Fujisawa-Hosai-vs-Go-Seigen-19430225-commentary-An-Younggil-8p.sgf

(;GM[1]FF[4]CA[UTF-8]AP[CGoban:3]ST[2]
RU[Japanese]SZ[19]KM[0.00]
GN[Fujisawa-Hosai-vs-Go-Seigen-19430225]PW[Go Seigen]PB[Fujisawa Hosai]WR[8d]BR[6d]DT[1943-02-25]GC[{
  "title": "Go vs Fujisawa",
  "authors": ["Josh Hoak", "Younggil An", "David Ormerod"\]
}]SO[http://gogameguru.com/]AN[An Younggil 8p]C[### Introduction

This was the second game in a ten game match between Fujisawa Hosai 9p and Go Seigen 9p. It was played on February 25 and 26, 1943.

Go Seigen is regarded as the greatest Go player in the history of Go. We refer to him as a 'living legend' in Korea. 

He was born on June 12, 1914, so next year (in 2014), he'll celebrate his 100th birthday. By the way, according to the way age is counted in some Asian cultures, Go Seigen will turn 100 this year (in 2013).

His style of play was so different from other top pros at the time, and he won nearly all of his ten game matches (jubango).

The only ten game match Go Seigen lost was this series against Fujisawa Hosai (4-6 in Fujisawa's favor). However, Fujisawa was taking black in all games and there was no komi. They played with a handicap called 'josen' (black in all games) because Go Seigen was 8 dan and Fujisawa was 6 dan at the time.

Fujisawa Hosai was very famous for his mirror Go. He kept experimenting with mirror Go because he wanted to learn all about it.

He didn't just mimic though, he spent more time than his opponents did, and he often got into byo-yomi situations (was short of time) in the middle game. He played with the sort of full concentration that Cho Chikun 9p does.

There were two famous players named Fujisawa in the Japanese Go world. Fujisawa Shuko (Hideyuki) 9p was the uncle of Fujisawa Hosai, even though he was 6 years younger.

Fujisawa Hosai was also the first professional player to be promoted to 9 dan under the Japanese Oteai system (Go Seigen was the second). They played two more ten game matches on even terms when they were both 9 dan (the only two 9 dans at the time) and Go Seigen won both of them.
]RE[W+Resign]OH[-B-]
;B[qd]C[Fujisawa plays black. There's no komi in this game.]
;W[dc]
;B[pq]
;W[dq]C[Even though many players, including (notably) Go Seigen and Kitani Minoru, had experimented with different openings in the 1930s (the shin fuseki period), the star point still wasn't especially popular during this period (1940s).

Compared to today's games, star point openings appeared much less often back then.]
;B[od]C[This opening seemed to be popular at the time.]
;W[po]
;B[qo]
;W[qn]
;B[qp]
;W[pm]
(;B[on]
;W[pn]
;B[op]LB[on:A]C[### An Old Fuseki

Even though it makes white's stones heavy, the peep at A is ``aji keshi'' --- reduces black's future opportunities. Peeping like that is uncommon in modern games.]
;W[qi]LB[ql:C][on:A][pn:B][rn:D]C[If black hadn't exchanged A for B, he could aim to invade at C, planning to connect under at D or run away later. The forcing move at A reduced black's future potential for invading.]
;B[ce]
;W[cn]LB[fc:A][gc:C][fd:B][gd:D][dh:E]C[This large knight's enclosure was Go Seigen's favorite. You can see this enclosure a lot in his games. Since there's no komi, white should play more actively, to lead things towards a complicated game.

Black should double approach in the top left next and there are several options (A-D). If black plays elsewhere, white can play a pincer extension at E, creating a good relationship with his enclosure in the lower left.]
;B[fc]LB[ce:A]TR[od][qd]C[Black chose this double approach. This is the most common move in this situation. Black aims to develop the top area while also helping his stone at A.]
;W[ed]
;B[fd]
;W[ef]
(;B[fe]
;W[ee]
;B[ch]LB[dg:A]TR[ed][ee][ef]C[In this case, this extension is better than A, because white's stones are more solid now.]
(;W[jd]TR[fc][fd][fe]C[This move is to split the top area and pincer black's marked stones at the same time.]
;B[lc]LB[mc:A]C[This is the right move. If black doesn't play here, white's approach at A will be very good. ]
;W[ff]C[This is a vital point.]
(;B[gf]
;W[gg]
;B[hf]
;W[hd]LB[ff:A]C[White hits another vital point. This has been white's plan since playing at A.]
;B[he]
(;W[le]LB[mc:A]C[This move was fast and flexible. It shows Go Seigen's style. White's aiming at the attachment at A.]
(;B[jf]C[This jump struck at a vital point in white's shape. Now black can aim to cut.]
(;W[ld]LB[kd:A][ke:B]C[This head-butt is a good response for white. It stops black from cutting at A or B.]
(;B[kc]C[Black extended here, with fighting spirit, but it was greedy.]
;W[mc]C[White can hane here now. Go Seigen's play is very lively and sharp. He likes ko fights and trades.]
;B[mb]
;W[nc]TR[hd][jd]C[White's willing to sacrifice the marked stones, because he knows he can get compensation in the corner.]
;B[kd]
;W[ke]
;B[je]C[This was black's plan, but]
(;W[pc]C[### The Brilliant Peep

