I've got an 808 and a 303 and a record collection like the ABC
Phil 20k (black) versus Andreas 18k (white), two stone handicap, 0.5 komi.
I have never met Andreas, and all I know about him is that a) he's Dutch, b) he's jovial, and c) he tends to play a better fuseki than me. Like yourself, he's one of those global villagers from teh intarwebs that my girlfriend hates to find passed out on the couch when she comes home after a quiet dinner with her colleagues. If I ever do meet him face to face, I'll be sure to upload a photo of the aforementioned tableau, beer bottles, go stones and all.
Anyway, since we couldn't meet over a real goban with beer and sake nearby, this game was played on the Dragon Go Server, with coffee and eyestrain -- the fukujinzuke of PBEM gaming -- as its accompaniments.
The game took place during August and September 2006, at a rate of about three or four moves per player per day.
Since reading a jargon laden session report of a themeless game is at best confusing and at worst boring, I suggest you download the SGF of the game and follow this comedy of errors move by move, in the go software of your choice.
The SGF of this game can be found here:
And a jargon-busting geeklist can be found here:
Opening (moves 1 to 40)
The go server automagically placed my handicap stones on the top right and lower left hoshi, then Andreas opened on the top left 3-4 point. I responded with a stone on the lower right 3-4 point, and was surprised when Andreas attached at the star point instead of playing a high approach. Figuring that a contact play would put me ahead in local liberties, and perhaps lure Andreas into an ill-advised crosscut, I played a hane above the star point. We both extended away from the potential crosscut, then Andreas had to solidify his position with an empty triangle in gote.
Happy to have sente in the opening, I played a high approach on white's top left stone, then for some odd reason answered his shape move by extending in the wrong direction... inwards towards the hoshi. A few more moves, and Andreas had substantial profit in the top left corner, while I had a shabby looking wall with a white stone poking around one end. Ouch.
Andreas now played a high kakari to the right of my lower left 4-4 stone, facing his empty triangle in the bottom right and sketching out some thin moyo. I perhaps should have played a pincer next, but instead I played keima on the other side, and let him play the bottom centre hoshi, loosely linking his empty triangle on the right and the high kakari in the left. The top right corner was still all black, so I played a keima shimari to reinforce my position. Andreas continued to attack the sides, this time playing a splitting move on the centre right handicap point. This looked a bit high to me, so I attached underneath, and after we exchanged a few more stones I had a good inside wall on the right meandering along the lines of three and four, which made up for my loss in the top left. When the wall was close enough to my shimari in the top right, I tenukied to the centre left, making a decent moyo there also, and Andreas played tengen in the hope of getting some utility out of his side of the wall.
In the end I think I got a bit more profit out of the fuseki than white, but less influence and fewer sente options. If it weren't for the handicap stones the game would've been a loss for me at this point.
Middlegame I: The Splittening (moves 41 to 112)
Using my lower left stones as a base, I jumped out towards the white stone in the centre. Andreas attached, I played hane, then we both extended from the cutting point (white towards his base on the lower edge, black advancing eastward into white territory). We both reinforced our stones in the area, but I like to think my shape was a bit better than that of my windmill-loving adversary. Black held on to sente through this, which I used to sketch out a little bit of territory on the top and bottom edges while pressing the white centre. It didn't last long, as I soon played a gote peep, letting Andreas go on the attack and indent my territory on the left a bit.
My mistakes in the opening came back to haunt me now, as white cut through my thin position in the top left and surged down from the corner. I was forced to defend, and Andreas laid down a series of contact moves in sente, making shape in the centre as he went, and threatening to split my upper and lower groups on the left. For some reason he didn't follow through with the split, and instead played a small yose move in the lower right, presumably keeping the split in hand as a ko threat (though playing it would've set up two money jump style extensions as threats). After a bit more wall-building it was white's turn to make a non-compelling peep, and I was able to indent his central territory a bit more.
We played a few more solidifying moves on the left side, but my eye kept returning to the area I'd just indented. It looked like something out of a tesuji book, with a bunch of cutting points and weird shapes full of potential for squeeze plays. I didn't think I could capture anything there, but I could hopefully squeeze white into a dumpling and get Andreas to waste a few stones fending off my attack. As soon as I got a chance I clamped a weak stone in the in in the middle of the area, with a view to setting up an oiotoshi or double atari.
Middlegame II: Clamp of Fury (move 113 to 154)
With the clamp in place, I studied the board a bit more. I could see an obvious defensive move (a dumpling-like solid connection), but figured I had nothing to lose if it were played... white would have to play a stone to match my investment, and I'd keep sente in any case. I waited for Andreas's move with bated breath, and he made shape instead of connecting solidly. The time was now right to beat down an innocent Dutchman.
I went in with a double atari, thinking that white would solidly connect and perhaps set up a ko. Instead he went for a sealing move then made shape again, leaving a conspicuous chink in the white wall. Andreas must have been really flustered now, because he kept making contact plays as I dripped stones through, when a net would have stopped me in my tracks. After a bit more dripping I was bearing down on the white group in the lower right corner, which was suffering a major shortage of liberties. Hoping to capture the lot with a squeeze play and a cut, I played a throw-in followed by atari on half of the group. Andreas saw what was coming and let me capture five stones while he sealed me in for good. I cut and squeezed a bit more to set up a ko threat with the remainder of the white group, then turned my attention elsewhere and pressed against some other weak areas in sente.
Now it was my turn to do something foolish, and, dazzled by my thrilling ko threat, I set up a gratuitous ko in the centre of the board (the excuse being that it would create some cutting points or something). Anyway, my reading was flawed and the net result was that white gobbled up two of my stones, took sente, and invaded my top right corner. Aargh!
Middlegame III: Dutch Invasion (moves 155 to 180)
My position was pretty thin at the top, but I couldn't afford to let white's invading group live small at this stage. I plugged some gaps while Andreas started making shape, then I invaded his invasion, aiming for a nakade shape or maybe bent four (I'm still at the stage of learning where seki never occurs to me as a target to shoot for).
White still had some outside liberties by the time I had something resembling bent four set up, so it came down to another ko fight. Andreas started the ko, so I replied with the thrilling ko threat that I'd set up earlier. Now it was white's turn to find a threat, and I waited for him to play the horrible splitting move on the left side that I had been fearing throughout the game. Happily he seemed to have forgotten about it, and cut in the lower right corner instead. I ignored the threat and reclaimed the top right corner, then we reinforced our positions some more.
Endgame (moves 181 to 215)
With the fighting over, we went into the endgame and solidified our territories with yose moves. Andreas kept sente for most of the endgame, picking up a goodly number of points, but when I had it briefly, the first thing I did was block the dreaded splitting move and the potential psuedo monkey jump extensions.
The territory looked pretty equal, so the final result was going to come down to the prisoner count. I had the edge here due to my clamp and subsequent rampage, and sure enough I eked out a +10.5 pt victory.
Going over the game, it's pretty clear that I was behind up until my lucky clamp, and would have suffered a resounding loss if Andreas had made more solid connections in the middlegame instead of worrying about shape. The lesson here (according to our 18k-ish analysis) is that you should aim for thickness rather than shape when your group is large and the threat is weak (or if you already have eyes!). There was never any serious chance of me capturing such a large group, so solid connections to defend against cuts were more appropriate than hanging connections and shape moves in general.
And the personal lesson that I took out of the whole thing is that I really should start reading that copy of Otake's Opening Theory Made Easy if I expect to defeat Andreas in an even game any time soon.
- Last edited Thu Oct 5, 2006 1:29 am (Total Number of Edits: 1)
- Posted Wed Oct 4, 2006 3:09 am