(;GM[1]FF[4]CA[UTF-8]AP[CGoban:3]ST[2] RU[Japanese]SZ[19]KM[6.50]TM[3600] GN[JRDK-mathmo(B) IGS]PW[JRDK]PB[mathmo]WR[4d]BR[5d]DT[2017-10-04]CP[ Copyright (c) PANDANET Inc. 2017 Permission to reproduce this game is given, provided proper credit is given. No warrantee, implied or explicit, is understood. Use of this game is an understanding and agreement of this notice. ]EV[Internet Go Server game: JRDK vs mathmo]PC[IGS: igs.joyjoy.net 6969]US[Brought to you by IGS PANDANET]C[ JRDK 4d?: 48 hi, have a nice game mathmo 5d : 48 have a good game board 1 super intense game. The opening was extremely good for me, but my middlegame was full of questionable move, but also had very slack moves from him. Missed that I had a life and death issue in the centre, and had to fight a 100 point ko for life, all while running out of time at 7 seconds/move After he threatened a 36 point group of mine, it looked I might be in trouble. But after a long think, I made the ko even heavier where it could be a 120 point ko, to threaten his eyespace. This gained me several more points, and I only had to give up a 25 point group while saving my 36 point group. I gained about 3 points in the simple endgame, and I won by 16.5. Actually, though I played well, this was extremely complicated with many variations I hadn't seen clearly at the time. My opponent made one huge (very innocent looking) mistake, but otherwise, it looks like in many of the variations I had a way to win, though some go crazy with first line connections further than the monkey jump that are easy to miss.]RE[B+16.50]LT[]NB[33]NW[32]RR[Normal] ;B[pd]BL[3578] ;W[dc]WL[3575] ;B[pq]BL[3564]C[random opening] ;W[dq]WL[3570] ;B[de]BL[3558] (;W[cg]WL[3563]C[I know alphago hates this, but I'm not particularly impressed by black continuing locally I know Mi Yuting played a sort of tenuki recently in this shape] (;B[co]BL[3527]C[not sure high or low] (;W[dl]WL[3521]C[human style, but I doubt it's alphago's move to pincer still, I'm not too clear how to deal with it] ;B[ep]BL[3493] (;W[dp]WL[3501] ;B[do]BL[3446] ;W[eo]WL[3498] (;B[eq]BL[3410] ;W[er]WL[3465] ;B[fr]BL[3407] ;W[dr]WL[3463] ;B[en]BL[3405] ;W[fo]WL[3455] ;B[gp]BL[3402]C[this is a very violent shape - B profits first on the lower side and makes W's corner small, and the D5 stones are stubborn to die] ;W[bp]WL[3372] ;B[bo]BL[3159]C[the white corner is locally ko after B B3] (;W[cp]WL[3364]C[but this is bad - too slack still, given that I've never seen this before, it took me a long time to understand how to play differently] (;B[dm]BL[2943]C[the fight is critical as W's corner has had to live small, B will be satisfied just to settle on both sides, as the F5 cutting stones will look silly.] (;W[gn]WL[3277]C[the fight is already very difficult for W (C4 is like a pass move) this shape is terrible] (;B[ip]BL[2562]C[the shape is thin, and perhaps this is a bit much given that alphago always defends at J3 the push and cut at F4 is basically defended against, but the H3 peep is severe. or the aji after W L3] ;W[el]WL[3252] ;B[fm]BL[2559] ;W[cm]WL[3232]C[this feels like a mistake, but I can't see a good move for W anyway (other than tenuki)] (;B[cl]BL[2054]C[despite the long think, this cut was premature __ but it is very complicated, I haven't found a satisfying conclusion that another move was better (see variations), and perhaps this was actually best.] ;W[em]WL[3124] ;B[dn]BL[2050] ;W[fn]WL[3117] ;B[bm]BL[2025] ;W[ck]WL[3114] ;B[bl]BL[2023] ;W[dj]WL[3086] ;B[cc]BL[2001] (;W[cb]WL[3064]C[so this makes sense] (;B[gd]BL[1948]C[so this move didn't take too long despite somewhat making it up on the spot. I was satisfied with the result, but for some reason, I've never seen it before in a professional game. And later even C17 for C18 became a good exchange once I got sente to