(;GM[1]FF[4]CA[UTF-8]AP[WebGoBoard:0.10.11]ST[2]RU[japanese]SZ[19]KM[6.5]HA[0]TM[604800]OT[86400 fischer]RE[B+R]DT[2012-12-24]CP[online-go.com]PB[Uberdude]BR[7d]PW[vitality]WR[4d]PC[OGS: http://online-go.com/game/368477]C[This is a game I played on OGS back in 2012-13 when I was much the same strength as I am now (sigh). I put a lot of thought into these games; vitality was one of the toughest opponents there and took some games of me in title matches. The opening was practically identical to a 2006 pro game (Niu Yutian 6p vs Zhou Heyang 9p http://ps.waltheri.net/database/game/47759/), but I was taking the role of the loser so tried to find his mistakes to find a good place to diverge. Comments with "Uberdude:" prefix were made by me after the game in mid 2013. Others are Leela Zero review now in 2018, but on weak computer with just a few hundred playouts per move.];B[pd];W[dc];B[qp];W[dq];B[do];W[co];B[dp];W[cp];B[eq];W[cn];B[dn];W[dm];B[cq];W[dr];B[bq];W[cr]C[LZ prefers outward turn but with so few playouts I'm not going to pay much attention on a joseki that's had so much human study (not that that means we aren't wrong as AlphaGo has some different ideas, and AG Teach on this exact board also prefers outward turn!)];B[cm];W[cl];B[br]C[LZ agrees with this whilst AG famously prefers the b7 "joseki mistake"];W[er];B[bm];W[bn];B[bl];W[bp];B[fq];W[dl];B[ap]C[LZ doesn't like, c9 was 51% this is 43%];W[ao];B[fr];W[bk];B[cs]C[AGT thinks this joseki result is 43% for black (starts at 47). I also understand pros decided it was locally slightly better for white (the e4 aji is worse than b10 so white thicker), but on this board the direction ok for black];W[oq]C[5 pro games (from mid 2000s) with prior position, all approached here. LZ agrees, white 58%](;B[ce]C[4 of them approached here. I can't remember my reasoning, mutual prevention of shimari? Could it also be a ladder maker? But the n4 pincer ladder already works. LZ likes.];W[cg]C[2 pros pincered. As I expected LZ doesn't like, too gote, corner soft. LZ would kick or tenuki.];B[mp]C[No pro games with whole board, but 17 with lower side.];W[mq]C[standard for pros, LZ wants to p5 then o5 at 1k.];B[lq](;W[lp];B[mo]C[Downside of c15 c13 exchange is now I don't have ladder, but upside is white would probably not pincer later. LZ agrees best.];W[kq];B[lr];W[mr];B[kr]C[LZ agrees last 4 moves];W[po]C[both pro games with this lower side press, LZ prefers p5 jump a little...](;B[qo]C[both pros blocked, but LZ says should push (and cut if white blocks)! -3%](;W[pn]C[one pro game Zho Heyang played here like vitality did. LZ agrees.](;B[qm](;W[kp]C[LZ says 7% mistake];B[jq];W[jp];B[iq];W[pm];B[ql];W[pl]C[LZ agrees to here](;B[pk]C[But -5 here, wants to jump];W[ok];B[oj]C[LZ agrees double hane](;W[nk]C[Alight minus, LZ wants to make the exchanges now](;B[lm]C[I don't think I'd have found this move myself, and LZ thinks it loses 4%!];W[qn]C[LZ agrees with following kikashi];B[rn];W[ro];B[rm];W[qq];B[rp];W[pj];B[qk];W[jm]C[and this jump](;B[pp]C[Uberdude: Seems I successfully improved on the pro's result =) So I pushed here to make defects in white's shape before hitting the vital point. LZ agrees this is a good move though k8 or o10 might be 1% better.];W[op](;B[nm]C[-1%](;W[oi]C[ok](;B[oo]C[LZ says a 1.5% mistake because W can sacrifice](;W[ol]C[Now we are truly in amateur land, the biggest mistake so far, -17%. Now black is winning.];B[pq];W[no]C[Uberdude: You should q2 first to get 1 eye on edge and stop my territory being so big. LZ agrees q2 a must, -10% until it sees o4 tesuji which I missed, then -18%. ](;B[or]C[I was happy to get this in sente I didn't look for something even better! -10%](;W[nq]C[-12%];B[mn]C[correct.](;W[on]C[-3](;B[jn]C[LZ would rather get out without sacrificing. -3];W[kn];B[km];W[in];B[jl];W[im];B[il];W[gm];B[lj]C[LZ agrees with my haengma](;W[nj];B[hl](;W[fo];B[lh]C[LZ agree](;W[pf]C[Uberdude: If you had q2 maybe you could treat that ponnuki as a strong group and approach on top side, this way I am happy to develop top. ];B[nd](;W[nf]C[Uberdude: Give you ended up doing r16 and r10, this looks slack territorially unless you can attack