This peep was brilliant!]
(;B[qc]C[Black played here in the game, but now]
;W[jc]TR[pc][qc]C[The marked exchange will help white with the fighting at the top.]
;B[nb]
;W[oc]
(;B[ob]C[In general, this sort of move is bad, because it damages the corner, but it was inevitable in this case.]
;W[pb]
;B[md]C[Black cut here, to counter-attack.]
(;W[me]C[This atari was the right move, and ]
;B[nd]
;W[pd]C[This is the right sequence. ]
;B[nf]C[This jump is the only move for black, and ]
;W[qe]LB[ne:A]C[This hane aims at A.]
(;B[ne]
;W[pf]LB[ib:F][kb:D][lb:E][kg:A][lg:B][mg:C]C[It ended with a big trade. Black got the top and the center stones, but black's top right corner turned into white's corner. The result was favorable for white, because white can still exploit aji at A-C later, or play at D-F, so black has bad aji at the top.

White caught up a bit here, but black's still slightly ahead on the board and white doesn't have komi.]
(;B[cc]TR[dc][ed][ee][ef][ff][gg]C[This attachment was good. Black should attack white's marked group to get compensation now.]
;W[cb]
;B[cd]
;W[eb]C[White exchanged these moves in order to settle his group.]
;B[fb]
;W[fa]C[It looks like it's too early to play here, but it helps white to settle the top left group.]
;B[ga]
;W[ea]
;B[hb]TR[jc][hd][jd]SQ[kg][lg][mg]C[Now, the aji at the top's gone, but white can still make use of the marked stones stones with the forcing moves in the center (marked).]
;W[cj]LB[bc:A]C[The hane at A would be the normal move now, but white played more actively here because it was a no komi game.]
;B[bb]C[This might look small, but it's the key point for the eye space of both groups and that makes it the key point for power in this part of the board. You should learn to spot these kinds of urgent points.]
;W[ba]
;B[ab]
;W[eh]LB[bg:A]C[This jump was very good. White aims to make a placement at A later.]
;B[di]
;W[fj]LB[pf:A]C[The opening up to A was good for white, but the game's still ok for black because there's no komi.]
;B[dj]
;W[ck]
;B[dk]
;W[cl]
(;B[fk]LB[ec:A][dd:B]C[Black should have exchanged A for B, before attaching here.]
;W[gk]
;B[gj]
;W[fi]
(;B[ec]C[This was a probe, but it was a bit too late.]
(;W[ek]TR[ba][ea][fa][cb][eb][dc]C[Go Seigen decided to give the marked stones away. ]
;B[dd]
;W[bg]C[This placement was very sharp. The timing was perfect.]
;B[cg]
;W[bh]
;B[bf]LB[cf:A]C[Black has to answer here, otherwise white can wedge and cut with A.]
;W[fl]LB[cc:A]C[This ponnuki is very nice. The result from A up to here was also good for white, and the game became even now. That means white had already caught up by as much as komi (i.e. 6-7 points).]
;B[iq]LB[jq:A]C[This was the right extension. It's better than playing at A, because black wants to limit white's influence in this game.]
;W[gq]
(;B[kp]LB[lp:A][mp:B]C[This was the proper move in this case. It's better than A because white can't easily invade at B in this case.]
(;W[oo]
;B[no]
;W[nn]
;B[mo]C[The game's even now. ]
;W[pp]C[It's a very good time to probe. ]
(;B[oq]C[This connection was right. ]
;W[qq]
;B[rq]
;W[qr]
(;B[rn]C[### A Small Mistake