live.] (;W[ce]WL[2906] ;B[kc]BL[1877] (;W[qo]WL[2797] (;B[np]BL[1848]C[to protect the aji] ;W[qk]WL[2681] ;B[qf]BL[1817] ;W[qh]WL[2324] ;B[cd]BL[1780]C[this works particularly well before W has invaded at J17. So B lives in sente maybe it was worth W protecting at D16 instead of R12] ;W[dd]WL[2317] ;B[be]BL[1777] ;W[cf]WL[2314]C[even here, C17 for C18 is working well as W should play D18 in response to C17 to stop the corner from being completely alive] (;B[bb]BL[1777] ;W[db]WL[2310] ;B[ba]BL[1776]C[the result is like invading a 4-4 with large knights enclosure, and living with sente while giving up very little on the outside, surely B is not unhappy despite living so small as W spent so many moves on this area.] ;W[bf]WL[2300] ;B[ae]BL[1764]C[this gets two more points than B16 while leaving one more ko threat clearly W playing here is too early - it could be used for ko threats] (;W[ee]WL[2233]C[this leaves J17 wide open] ;B[in]BL[1700] (;W[gm]WL[2201]C[??? surely too solid this aims at the huge G2 clamp as F4 is not sente now though, which concerns the B9 move as well so it makes sense. But in that case B gains a lot in sente, and perhaps W should have played J6 earlier? ] (;B[qm]BL[1581]C[I thought that this might be the sort of move that I should play but normally don't. But the result was no good. I didn't really think of my top side as large potential. That was probably a mistake. This is a fighting move that aims to fix B's shape both above and below, but either I didn't know how to use it (true) or it isn't a good move.] ;W[oo]WL[2100] (;B[no]BL[1361] ;W[on]WL[2038] (;B[pl]BL[1359]C[this feels like either overplay or I didn't use it well] ;W[nn]WL[2004] (;B[pk]BL[1350] ;W[qj]WL[2000] (;B[oi]BL[1312]C[this feels dumb, too direct once again, I thought I was attacking, but that is unreasonable] ;W[oh]WL[1951] (;B[pi]BL[1257]C[too direct, playing as if W is much weaker than B here] ;W[ph]WL[1934] ;B[mi]BL[1252]C[up to here, it seems like I played a series of small mistakes, enough to make the game much more even.] ;W[nh]WL[1886] (;B[ln]BL[1201] ;W[qq]WL[1812] ;B[qr]BL[1200] ;W[pp]WL[1809] ;B[oq]BL[1200] ;W[rq]WL[1791] ;B[mk]BL[1181] ;W[rg]WL[1769] (;B[md]BL[1059] ;W[mn]WL[1745] ;B[lo]BL[1058] ;W[lg]WL[1717] ;B[ji]BL[1017]C[B is annoyingly thin here, and at the top too, so I can't see any profit from R7 up to here W didn't have much territory there anyway.] ;W[pc]WL[1588]C[painful though I'm not sure how B should protect against this aji if it were B's move: R17 or R16 perhaps?] ;B[qc]BL[969] ;W[oc]WL[1489] ;B[od]BL[929] ;W[qb]WL[1478] ;B[nc]BL[920]C[it feels like B haemorrhages territory again] ;W[rc]WL[1470] ;B[qd]BL[919] (;W[ra]WL[1337]C[but this move ...] ;B[gk]BL[756]C[make W feel the pain] ;W[fl]WL[1302] ;B[gb]BL[702]C[huge - B is still a bit thin and not completely alive otherwise] (;W[lm]WL[1297] (;B[rn]BL[548]C[I was getting worried about my eyespace] ;W[ml]WL[1188]C[ Uberdude 3d : i fee like m6 shoulda been r4 ] ;B[lk]BL[501] ;W[rr]WL[1158]C[huge loss of points given S6] (;B[rl]BL[472]C[dumb in combination with S6 should have played here without playing S6] ;W[ne]WL[1133]C[questionable to give up playing O18 I thought] ;B[nb]BL[469] ;W[pa]WL[1127] (;B[oa]BL[434] ;W[pb]WL[1124] (;B[me]BL[423]C[ very slightly questionable] ;W[nf]WL[1118] ;B[nd]BL[422] ;W[km]WL[1037]C[ Uberdude 3d : n15 looks like wrong kikashi Uberdude 3d : p14 could be better for detroying w eyes Uberdude 3d : b has to get his profit somehow ] (;B[jn]BL[302]C[ SimbaRules 2k+: Maybe K6 a little slow, could tenuki. __ you might be right, I'm not clear. The aji feels bad, but I guess W cutting off J6 is not worth spending a move ] ;W[nl]WL[1002] ;B[nk]BL[243] (;W[kf]WL[957]C[ Uberdude 3d : so generous ] ;B[gf]BL[584] ;W[jd]WL[933] ;B[kd]BL[569]C[this move actually shows some agitation with the game, the last 50 moves felt