my centre group. I was actually expecting that as you like to attack thickness, but I had k11 answer prepared for l11 m11 l9 m9 l8 LZ does consider bt prefers more active];B[dg]C[LZ agrees with timing. Black 90% now](;W[ch](;B[gc];W[ed];B[jc];W[qd]C[Uberdude: I think you need c16 or d14 honte ];B[qc];W[qe];B[rc];W[qj];B[bc]C[Uberdude: Now I don't see how white can make enough points. ];W[cf]C[Uberdude: It attacks the shape, but helped your strong group and weakened your weak group. Expected d14 c14 e14 c18 ];B[de];W[be];B[bd];W[bf];B[cc]C[Uberdude: Maybe I should e15 first ];W[ee];B[df];W[gd];B[fc];W[fg];B[db];W[eb];B[ca];W[ec];B[hd];W[he];B[id];W[hg];B[hi];W[gi];B[gj];W[fi];B[ih];W[me]C[Uberdude: I expected desperate m17 as looking for place to resign, as black wins big this way. LZ agrees m17 best.];B[md];W[fj];B[gk];W[ep];B[en];W[em];B[fn];W[gn];B[es];W[rd];B[rj];W[ri];B[rk];W[ke];B[ld];W[sc];B[sb];W[sd];B[rb];W[pe];B[od];W[hb];B[hc];W[ib];B[jb];W[kg];B[kh];W[si];B[dh];W[di];B[ml])(;B[ed]C[LZ would press to settle rather than trade like I did.]))(;W[ki];B[li];W[kk];B[lk];W[kl];B[ji]))(;W[ii]))(;W[en];B[ep];W[ii]C[LZ wants to stop me raisig a flag.]))(;W[ep];B[fo];W[fn];B[eo];W[go];B[lh]))(;W[mi]C[probably he didn't like giving r10 sente but LZ prefers.]))(;B[jl])(;B[kl]))(;W[nn]))(;W[on];B[nq];W[np];B[nr];W[mn];B[ms];W[lo];B[mq];W[nn]C[better for white to give some territory to strong groups and make strength for himself.]))(;B[np];W[on];B[nq];W[or];B[nr]))(;W[rq](;B[ol];W[pq];B[so]C[White 64%.])(;B[pq]C[slightly better for black to push here.])))(;B[pq];W[pr];B[qr];W[rq];B[so];W[mn];B[lo];W[nn];B[ln];W[or];B[jn];W[in];B[im]))(;W[no]C[defend in sente first];B[lo];W[oi]))(;B[jl];W[im];B[kl]C[LZ wants to get out first]))(;B[nm]C[Niu Yutian 6p played here, but I thought this was a mistake to allow white perfect q4 shape. LZ agrees, -4%. Yay, go me!];W[pp];B[oi]C[(remember top left was a 4-4 but otherwise board the same)];W[kk];B[jn];W[kn];B[km];W[in];B[jl];W[im];B[kl];W[lk];B[il];W[gm];B[ij];W[kh];B[ih];W[jf];B[hf]C[I didn't like his group running like this.]))(;B[pj]C[LZ says this simple defence better, sac 2 stones, Black 46%](;W[ln];B[nc])(;W[ip]C[this is better for w, threaten to gobble whole];B[ln]C[LZ might try saving them, tricky vars])))(;W[qn];B[rn];W[ro];B[rm];W[qq];B[rp];W[pj];B[qk];W[nk];B[nj]))(;B[mm];W[qn];B[rn];W[ro];B[rp];W[rm];B[so];W[qk];B[rl];W[rk];B[sm]C[getting wrapped around feels bad to me];W[oj]C[but probably LZ says ok because b gets sente again];B[jm]C[the 4 W stones aren't captured but LZ wouldn't try to save them.]))(;W[lo]C[interesting idea];B[mn];W[ln];B[mm];W[jq];B[jr];W[ir];B[hr]))(;B[pq]C[LZ 2% pref to go for corner eyes instead of jump which allows white leaning.]))(;W[kp]C[one pro game Luo Xihe connected here (top left was white 4-4, but LZ analysis with this board, let's hope no ladder differences!) LZ -3%];B[jq];W[jp];B[iq];W[pn](;B[qn]C[Then Paek Hongsuk pushed here to get sente, maybe he thought white made l4 heavier so worth taking local loss in corner to chase them];W[pm](;B[kn](;W[ip]C[LZ would rather hit the shape point](;B[nn](;W[pp];B[qq];W[hp];B[hq];W[hm]C[agrees better than j7](;B[nl];W[qm];B[pk](;W[pr];B[qr];W[np];B[nc];W[jm];B[qi]C[fairly peaceful (top left was 4-4)])(;W[rm];B[rn];W[qj]C[LZ]))(;B[nk]C[LZ thinks black can be more adventurous])(;B[ol]))(;W[hq];B[hr];W[gr]C[LZ wants to punish the tenuki of j4 first ]))(;B[mm]C[also good])(;B[nm]))(;W[mm];B[mn];W[lm];B[km];W[qm];B[pq];W[pp];B[qr]))(;B[mm]C[prefers this, at 47% because..]))(;B[rm]C[8% better to not push from behind but get ahead says LZ. Seems legit.])))(;B[pp];W[kp];B[jq];W[no];B[mn];W[nn];B[mm];W[qo];B[or]C[nice tesuji](;W[np];B[pq];W[ol];B[jp];W[ro];B[rp]C[white 51%])(;W[pq];B[nq];W[op];B[pr];W[qr];B[np];W[qq]C[white 39%])))(;W[mr]C[LZ likes this too, looks passive to human?](;B[lp];W[np])(;B[dg])))(;B[mp]C[Lee Changho and Cho Hunhyun played this lower side shape first in 2003 (top left was 4-4)];W[mq];B[lq];W[lp];B[nq];W[mr];B[mo];W[nr];B[kp]C[this ladder already works, but overconcentrated anyway?];W[po];B[qo];W[pn];B[qn];W[pm];B[lo];W[qr];B[ql]C[Actually Cho and Lee played this]))