This hane showed fighting spirit, but it was too much.]
;W[rr]C[White seized a good chance here. Go Seigen doesn't miss his opponent's tiny mistakes.]
(;B[rm]C[This is the only choice for black. ]
;W[rp]
;B[ro]
;W[sq]C[White lived in the corner, and black's lower side became thin.]
;B[qk]C[Black has to look after this group now.]
(;W[ok]
(;B[ri]C[Attaching like this is a good way to settle or make shape.]
;W[rj]
;B[qj]
(;W[rh]C[This atari was right. ]
;B[pi]
;W[qh]
;B[rk]
;W[si]
;B[oj]C[Black makes good shape and the game's still close, but white's slightly ahead now.]
;W[pk]
;B[pj]
;W[mn]
;B[nk]C[The hane at the head of two stones is nearly always good.]
;W[nl]
;B[mk]
;W[ml]
;B[lk]
;W[km]
(;B[lo]C[This move looks small and slow, but it was necessary.]
;W[jk]C[This knight's move was very good, and it's hard for black to make territory in the center now. The game's still slightly better for white.]
;B[pa]C[### The Losing Move

This hane was a probe, but it was premature, and it became the losing move. ]
;W[qa]LB[qb:A]C[White doesn't want to pull back at A, because the game's very close.]
;B[qb]LB[oa:A]C[Black can't connect at A, because he'd lose two points, so he cut here. ]
;W[oa]
;B[ra]
;W[pa]
;B[sb]C[Black lived in the corner. It appeared to be very sucessful for black, but white had a plan at the top.]
;W[re]C[The timing of this forcing exchange was exquisite.]
;B[rd]LB[rc:C][sc:B][sd:A]C[Black has to answer, otherwise white can capture black's top right group with A (which makes miai of B and C).]
;W[kb]C[### White fights back

This is what white was aiming at. ]
;B[lb]
;W[la]C[This double hane was brilliant, and white created a big ko.]
;B[jb]
;W[ja]
;B[ka]
;W[mq]C[It was very unfortunate for black that this area was a ko threat factory for white.]
;B[nr]
;W[kb]
;B[mf]C[This was a good move for black, and now, white has two options.]
(;W[na]LB[ma:A][pe:B]C[White played atari here instead, making the ko even bigger. Go Seigen is extremely good at ko fights and trades, and he must have been sure that he was winning at this stage.]
;B[ka]
;W[jq]C[This was a good ko threat. ]
(;B[jp]C[Black must answer here. ]
;W[kb]
;B[lm]
;W[ll]
;B[ka]
;W[ir]
;B[jr]
;W[kb]
;B[kl]C[Black had some ko threats here, but they damaged the center.]
;W[kk]
;B[ka]
;W[kq]
;B[hr]
;W[kb]
;B[jl]
;W[ln]
;B[ka]
;W[ko]C[This was a good ko threat, and black doesn't have any ko threats after this.]
;B[jo]
;W[kn]
;B[ma]C[So black had to end the ko, but black wouldn't have felt happy about that.]
;W[kr]LB[lp:A]C[This atari was right. If white just cuts at A, black can atari at this point.]
(;B[is]
;W[lp]C[This was so big, and it was big enough to get compensation for losing the ko at the top.]
(;B[gr]C[Black pushed here to live, ]
(;W[ip]C[This cut was good. ]
(;B[eq]C[Black attached here to take sente, but it was questionable. ]
;W[er]
;B[fr]
;W[dr]C[White's corner became solid territory, and the game became easy for white.]
;B[pe]LB[rd:B][re:A]C[This move was very big, but black's bottom group wasn't alive yet. And now, you can see that the exchange of A for B earlier was very far sighted and good for white.]
;W[hq]C[This atari was good, and it's painful for black because he's of short of ko threats.]
;B[fq]
;W[ir]
;B[dp]
;W[cp]
;B[iq]
;W[io]C[White's position was already good enough, and black still wasn't alive.]
;B[do]
;W[co]
;B[hp]
;W[ir]
;B[cq]
;W[cr]
;B[iq]
;W[gp]
;B[ho]
;W[ir]
;B[bq]
;W[br]
;B[iq]
;W[in]
;B[go]
;W[ir]
(;B[sp]C[This ko threat was too small, but it couldn't be helped.]
;W[iq]C[White eliminated the ko, and the game was practically over now.]
;B[rq]
;W[pr]
;B[or]
;W[rp]
;B[pl]
;W[ol]
;B[rq]
;W[mr]
;B[sr]C[Even though black won this ko, white's lower side territory was huge.]
;W[np]
;B[rs]
;W[ki]C[White's winning by about 15 points.]
;B[dl]
;W[dm]
;B[ii]
;W[ji]
;B[gh]
;W[fg]
;B[ih]
;W[ij]
;B[hj]
;W[hk]
;B[jh]
;W[mi]
;B[nj]
;W[lg]
;B[kg]
;W[li]
;B[lf]
;W[ql]
;B[el]
;W[em]
;B[hg]
;W[gi]
;B[hi]
;W[ci]
;B[dh]
;W[df]
;B[cf]
;W[qf]LB[pa:D][mc:A][fl:B][rn:C]C[### Black resigns

It was an interesting game.