pretty bad for me. it may well be best, but it does leave some bad aji, which I ignored out of emotion. Most likely W doesn't have anything as W is thin too, but B is annoyingly thin with insufficient eyespace (as we see later!) and the weak G9 stone. ] ;W[je]WL[930] ;B[jc]BL[544] ;W[bk]WL[915]C[this was so incredibly big, that I should have played here 40 moves before. I had thought about it, but obviously, my thinking was insufficiently clear. I thought A5 hurt the W corner sufficiently much I didn't need to worry so much (I would rather jump to B10 rather than B9 for endgame) but really, with the clamp at G2, the white corner is fine - I lose many points if I want to keep its eyespace low. Furthermore, if I had played B9, there would be aji at B11 if B has a stone at E12. And also the W corner has seki aji with B2, so using up the aji with A5 is not very profitable. ] ;B[ao]BL[422]C[very painful for B] ;W[gr]WL[875]C[around 10 points] ;B[gq]BL[418] ;W[fs]WL[872] ;B[hr]BL[416] ;W[fq]WL[871] ;B[fp]BL[416] ;W[fr]WL[739]C[much bigger than it looks as it leaves the big endgame of H2 and some aji inside with H5 and J2 or N5] ;B[pr]BL[341]C[trying to fix some aji, this is also at least 11 points + W eyespace] ;W[jl]WL[668]C[highly questionable to play dame as this W group is already alive, just exchange O11 or P8 in sente if W is still worried] ;B[ik]BL[237]C[W helps B fix down to 16 moves in 4 minutes, pretty tight.] ;W[jk]WL[609] ;B[jj]BL[236] ;W[hs]WL[588] ;B[ir]BL[235] ;W[is]WL[586] ;B[js]BL[234] ;W[gs]WL[585]C[this looks normal, but this leaves a potentially critical set of extra ko threats (at least 2) in the corner with the shortage of liberties] (;B[kr]BL[195]C[questionable, the ko threat at K2 is pretty big. I read afterwards that if W plays K2 and K3 though, B can still play L1 to kill W! But the aji is extremely bad. I thought this might be better for aji/points, but there seems no reason to think so.] ;W[il]WL[555]C[???? W has two eyes already] (;B[fg]BL[164] ;W[af]WL[534] (;B[bd]BL[162]C[2 seconds thought, leads to this embarrassing move, leaving an extra ko threat. I noticed as soon as I played it.] (;W[eh]WL[530] ;B[eg]BL[161] ;W[dg]WL[525] ;B[jg]BL[154] ;W[sd]WL[500]C[B is relieved, and should win comfortably now] (;B[fh]BL[144]C[embarrassing endgame again It's as if I took gote to play a 1-2 point move compared to F10] ;W[hk]WL[494] ;B[gj]BL[106] ;W[ij]WL[485] ;B[ii]BL[95] ;W[ic]WL[476]C[overplay] (;B[ib]BL[61]C[ SimbaRules 2k+: not convinced lol My loss is small - I lost around 2 points locally and W even got sente] ;W[hc]WL[466]C[ SimbaRules 2k+: time pressure I guess SimbaRules 2k+: I think J16 works ] ;B[gc]BL[45] ;W[he]WL[455] ;B[ap]BL[34]C[sente to unconditional kill with B3! its a nice go problem you can try, this is why G1 might lead to problems this is 4 1/3 points in sente, so it is huge in comparison E11 is around 7 points ] ;W[br]WL[443] ;B[hb]BL[26]C[trying to get an extra half point or so compared to G15 but ...] ;W[ni]WL[411] ;B[nj]BL[22] ;W[ol]WL[408] ;B[pj]BL[597] ;W[ok]WL[403] ;B[oj]BL[596] ;W[li]WL[398] ;B[rk]BL[524]C[losing my monkey jump sadly but it was inevitable] ;W[rj]WL[395] ;B[ro]BL[512] ;W[kj]WL[374]C[this felt like W is just losing more points, but actually it seems to be good endgame to get T9 in sente] ;B[rp]BL[509]C[ SimbaRules 2k+: r8? it is painful for white to lose these 3 points in sente ] ;W[qp]WL[372]C[ SimbaRules 2k+: murder was gotten away with... ] (;B[sm]BL[480]C[I had no idea my centre was going to get into trouble - the eyespace looked plenty, and I was still thinking about using N12 for endgame, when really, this is an important timing to peep at L12] ;W[ge]WL[363] (;B[hf]BL[469] ;W[if]WL[358] ;B[ig]BL[467] ;W[kh]WL[329]C[ SimbaRules 2k+: e11 big SimbaRules 2k+: 7 seconds/stone, yikes I was horrified to find myself