| 10