The first fight, starting with A, was very interesting, and the result of the trade was good for white.

The second fight, ending at B, was also good for white, and the game became even (without komi).

Black C was a questionable move, and white took the lead in the game.

Black D was a bit too early, and Go Seigen showed one of his specialties - fighting ko. He was able to use the ko to finish the game decisively.

Throughout the whole game, I couldn't find any mistakes or bad moves for white, and that's very impressive.

Top players today still make lots of mistakes. Even though the short time limits of today's games can partially explain this, it doesn't diminish Go Seigen's greatness and uniqueness.

Fujisawa Hosai was also one of the best players of his generation, but because Go Seigen was too special, Fujisawa's moves don't seem to be that special in comparison. His greatness was overshadowed by Go Seigen.

However, we should remember that he was the only player who ever managed to defeat Go Seigen in ten game match.

I hope you've to enjoyed seeing this classic game and can feel some of the differences between this game and modern games. :)

Thanks, 

Commented by An Younggil 8p

http://gogameguru.com/])
(;B[pl]C[This is a good ko threat, but ]
;W[ol]
;B[iq]
;W[fp]
;B[fo]
;W[ir]
;B[om]
;W[ql]
;B[iq]
;W[rl]C[Even if Black takes at White (a), Black can't answer this ko threat, and]
;B[ep]
;W[sk]C[Black lives, but White's winning by about 10 points. ]))
(;B[fq]LB[io:B][dp:C][fp:A][cq:D]C[Black should have played here instead. A is sente (because it aims to play B), so black can attach at C or D later. This way would have been more complicated than the game. Since black's already losing, he should look for a way to complicate things.])
(;B[io]
;W[hp]TR[io][jo][jp][kp]C[There's no way for black to save the marked stones. ]))
(;W[fq]C[If white just answers here, ]
;B[mr]C[Pushing in here is good, and]
;W[np]
;B[lr]
;W[lq]
;B[ks]C[Black can connect under to catch up.]
;W[pe]
;B[ho]C[The game becomes close again. ]))
(;B[np]C[If black connects here to fight,]
;W[ip]C[This cut is good. ]
;B[io]
;W[hp]
(;B[ho]
;W[hq]
;B[ir]
;W[gr]C[Black's in trouble. ]
;B[ks]C[If black keeps going, ]
;W[lr]
;B[mr]
;W[ls]
;B[ms]
;W[js]
;B[pl]C[This is black's only big ko threat.]
;W[om]
;B[ks]
;W[lq]C[But white can connect here now. Black needs another ko threat, but he doesn't have any that are big enough.])
(;B[mr]C[If black plays here,]
;W[ho]
;B[in]
;W[hq]
;B[ir]
;W[im]
;B[hn]
;W[gn]
;B[hm]
;W[hl]
;B[gm]
;W[fm]C[Black's cutting stones can't escape.])))
(;B[lp]C[Black can't connect here, because]
;W[js]
;B[hq]
;W[gr]
;B[hs]
;W[mr]TR[jo][lo][mo][no][jp][kp][lp][op][hq][iq][oq][pq][hr][nr][hs]C[There's no way for black's marked stones to live.]))
(;B[ma]C[If black doesn't answer, ]
;W[jp]
;B[jo]
;W[ip]
;B[kq]
;W[jr]
;B[kr]LB[iq:A]C[White captures the stone at A in sente, and]
;W[pe]C[White is winning now.]))
(;W[ib]C[This is possible, and ]
;B[kf]
;W[na]
;B[lf]
;W[ma]LB[qb:A]C[White can connect under, but the game's still very close. Black lived in the top right corner, and white saved the stones at the top. The result is pretty even from A up to here.]))
(;B[jk]C[If black doesn't reinforce at the bottom, ]
;W[mq]C[White can invade here. ]
(;B[nr]C[If black responds here, ]
;W[lo]
;B[lp]
;W[mp]
;B[np]
;W[mr]
;B[lr]
;W[pr]
;B[or]