locally dead ] (;B[fe]BL[115]C[17 moves in 2 minutes] ;W[fd]WL[317]C[ aitkensam 4d?: uk players really like using their time it seems :) ] ;B[ef]BL[111] ;W[ff]WL[314]C[ SimbaRules 2k+: I used like 14 minutes for the whole game lol ] ;B[ei]BL[89]C[ SimbaRules 2k+: yay, e11 ] (;W[fj]WL[237]C[correct] ;B[fe]BL[87] ;W[ql]WL[220] ;B[pm]BL[70] ;W[ff]WL[218] ;B[fi]BL[63] (;W[hh]WL[166]C[but even though it's necessary for W to respond, this is a HUGE mistake as it gave up several points and B still had more threats ] ;B[fe]BL[62] ;W[qn]WL[152]C[big mistake, T9 is better by 3 points and one ko threat] ;B[sk]BL[61] ;W[ff]WL[149] ;B[hg]BL[58]C[ SimbaRules 2k+: hmm r6 instead of t9 for white SimbaRules 2k+: bad endgame ] ;W[gh]WL[139] ;B[fe]BL[57] ;W[sp]WL[135] ;B[so]BL[53] ;W[ff]WL[132] ;B[di]BL[49]C[so painful on W as W has run out of huge ko threats] ;W[dh]WL[126] ;B[fe]BL[48] ;W[df]WL[106] ;B[jf]BL[46] ;W[ie]WL[81] ;B[ke]BL[45] ;W[ff]WL[80] ;B[ej]BL[23]C[the order of my ko threats is precise If I'd played this first, D11 wouldn't be a threat as this shorts my own liberties and loses me the eye at H12.] ;W[fk]WL[46] ;B[fe]BL[22] ;W[al]WL[37]C[ SimbaRules 2k+: I guess you have to admire his confidence with time management and the endgame, figuring that he can just ultra-blitz out a near perfect global sequence having read it up front at the start SimbaRules 2k+: still would make me far too nervous using most of my time on the first stone, just in case something happened ] (;B[fc]BL[167]C[thinking for 7 minutes+ 24 stones in 167 seconds is 7 seconds/move again this is a lovely move that wins me an extra 10 points or so over the variation I was reading, but actually only 6 points. I only had so much fighting spirit because I wasn't convinced that I would win if I ended the ko. Probably I would have, but it would be close] ;W[ff]WL[11] ;B[cj]BL[165]C[this is a huge ko threat as W has just spend a move to kill at A8] (;W[dk]WL[528]C[of course] ;B[fe]BL[164] (;W[ac]WL[513]C[ SimbaRules 2k+: maybe e16 is enough even? ] ;B[ed]BL[118]C[ SimbaRules 2k+: seems so SimbaRules 2k+: a7 bigger than the corner SimbaRules 2k+: so w needs to fix d15 too ] ;W[bc]WL[466] ;B[am]BL[116] ;W[ci]WL[452] ;B[ak]BL[112]C[dumb move as H10 is slightly bigger (locally B continues at B11)] ;W[aj]WL[447]C[W should have H10] ;B[hj]BL[108]C[ SimbaRules 2k+: b+10-15ish I think, should be clean from here, all settled. B is very happy to get the last big 8 point move when the rest of the endgame is small (4-5 points) and easy (and easy to play fast) My analysis is that W loses 3 more points from here.] ;W[sj]WL[379]C[looks like 3 point reverse sente, but really around a pure 4 points, so not quite biggest] ;B[ar]BL[99]C[4 2/3 points] ;W[bq]WL[372] ;B[aq]BL[98] ;W[kn]WL[347] ;B[ko]BL[97] ;W[ec]WL[339] ;B[de]BL[96] ;W[mo]WL[337] ;B[mp]BL[95] ;W[ee]WL[336] ;B[fd]BL[84] ;W[ho]WL[323]C[3 points (looks like subtract 1 for liberty shortage, but not actually) with a 2 point follow up. So this is actually bigger than T10 and A2 though it doesn't look it. B should have played H6.] ;B[io]BL[77] ;W[qs]WL[309] ;B[ps]BL[72] ;W[rs]WL[307] ;B[rf]BL[70] ;W[sf]WL[303] ;B[mh]BL[66] ;W[mg]WL[301] ;B[qi]BL[54] ;W[ri]WL[299] ;B[op]BL[48]C[reverse sente 1.5] ;W[ob]WL[263] ;B[na]BL[47] ;W[lf]WL[261]C[2 points] ;B[pn]BL[41] ;W[po]WL[259] ;B[re]BL[37] ;W[se]WL[257] ;B[of]BL[32]C[3.5 points ish, this was hard for me to count (I thought I was 2 points initially), but there is a cut at L9] ;W[lh]WL[220] ;B[mj]BL[30] ;W[le]WL[218] ;B[sq]BL[17]C[about 2 7/12] ;W[ld]WL[198] ;B[lc]BL[599] ;W[kk]WL[188]C[big, but should be at G8] (;B[hp]BL[588] ;W[jh]WL[598] ;B[ih]BL[587] (;W[gl]WL[589]C[mistake this looks big, but it is only about 1 1/3] ;B[ik]BL[586] ;W[hl]WL[583] ;B[om]BL[581] ;W[nm]WL[572] ;B[sr]BL[565]C[looks like mistake only 1 5/6 I thought I was