I developed a completely new computer assisted opening theory that is aimed for the KGS blitz games. The idea is that I draw randomly first four moves and does not care where opponent plays their own opening moves. I use a php script for generating random moves, thus computer assisted. I got the inspiration for this because Zen is my favorite go player and I wanted a little irrational MC gobot feeling for my kgs games.

There are two aspects that are little surprising. There is little relevance for the final outcome where the first four moves are. Especially when playing black, result after opening is usually roughly even. When playing as white games are sometimes more difficult than normally, but still often manageable. However other thing that is even more surprising is that I have not seen any other using such opening theory. Thus I declare that it is new opening theory.

Anyways I have now played some 35 rated blitz games on KGS and I have had no difficulties of keeping 2d rank there. Although I am not that particularly strong for 2-dan so handicap what is given by placing first four stones randomly is far less than one stone. I usually assist a little if it seems that local response is urgent. When playing as black assistance is hardly never required, and when playing as white assistance is required sometimes but perhaps less than one out of 4 games.

I would like to discuss about people's personal likings towards highly experimental fuseki theory.

"I usually assist a little if it seems that local response is urgent." Does this mean you don't always play the first 4 stones randomly, but only when you decide it's OK? If so, that's not really the same as true randomness. I'm not really surprised to hear opening doesn't matter too much in blitz, though.

No it does not mean that. It just means that if randy draws a random move and opponent replys with tsuke, it is reasonable to reply nobi locally. Like I stated that these things are rare and relevant only if opponent's moves are urgente sente moves. Sente is a move where tenuki to any other part of the goban will lead to the immediate loss of points.

I'd be interested to hear how it does in non-blitz games. I find new and experimental fuseki interesting in theory, but I haven't tried any in practice because I'm not strong enough to play out an opening that I don't yet know the general properties of.

I'm assuming you exclude plays on the first line at least.

Interesting idea and I might try it.

However you should either state up front that you'll respond to a contact move locally, or not deviate from those four moves no matter what. I think I will try the latter.

_________________
That which can be destroyed by the truth should be.
--
My (sadly neglected, but not forgotten) project: http://dailyjoseki.com

Yes, I have excluded intersections below third line and also I have excluded coordinates that are next to each other (nobi). I do not have had courage to try with non blitz timings =)

You might be interested to check out the games of Robert Jasiek (sum on KGS). His account is 5d and his first few moves often look pretty random.

dezomb is another KGS character with, how to phrase it, imaginative openings.

This just shows how little all this theory stuff really matters

I've heard more than once before the notion that your first two moves can be anywhere (except the first line), and only from move three should you be thinking about coordinating the stones.

No, it just shows how few people understand opening theory. If Liisa encounters someone who does, he/she is going to get clobbered.

_________________
Help make L19 more organized. Make an index: https://lifein19x19.com/viewtopic.php?f=14&t=5207

It's quite astonishing how often you will find "wrong direction" in professional's comments on even high amateur Dan's games.

This will be true with "normal" Fuseki as well as with "unusual" Fuseki.

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

Actually, I suspect this is a testament to how well Liisa does understand opening theory, in knowing what to do with those 4 stones.

Yes. He has written before, I believe on gd, that he sometimes uses random moves for the first two moves.

_________________
FusekiLibrary, an opening library.
SGF converter tools: Wbaduk NGF to SGF | 440 go problems | Fuseki made easy | Tesuji made easy | Elementary training & Dan level testing | Dan Tutor Shortcut To Dan

I do not think so. First if someone understand's fuseki significantly better than my randy, their rank would not be 2d. And second is that in Kgs blitz wins the player who has bigger muscles. It is not about winning by four points, that is what you might loose by playing non ideal opening moves, but finding a sequence that opponent cannot read in 10 seconds.

When Wang Yang 5p was the main teacher on a Go 'n' Games Go to China trip I asked him about how much "worse" eg the Big Cross or Great Wall openings are in comparison to more traditional openings.
As far as I remember he said that for amateurs it makes no difference, it is more important to play a opening that you "know", so if you study the Great Wall opening it will suit you just as good as eg San-Ren Sei or Kobayashi.
Again if I remember correctly he said that on a professional level, there may be a difference of 4-5 points to use a less optimal opening.

/Mats

_________________
mohsart - games & books
http://spel.mohsart.se/

To paraphrase some things I've heard before:

Playing strange (random certain qualifies as strange) starting moves stops your opponent playing the opening by memory. ## I would say this is your key advantage, and this is a reasonably common idea

In amateur games, by pro standards many game losing mistakes are typically made on both sides.

And Kajiwara:
There are people who think that it makes little difference how they play in the opening. Ridiculous! A game is often decided in the opening.


Expanding on Kajiwara's thoughts; I think that if you are at a territorial disadvantage after the first few moves, you must struggle to catch up (which is hard/risky).

_________________
Revisiting Go - Study Journal
My Programming Blog - About the evolution of my go bot.

I agree that the opening is far less important than most players I have encountered (in the U.S.) think. If you look at the Korean approach to go (heavy emphasis on reading, playing faster games, playing more games as opposed to "studying"), they are on average far stronger players who improve much more rapidly. I would go as far as to say that playing random moves could actually be more useful in getting weaker players away from the bad habits of memorizing joseki, trying too hard to punish moves in the opening they see as "wrong", and seeing the whole board. I realize that a lot of players find the opening more interesting to study, but if the main goal is to further understand and improve at the game of go- then it's really not the most important thing.

I've played many games on KGS with the first two moves or so on semi-random points (3rd line or higher, away from other stones). I don't think it makes any difference to the final result in blitz games. If anything, I think it's better as I become interested in the game from an earlier stage as I try to make my initial stones useful. Another advantage is the sheer number of people (at ~KGS 2D at least) that play out joseki even though their position falls apart only a few moves down the line due to a nearby 8-6 point or something. It's definitely worth trying as a good example of the proverb that memorising joseki is bad for you.

I think that the issue is that 2 dan players are really pretty weak in the opening. Even in slow games they make many mistakes but in blitz games it really doesn't matter. Try this strategy against a 5 dan and I think you will be well behind at the end the fuseki in every game.

_________________
Still officially AGA 5d but I play so irregularly these days that I am probably only 3d or 4d over the board (but hopefully still 5d in terms of knowledge, theory and the ability to contribute).

Actually I tried this against KGS 5 dan (about ORO 6-dan), and I won three stone random handicap game by 6 points (NZ rules was used). Then I played right away another regular handicap game with Japanese rules and I got crushed by 101.5 points. I did not have much chances during the game and I got continuously outread by white.

Sample size of course is not statistically significant, but if there is much bigger difference than one stone, this should count something. So even 5-dans has some memorized patterns that are effective strategies against 4-4 handicap stones what they cannot utilize against random handicap stones.