;W[ms]TR[mo][no][np][op][oq][pq][nr][or]C[Black's marked stones are captured.])
(;B[mp]C[If black answers here, ]
;W[lq]
;B[kq]
;W[nq]
;B[np]
;W[or]C[White can easily connect under like this. ])))
(;W[rk]C[Pushing here doesn't work, because]
;B[pl]C[This attachment is very good, and ]
;W[om]
;B[rh]TR[rj][rk]C[White can't save the two stones.]))
(;B[pj]C[Moving out like this would be dangerous.]
;W[ni]C[White can keep attacking like this and,]
;B[pi]
;W[qh]LB[lf:B][lg:A][ng:C]C[Black's in trouble because white has forcing sequences like A-C]))
(;W[pk]C[If white attached here,]
;B[ql]C[Black can just pull back,]
;W[pj]
;B[ri]C[And attach.]
(;W[rh]
;B[rj]LB[re:C][rg:A][qh:B][so:D]C[White wouldn't like to leave black the aji of A-C and black's group is already alive because D is sente.])
(;W[rj]C[If white cuts here,]
;B[qj]
;W[rh]
;B[pi]
;W[qh]
;B[pl]
;W[ol]
;B[ok]
;W[oj]
;B[nk]LB[oi:A]TR[ol][pm][nn][pn][qn][oo][po][pp]C[Black will break through and white will lose control of the fight. Black can ladder at A next or capture white's marked stones.])))
(;B[rp]C[Black can't connect here, because]
;W[rm]
;B[sr]
;W[sn]
;B[pr]C[Even though it's no problem to capture the corner,]
;W[ro]
;B[qs]C[Black would lose four points compared to the variation I showed at move 101. That's why black couldn't play like this.]))
(;B[rr]C[Black should have played here,]
;W[rn]C[White gets to force here,]
;B[sp]C[But the game's still very close. ]))
(;B[qq]C[If black connects here, ]
;W[lo]LB[np:A]C[This attachment (aiming at A) is good and,]
;B[lp]
;W[mn]
;B[np]
;W[lg]C[White gets sente, and plays here.]
;B[lf]
;W[ng]
;B[mg]
;W[mh]
;B[mf]
;W[kg]
;B[kf]
;W[og]C[White's right side becomes very big, and white's winning now.]))
(;W[mp]C[If white invades here, ]
;B[lo]C[This is a good response, and ]
;W[mo]
;B[mn]
(;W[oo]
;B[no]
;W[np]
;B[nn]C[It's hard for white to fight or live. ])
(;W[no]
;B[oo]
;W[nn]
;B[nm]C[White can't escape like this.])))
(;B[lp]C[If black plays here, ]
;W[ip]C[This attachment is good. ]
;B[jp]
;W[io]
;B[hq]
;W[gp]LB[jq:A]C[White can cut at A to squeeze black later, and the result is good for white.]))
(;W[dd]C[If white answers, ]
;B[hj]C[Black can extend here to fight. ]
;W[ek]
;B[fl]
;W[el]
;B[ej]LB[db:A]C[White can't make an eye at A, so it's no good for white.]))
(;B[hj]LB[ec:A]C[If black extends here, instead of playing at A,]
;W[ek]
;B[fl]
;W[el]
;B[ej]
;W[em]
;B[gl]
;W[dg]
;B[cg]
;W[ec]C[White can live here, and it's ok for white.]))
(;B[ec]C[If black had played here now,]
;W[dd]C[White would answer and,]
;B[fk]
;W[gj]
;B[gk]
;W[hj]
;B[iq]
;W[gq]
;B[il]C[The game could continue like this. Black still has a small advantage (because there's no komi).]))
(;B[bc]C[This knight's move might look good, but it's too loose.]
;W[cf]C[White can attach here and crosscut.]
;B[bf]
;W[be]LB[cd:B][bg:A]C[White makes miai of A and B.]
;B[bd]
;W[bg]
;B[ae]
;W[cg]C[Black won't be happy with this result.])
(;B[ck]C[If black extends here instead,]
;W[be]C[This attachment is good, and]
;B[bf]
;W[bd]C[White can make a base for his group in the corner. Next...]
;B[iq]C[If black takes a big point, like this,]
;W[kb]
;B[lb]
;W[ib]LB[la:A][gb:B]C[White will bring his stones at the top back into play, making miai of A and B. It's quite annoying for black.]))
(;B[re]C[If black plays here, for example,]
;W[ne]
;B[oe]
;W[of]C[White captures the cutting stones.]))
(;W[oa]C[If white still tries to capture black, ]
;B[me]