gaining points in sente, but actually I could lose a tiny bit in sente. Furthermore, there are moves bigger than 2 points. But as my opponent responded, this became 2/3 in sente.] ;W[ll]WL[565] ;B[eb]BL[557]C[this is 2 1/3 (it concerns D15)] ;W[ea]WL[561] ;B[fb]BL[557] (;W[de]WL[557]C[slightly incorrect perhaps] ;B[fa]BL[555] ;W[da]WL[555] ;B[oe]BL[550] ;W[pg]WL[552] ;B[pf]BL[547] ;W[bs]WL[530] ;B[ij]BL[526] ;W[hn]WL[520] ;B[im]BL[524] ;W[al]WL[518] ;B[sb]BL[511] ;W[rb]WL[507] ;B[ak]BL[510] ;W[ss]WL[494] ;B[sp]BL[508] ;W[al]WL[492] ;B[bi]BL[503] ;W[bj]WL[473] ;B[ak]BL[503] ;W[jr]WL[455] ;B[jq]BL[501] ;W[al]WL[453] ;B[sh]BL[501] ;W[rh]WL[431] ;B[ak]BL[501] ;W[ks]WL[427] ;B[ls]BL[500] ;W[al]WL[424] ;B[ch]BL[500] ;W[bh]WL[422] ;B[ak]BL[500] ;W[bn]WL[408] ;B[cn]BL[499] ;W[al]WL[407] ;B[og]BL[598] ;W[ng]WL[404] ;B[ak]BL[598] ;W[lj]WL[593] ;B[rm]BL[595] ;W[al]WL[585] ;B[ab]BL[595] ;W[ad]WL[581] ;B[ak]BL[595] ;W[rd]WL[559] ;B[al]BL[594] ;W[jm]WL[546] ;B[hm]BL[591] ;W[go]WL[544] ;B[kg]BL[589] ;W[ki]WL[542] ;B[ek]BL[588] ;W[mf]WL[539] ;B[as]BL[587] ;W[qg]WL[538] ;B[]BL[572] ;W[]WL[534] ;B[]BL[572]C[ mathmo 5d : mathmo dead @ J15 mathmo 5d : mathmo dead @ B18 mathmo 5d : mathmo dead @ G12 JRDK 4d?: JRDK dead @ B11 mathmo 5d : mathmo dead @ H17 mathmo 5d : mathmo dead @ T18 JRDK 4d?: JRDK dead @ T12 mathmo 5d : mathmo dead @ B6 JRDK done ] ;TW[aa][ba][ca][qa][sa][ab][bb][sb][cc][sc][bd][cd][ae][be][ag][bg][sg][ah][ch][sh][ai][bi][si][cj][kl][fm][mm][cq][cr][cs][ds][es]TB[ga][ha][ia][ja][ka][la][ma][jb][kb][lb][mb][hc][ic][mc][hd][id][jd][ge][he][ie][je][pe][qe][ff][if][gg][gh][hh][gi][hi][sl][cm][an][bn][sn][jo][jp][kp][lp][hq][iq][kq][lq][mq][nq][jr][lr][mr][nr][or][ks][ms][ns][os]OS[ABQNM][Denmark5][FrenchDude][Husrev][Laudrup][MartinSWE][Mercian][Podiceps][SimbaRules][Uberdude][afar][aitkensam][bunzu][erdem][gasana][hogweed][hwspan][tiaxSWE][tvh58]) (;W[oe]C[better as this is the last endgame as there is a ko at A8 and B1 and J9, so it could gain a point if W can hold the D15 ko. But I'm not sure W can hold onto the D15, So it's probably the same or worse ])) (;W[hl]C[locally this is better by 1/3])) (;B[hl]C[this loses 1/3 of a point if I can't win the A8 ko, so I didn't bother, but I do actually have several threats at T12 and B1. If I win the A8 ko this gains only 2 points anyway, so it's not really worth taking gote]) (;B[sr]C[this is bigger])) (;W[cn]C[so tempting to kill the whole group of over 36 points but and this is the best technique to kill, leaving just 2 ko threats after an hour of analysis of complications ... I think this is 3 points better for W than the game !!!!] ;B[ed]C[I spend an extra move locally to win the ko, but it is totally worth it as W owes a move at D15, but also at C11 (the whole W group is dead otherwise)] ;W[ci] ;B[de]C[and the key is that my calculation was that with a dead group, I could win this ko, and hence gain around 4 points compared to the game. If on the other hand I lost the ko, I would lose around 4 points compared to the game. (I'm not sure about these numbers at all as they depend on the size of ko threats and subsequent endgame) Right now, I'm not sure if I could win the ko or not.] ;W[ac]C[this extra ko threat might prove critical] ;B[ab] ;W[ee] ;B[am] ;W[bn] ;B[de] ;W[bc] ;B[ad] ;W[ee] ;B[an] ;W[cn] ;B[de] ;W[bc] ;B[ac] ;W[ee]C[it's important not to use the B11 ko threat yet] ;B[bq]C[this is a ko threat (you can read the unconditional kill)] ;W[cq] ;B[de] ;W[mo]C[this is a huge threat too] ;B[mp] ;W[ee] ;B[ar]C[this is an absolute threat because B hasn't used up the massive B11 ko threat] (;W[bs] ;B[de] ;W[jr] ;B[jq] ;W[ee] ;B[ds] ;W[cs] ;B[de] ;W[hq] ;B[hp] ;W[ee] ;B[bi] ;W[ch] ;B[de] ;W[ks] ;B[ls] ;W[ee] ;B[as] ;W[aq] ;B[de] ;W[kn] ;B[ee]C[this is a reasonable result to expect I think W captures a 11 point larger group, B captures an