;W[lf]
;B[mf]
;W[lg]
;B[jb]
;W[ib]
;B[kb]
;W[la]
;B[lb]
;W[ic]
;B[pd]C[White can't extend his liberties now, so black will win this capturing race.]))
(;B[md]LB[ob:A][pb:B]C[If black cuts here, without exchanging A for B first, ]
;W[ob]C[Blocking here is good.]
;B[me]
;W[lf]
;B[mf]
;W[lg]
;B[jb]C[Black needs to come back here, but ]
;W[ib]
;B[kb]
;W[la]
;B[lb]
;W[ic]LB[lc:A][mc:B]C[Black can't win the capturing race between A and B. ]
;B[pd]
;W[qb]
;B[rb]
;W[na]C[White will win this capturing race. ]))
(;B[oc]C[If black pushes down here,]
;W[nb]
;B[lb]
;W[ob]
;B[qc]
;W[qb]
;B[pd]
;W[pb]
;B[qg]C[The result looks ok for black, but ]
;W[cd]C[White can come back here, and]
;B[ck]
;W[gb]C[This move will be annoying for black.]
(;B[id]
;W[fb]LB[ib:A]C[A is sente for white later, and black's territory at the top is so small. ])
(;B[fb]C[Black can't block here, ]
;W[jc]
;B[jb]
;W[ib]
;B[ia]
;W[hb]
;B[ka]
;W[ma]C[It's a ko, and black's in trouble.]))
(;B[pd]C[If black simply connects,]
;W[nb]C[Go Seigen said he planned to trade like this.]
;B[ic]
;W[qc]
;B[qg]
;W[rd]
;B[re]
;W[rc]C[White lives comfortably in the corner and black's top right group is still unsettled.]))
(;W[jc]LB[pc:A]C[If white blocks here, without peeping at A first,]
;B[nb]
;W[oc]
(;B[md]TR[mc][nc][oc][ld][ke][le]C[Black can cut here now. White seems to be in trouble, but it's very complicated. ]
(;W[nd]C[If white plays here, ]
;B[me]
;W[ne]
;B[mf]TR[od][qd]C[It's hard for white to capture the marked stones, so it's good for black. ])
(;W[pc]C[This move looks similar to the game, but ]
;B[mf]C[This move is very nice.]
(;W[pd]
;B[me]C[This is better for black. ])
(;W[me]C[If white keeps going, ]
;B[nd]
(;W[pd]
;B[nf]TR[ld][ke][le][me]C[White still can't save the marked stones. ])
(;W[nf]
;B[pd]
;W[jb]
;B[lb]
;W[qc]C[It looks complicated. ]
(;B[rc]C[This hane doesn't work well for black.]
;W[rb]
;B[rd]
;W[ob]C[It's a capturing race, but ]
;B[pa]
;W[qb]
;B[oa]
;W[sb]
;B[na]
;W[la]TR[na][oa][pa][lb][mb][nb][kc][lc][kd]C[Black's short of liberties, so instead...])
(;B[ob]C[This push is right and,]
;W[rd]
;B[re]
;W[rc]
;B[qf]LB[pb:B][mg:A]C[Now, A and B are miai for black. ]
(;W[pb]C[If white blocks here to fight, ]
;B[mg]C[This extension is good, and ]
;W[ne]
;B[ng]TR[ld][ke][le][me][ne][nf]C[White can't save the markes stones. ]
(;W[pf]
;B[pg]
;W[oe]
;B[qe]
;W[qg]
;B[og]
(;W[rg]
;B[of]C[White dies. ])
(;W[se]
;B[rg]
;W[qh]
;B[of]C[It's about the same. White's still dead.]))
(;W[og]
;B[pg]LB[of:A][oh:B]C[A and B are miai for black. ]))
(;W[ne]C[If white reinforces here, ]
;B[pb]C[Black can capture the corner. It's a catastrophe. ]
;W[qb]
;B[ra]C[This is a tesuji for capturing races, and white's corner stones are dead now.])))))
(;W[jb]
;B[lb]
;W[ob]C[If white plays like this, which isn't good,]
;B[nd]LB[pc:A][mf:B]C[This connection is good for black. A and B are miai for black now.])
(;W[ob]C[Blocking here is the best white can do now, and ]
(;B[nd]C[This move doesn't work anymore. ]
;W[na]
;B[jb]
;W[ib]
;B[lb]
;W[ma]C[Black can't extend his liberties. White wins the capturing race.])
(;B[me]C[Pushing here isn't good either.]
;W[lf]
;B[mf]
;W[lg]
;B[jb]C[Black has to come back here, and ]
;W[ib]
;B[kb]
;W[rc]LB[la:A]C[Black needs one more move at A, so this isn't a satisfactory result for black either.])
(;B[mf]C[If black caps here,  ]
(;W[na]C[This hane is wrong, ]
;B[jb]
;W[ib]