extra 17 points compared to the game, but W pushes an extra 5 points into the lower side, H3 loses 2 points, but W gest sente for the huge move at H10. Overall, this is probably about 3 points better for W than the game.]) (;W[de] ;B[bs] ;W[aq] ;B[bi] ;W[ch] ;B[bq] ;W[hq] ;B[cr] ;W[iq] ;B[jr] ;W[hp] ;B[jq] ;W[ho]C[B wins by miles, something like B+30]))) (;W[fe] ;B[dk]C[this is about 20 points better for me than the game!])) (;B[ff] ;W[am] ;B[ci] ;W[bh] (;B[fc]C[I didn't notice that this is much bigger] ;W[bi] ;B[hj]) (;B[bi]C[this is a close game] ;W[fc]C[despite having a ko to kill W, the ko is actually very heavy for me too, and W has several local ko threats, and B doesn't] ;B[aj] ;W[ak] ;B[ah] ;W[ag] ;B[ch]C[It felt to me that I had so few threats, that it was as if there wasn't a ko here. I think my analysis was correct]))) (;W[fe]C[even though B is locally dead ...] ;B[hj] ;W[jh] ;B[ih] ;W[hh] ;B[gh]C[H13 is actually better, but this still seems to live anyway] ;W[hg] ;B[gg] ;W[hi] ;B[di] ;W[dh] ;B[ci]C[this gives W a serious problem] (;W[df] (;B[bj]C[to stop the weird A10 connection] ;W[ak] ;B[hl] ;W[ik] ;B[gl] ;W[hm] ;B[im] ;W[bi] ;B[bh] ;W[ai] ;B[ah] ;W[aj] ;B[ch]C[W dies]) (;B[hl]C[this would be easy to miss and embarrassing for B] ;W[ik] ;B[gl] ;W[hm] ;B[im] ;W[aj]C[this is a connection!])) (;W[aj]C[this doesn't help] ;B[df] ;W[bh] ;B[gi] ;W[hh] ;B[bi] ;W[ai] ;B[hl] ;W[ik] ;B[gl] ;W[hm] ;B[im]C[this is seki or something])) (;W[hj]C[I expected this. hmm, I think this is so much better if B responds I think B is dead (or a ko where W has a zillion local threats] ;B[hi] ;W[df] ;B[hl] ;W[ik] ;B[gl] ;W[hm] ;B[di] ;W[dh] ;B[fe] ;W[ke] ;B[ci] ;W[aj]C[this is a weird capture race but W has a zillion ko threats] ;B[im] ;W[ff])) (;W[df]C[at the time I thought that W could seriously consider capture as I am low on liberties, but I didn't have the time to read that I am fine] (;B[fe]C[I thought I still have to fight the ko ]) (;B[ej] ;W[jh] ;B[ih] ;W[hj] ;B[hi] ;W[fk] ;B[fj] ;W[gg] ;B[hg] ;W[gh] ;B[dh] ;W[gi] ;B[di])) (;W[fe]C[B lives] ;B[ej] ;W[hj] ;B[hi] ;W[jh] ;B[ih])) (;B[ei] ;W[fj]C[this is dead!]) (;B[gh] ;W[fi]C[this is dead!]) (;B[ej] ;W[ei] ;B[fi] ;W[jh] ;B[ih] ;W[hj] ;B[hi] ;W[gh])) (;B[kh]C[once again this is critical])) (;B[kh]C[critically important])) (;B[id] ;W[hd] ;B[ie] ;W[he] ;B[if] ;W[hc] ;B[gc])) (;B[hj]C[even this is better]) (;B[fj]C[this seems best My estimate is that B is leading by around 10 points.])) (;W[ih]C[very important area, but] ;B[df]C[A14 shorts W liberties critically] ;W[ef] ;B[dg] ;W[eg] ;B[dh] (;W[eh] ;B[ch]) (;W[ch] ;B[eh] ;W[fh] ;B[fi] ;W[gh] ;B[ei] ;W[gi]C[this actually seems to work for W though]))) (;B[bc]C[this leaves only 2 threats])) (;B[eg]C[I found it hard to choose this move is thinner, but it might be fine])) (;B[jr]C[better])) (;W[ge]C[this remains huge])) (;B[hi] ;W[jo] (;B[io] ;W[mo] ;B[mp] ;W[lp] ;B[lq] ;W[kp] ;B[kq]) (;B[jp] ;W[mo] ;B[lp] ;W[mp] ;B[mq] ;W[lq]C[my shape is quite awkward]))) (;B[kf]C[there is an idea here] ;W[le]) (;B[le]) (;B[of]C[I'm not sure about uberdude's suggestion, it looks slightly risky, and I don't see the value either]) (;B[mf]C[more of a shape, but I'm not sure it helps])) (;B[ob]C[I spent time wondering about the (almost zero) difference with this])) (;B[km]C[this is actually a huge move - B is thin otherwise])) (;B[rl]C[this is still probably better] ;W[rn]C[I feared this, but I don't think I should, my moves in the game were too small] ;B[pn])) (;W[ge]C[I expected this sort of move for the longest time, aiming at G11 or even H11])) (;W[nb]C[I thought it was surely here] ;B[mb] ;W[ob]C[with the weakness at O15 and M11, there doesn't seem sufficient hope to attack the R12 group when it can live with on move on the right])) (;B[nd]C[I wasn't sure if this was better]) (;B[oc]C[this would feel slack given that W's eyespace is still important, and there is still the J17 invasion and S15 into the corner])) (;B[qp]C[of course this is better] ;W[lo]C[this aji is nothing if W is not alive])) (;B[nh]C[once again, this is probably better] ;W[og] ;B[pi] ;W[qg] ;B[pf] ;W[ni] ;B[nj] ;W[mi] ;B[mj])) (;B[mk]C[just here is fine] ;W[oi]C[this gave me bad shape, but it should be fine still] ;B[qp]C[this is too severe?]) (;B[pi]C[this is also an idea] (;W[ph] ;B[oh]C[at the least this hane is better shape than the game]) (;W[pj] ;B[oj] ;W[ok]C[this is a problem, but B can still sacrifice the stones] ;B[oi] ;W[nk] ;B[qi] ;W[ri] ;B[rh] ;W[rj] ;B[rg] ;W[rm]C[there is still the aji of S8]))) (;B[nl]C[I thought too much about attacking W, and that was unreasonable])) (;B[pj]C[this might be sufficient as W has already made a bad exchange at P5] ;W[qj] ;B[qp] ;W[ro] ;B[ph])) (;B[pl]C[is this better? I considered it for a while.] ;W[no]C[I was scared of being pushed, as still losing the 3-3, but is this really a fear?] ;B[mp] ;W[mo] ;B[lp] ;W[qq] (;B[qr] ;W[pp] ;B[oq] ;W[ml]) (;B[pp] ;W[qn] ;B[qp] ;W[rp] ;B[rq]C[maybe B is still happy?]))) (;B[bk]C[this was good timing to exchange this. I had considered it, but didn't realise its importance given that W defended solidly at G7 rather than at H8 Like this I can aim at seki with B2 If I don't play here W B9 is horrifically painful as there is no longer aji at B11, and the W corner is much healthier with the G2 clamp. But I didn't think so clearly at the time.]) (;B[qp] ;W[po] ;B[on] ;W[pm] ;B[pj] ;W[qj] ;B[ph]C[this should be my dream result, I was scared of the bad aji around L3 still, but it feels better than the game where my top side got reduced so small])) (;W[im]) (;W[hl])) (;W[fe]C[very tempting to my eyes, but W plays solidly the whole gaame])) (;AE[cb]AW[db]C[this seems better for W])) (;B[qk]C[though alphago rarely pincers and hence I don't either, another game I would be tempted by this despite the aji])) (;W[hq]C[this is very bad aji])) (;W[ee]C[it seems that W should have cut, maybe C13 is better placed for this] ;B[ef] ;W[ed]C[even here may be sufficient])) (;B[cd]C[but similarly, it didn't feel like it made sense to fight here - W is too solid for B to create a weak group here.]) (;B[dg]C[I was tempted by this, but hadn't read a clear variation])) (;W[gc]C[I actually envisaged this] ;B[dd] ;W[db] ;B[cf] ;W[dg] ;B[cb]C[W is overconcentrated everywhere, and B has a good result given that this is W's side]) (;W[db]C[an idea I just had to prevent B G16] ;B[dg] (;W[df] ;B[cf] ;W[ef] ;B[bg] ;W[ch] ;B[bh] ;W[ci] ;B[ce] ;W[ee] ;B[cb]C[how is this result?]) (;W[dh] ;B[cf]C[good for B?]))) (;B[hm]C[I had been reading that silly things like this didn't work]) (;B[fn]C[but in hindsight, just defending properly and crudely was much better the idea is to make W heavier in a fight where W's stone already make no sense hmm but I didn't yet consider W tenuki] (;W[fd]C[my feeling is that W should tenuki]) (;W[go] ;B[hp]C[the problem is that B is threatening C8 and H7 and W has a difficult job making the E5 cut look meaningful if B lives on both sides.] (;W[in] ;B[cl]C[this would look stupid for W]) (;W[cj]C[it took me a while to figure this one out as the cut at C8 will be met by B7 and though B has the B8 tesuji to capture the G5 stones, B gives up too much.] ;B[bm]C[but simple as it is, it took me a long time to find the correct tesuji] ;W[bl] ;B[hm]C[now this works] ;W[gm] ;B[gl] ;W[fl] ;B[hn] ;W[dn] ;B[cn] ;W[em] ;B[dn]) (;W[ck]C[W has this powerful move to save both sides.] ;B[gm] ;W[in] ;B[im]C[but W's shape remains bad. This will be a fight though, and I'm not completely confident this will be better for B than the game. (though it feels that way) B will have more potential to do things on the left with the D15 stone, but W will threaten to