;B[me]TR[jb][mb][nb]C[White can capture either of black's marked stones, but black will still capture the three cutting stones. This result is good for black, becase black's outside influence will be very powerful.]
;W[kb]
;B[lb]
;W[ja]
;B[kf]
;W[la]
;B[lf]
;W[ma]C[White can connect under, but ]
;B[cc]
;W[cb]
;B[cd]C[Black can come back here to attack white. This is good for black. ])
(;W[la]C[This is the vital point, and black's captured. ]
(;B[lb]
;W[na]
;B[ka]
;W[jb]
;B[ma]
;W[oa]C[The game's over. ])
(;B[jb]
;W[lb]
;B[kb]
;W[ib]TR[jb][kb][mb][nb][kc][lc][kd]C[Black can't atari on either side.])))
(;B[jb]C[Black should answer here. ]
;W[ib]
;B[kb]
(;W[rc]C[If white plays here, ]
;B[nd]C[Black can connect here now. ]
;W[la]
;B[lb]
;W[mf]
;B[qc]
;W[qb]
;B[rb]C[This is good for black. ]
;W[pb]
;B[rd]
;W[ic]
;B[na]C[This is good to avoid ko, and white's short of liberties.])
(;W[nd]C[If white wants to trade, ]
;B[me]
(;W[oe]
;B[mf]
;W[qf]C[It looks good for white, but there's bad aji in the corner. ]
;B[rf]C[White can't simply capture the corner. ]
(;W[rg]C[If white answers like this,]
;B[pf]C[This attachment is good. ]
;W[pe]
;B[qe]
;W[qg]
;B[qb]LB[of:A][sf:B]C[A and B are miai for black. ]
;W[of]C[White need to play here, and ]
;B[sf]C[Black can live.])
(;W[qg]C[If white extends solidly,]
;B[qb]C[Black's shape is flexible, and not easy to kill.]
;W[re]
;B[qe]
(;W[rd]
;B[rc]
;W[rg]
;B[of]C[White is cut off and in trouble.])
(;W[rc]
;B[rd]
;W[qc]
;B[rb]
;W[pc]
;B[sc]
;W[sb]
;B[sa]
;W[qa]
;B[rg]
;W[rh]
;B[se]
;W[sg]
;B[of]C[White is in trouble like this too.])))
(;W[pd]C[This should be the best for white, ]
;B[mf]
;W[qe]C[However, this is still better for black than in the actual game. That's why Go Seigen's move in the game was brilliant.]))
(;W[pd]C[If white wedges here first, ]
;B[pe]
;W[nd]
;B[pc]
;W[ne]
;B[mf]
;W[me]
;B[la]C[This is playable for black. ]))))
(;B[ob]C[Black can also push here first, and]
;W[pc]
;B[md]C[Then cut here.]
;W[me]
;B[nd]
;W[pd]
;B[nf]
;W[qe]
;B[ne]LB[pb:B][qc:A]C[It's similar to the actual game, but there's no exchange of A for B. White's corner is small now.]))
(;W[nb]C[If white turns here now,]
;B[ic]
;W[pc]C[And then peeps here,]
;B[qc]C[Black will just block off the corner now.]
;W[pb]
;B[qg]C[And black can make a base like this. The move order in the game was better for white.]))
(;B[mc]C[Black should have pulled back.]
(;W[kc]C[If white blocks here,]
;B[lg]C[Capping here is nice. ]
;W[ne]
;B[ng]C[The flow of the stones is good for black. ])
(;W[lg]C[If white jumps here, ]
;B[kc]C[Now black can extend here and there are no problems for black.]
;W[je]
;B[ib]C[Black can connect under like this. ])))
(;W[mc]C[If white attaches here now,]
;B[mb]C[This hane is a good response. ]
(;W[nc]
;B[ld]
;W[me]
;B[ke]LB[jf:A]C[White's in trouble. A is really helping black a lot.])
(;W[ld]C[If white comes back here, ]
;B[nc]
;W[kc]
;B[md]TR[mb][lc][nc]C[Black's shape in the corner is very solid and strong. This is bad for white, even though white has sente.])))
(;B[hg]C[If black pushes here now,]
;W[mc]C[This attachment is good.]
;B[md]
;W[ld]
;B[nc]
;W[mb]
;B[kc]
;W[jc]
;B[nb]
;W[me]
;B[nd]
;W[gh]TR[mb][mc]C[White's made good shape by sacrificing the marked stones.])
(;B[ke]C[You might worry that black will cut here, but]
;W[kd]
;B[ld]
;W[kf]
;B[je]
;W[ie]
;B[jf]
;W[if]
;B[jg]
;W[hg]C[This would be a disaster for black.])
(;B[kd]C[Cutting here isn't a good idea either.]