play moves like L4 and hurt the G4 group.]))) (;B[gm]C[actually, maybe this is even better I seriously considered this in the game, but eventually went for simplicity. But this looks clearly better to keep fighting and W tenuki is a little harder as B H6 is a clean kill.] ;W[hn] ;B[fn] ;W[go] ;B[hp] (;W[jn] ;B[cl]C[W's shape is uglier and W still can't play L4 (the K5 cut), so this does look even better than my suggested variation.]) (;W[cj]C[but maybe this remains an issue.] ;B[bm] ;W[bl] ;B[im] ;W[in] ;B[jn] ;W[jo] (;B[kn] ;W[io] ;B[lp]C[is this possible? looks like it barely doesn't work for B. with the H7 cut]) (;B[hm]C[if so, this is still a good shape for B] ;W[kn] ;B[jm] ;W[io] ;B[kq] ;W[jq] ;B[jp] ;W[kp] ;B[lp] ;W[ko] ;B[jr] ;W[lq] ;B[kr] ;W[mp]C[my feeling remains that B is happy here, but compared to the game, perhaps my result was still fine.])))) (;B[el]C[this is so automatic, that it took me a long time to realise that the cut at F4 was too severe (as C4 makes the corner alive) hmm, but at the time I didn't dream of not defending the E4 stones, but B seems fine just sacrificing them.] ;W[fp] (;B[fq]C[it wouldn't seem right to connect here] ;W[gq]C[G6 defends against the ladder, so B crawling on the 2nd line has no purpose, as C4 makes the corner alive]) (;B[dk] ;W[ce] ;B[pn] ;W[gq]C[but this sort of play seems possible]))) (;W[hp]C[someone suggested this, and it seems surprising, but I haven't found a great way to play against it. So it feels like a good move in this position] ;B[go] ;W[fn] (;B[fm] ;W[gn] ;B[ho] ;W[fp]C[B feels in some trouble]) (;B[el] ;W[fp] ;B[gq] ;W[fq] ;B[dk] ;W[jp]))) (;B[el]C[there are many variations where this is good] ;W[dn]C[but this cut is huge (now C4 has a purpose)] ;B[fn]C[the problem was that I had seen variations like this played by leelabot, but probably in different situations, so I spent time considering it, when in reality it seems to make no sense here.] ;W[go] ;B[dm] ;W[cn] ;B[bn] ;W[cm] ;B[gn] ;W[ho] ;B[cl] ;W[bm] ;B[bl] ;W[am] ;B[dk] ;W[ao]C[this feels good for W, but I can't remember in what circumstances I saw leelabot play this sort of thing]) (;B[gn] ;W[fn] ;B[fm] ;W[em] ;B[go] ;W[dn] ;B[fp] ;W[en]C[this has been played professionally, but I didn't dare give up so much territory] ;B[fl] ;W[dj])) (;W[dn]C[in early 2017, the Chinese national team said this was the result of their analysis, but that this was veyr complicated] ;B[dm]C[something like this] ;W[em] ;B[fn] ;W[go] ;B[cn] ;W[gq] ;B[cl] ;W[qo]) (;W[go] ;B[hp] ;W[ho] ;B[ip] ;W[io] ;B[jp] ;W[jo] ;B[kp] ;W[em]C[this is an idea] ;B[bq] ;W[br] ;B[ar]C[the corner is ko though] ;W[cq] ;B[ap] ;W[aq]C[but B doesn't seem to have a good ko threat])) (;B[dm] ;W[fo] ;B[em] ;W[cl]C[this remains the issue - W has stones to help in the fight])) (;W[eq] ;B[fp] ;W[gq]C[alphago would presumably do this. I can't be sure of my follow up.] (;B[cc]C[probably aim to take sente for R5 is the key in this position. Moving this stone might be counter productive in fact] ;W[gc] ;B[dd] ;W[db] ;B[be] ;W[di] ;B[cb] ;W[jc] ;B[qo]C[this would be one of the ideal variations for B]) (;B[qo] ;W[fd]C[there is an obvious weakness at D10, but with some reading, this does feel like a good shape for W. the high and far placement of D8 makes moving D15 out very awkward] ;B[dj]C[this isn't really a problem] (;W[ef]C[even after something like this, I don't see a good follow up]) (;W[fl]C[or this if locally above C17 is shape, is there really any good move to work with D10?] ;B[cc] ;W[cb] ;B[fe] ;W[ge] ;B[cd]C[maybe B is ok here though]) (;W[en])))) (;W[qo]C[I'd expect alphago to play here])) (;B[cc]C[standard joseki, but I've long not been convinced (by either side's moves])) (;W[ce]C[see alphago self play 6 for example] ;B[cf] ;W[cd] ;B[df] ;W[fc] ;B[co] ;W[fp] ;B[dl] ;W[qo] ;B[pl] ;W[qc]C[sort of idea]))