;W[ke]
;B[je]
;W[ie]
;B[jf]
;W[if]
;B[ig]
;W[id]LB[hg:B][jg:A]C[The ladder at A favors white, so A and B are miai.]
;B[jg]
;W[hg]
;B[ge]
;W[gb]
;B[fb]
;W[ib]C[White will win this capturing race.])
(;B[md]C[If black plays solidly like this,]
;W[kc]C[White can start making shape by playing here.]))
(;W[ke]C[White could also move out more carefully like this, but it's slower and Go Seigen thought the move he played in the game was better.])
(;W[jf]C[If white jumps like this,]
;B[ke]C[Black will peep first, to make white heavy, then]
;W[je]
;B[hg]C[Lean against this group to build power.]
;W[gh]
;B[hh]
;W[gi]
;B[jh]LB[hb:C][ic:A][id:B][eh:D]TR[hd][jd][je][jf]C[And cap here. Black's attacking in good form and white's group is heavy. Black still has the sequence A-C in reserve, so he can connect under or live easily if he has to later. There's also a vital point at D now.]))
(;B[ie]C[Sometimes it's good to move out by leaning on the pincer stone, with a shoulder hit like this. However, in this case,]
;W[je]
;B[if]
;W[id]LB[ff:A]C[Black's shape becomes awkward after white plays like this and it's hard for black to find a good move now.]))
(;W[jc]C[White could also play here, but the 4th line pincer can move into the center more quickly and easily.]
;B[lc]
(;W[je]
;B[gg]C[The game might continue like this now.])
(;W[ff]C[If white still plays here,]
;B[he]LB[je:B][gf:A]C[Black would choose this shape now, which makes miai of A and B.]
(;W[je]
;B[gf]C[Black moves out with good shape.])
(;W[gf]
;B[je]C[Black can jump here and it's not easy for white.]))))
(;B[dg]C[This is also conceivable. ]
(;W[cd]C[If white attaches here, ]
;B[eg]C[Pushing up here is good.]
(;W[be]C[This hane is the right move for white.]
;B[ff]C[It's another game. ])
(;W[ff]C[White shouldn't play here, because]
;B[de]C[This is painful. ]
;W[ee]
;B[dd]
;W[ec]
;B[cc]
;W[cb]
;B[bd]
;W[bc]
;B[jc]C[The result it better for black. ]))
(;W[jd]C[White would pincer here, and it's similar to the game.])
(;W[fe]LB[de:A][ee:B]C[If white plays here, white will have bad shape after black exchanges A for B.])))
(;B[nq]C[This is more common.]
;W[qi]LB[qg:B][ql:A]C[Black can aim to invade at A later. For example, after approaching at B.])
(;B[pj]TR[od][qd][pj]C[Black might also consider pincering here, to make his stones in the top right corner work effectively.]
(;W[oq]LB[op:B][pp:A]C[It used to be that white would play here now, and this leads to fighting after black cuts with A or B.]
(;B[pp]
;W[op]
;B[oo]C[Black might cut and fight like this.]
;W[pn]
;B[no]
;W[or]
;B[pr]
;W[lq]
;B[nm]
;W[rq]C[White should create some aji in the corner first.]
;B[rr]
;W[sr]
;B[rs]LB[rm:A][rn:B]C[Now moves like A and B will be sente for white later and white can use them to help his out four stones on the right side.]
(;W[qj]C[White could settle his group like this.]
;B[qk]
;W[pk]
;B[qi]
;W[rj]
;B[ok]
;W[pl]
;B[pi]
;W[ri]
;B[rh]
;W[rk])
(;W[ok]C[White could also choose play here.]))
(;B[op]C[Black could play here instead, which is also severe. I normally tell my students to play like this.]
;W[pp]
;B[qr]
;W[nq]
;B[mp]))
(;W[np]C[I think it's better for white to press more gently like this.]
(;B[nq]C[If black attaches,]
;W[mp]C[White can just extend.]
;B[mq]
;W[lp]C[This result is satisfactory for white.])
(;B[nr]LB[np:A]C[If black plays here, then white A becomes a forcing move and white's happy.]))))