(;GM[1]FF[4]CA[UTF-8]AP[CGoban:3]ST[2]
RU[Japanese]SZ[19]KM[6.50]TM[3600]
GN[mathmo-GGroenen(B) IGS]PW[mathmo]PB[GGroenen]WR[4d]BR[5d]DT[2017-01-11]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: mathmo vs GGroenen]PC[IGS:  igs.joyjoy.net 6969]US[Brought to you by IGS PANDANET]C[
 mathmo 4d+: 88 board 3
 GGroenen 5d?: 88 have a nice game!
]RE[W+4.50]LT[]NB[33]NW[32]RR[Normal]
;B[pd]BL[3577]
;W[dd]WL[3598]C[
 mathmo 4d+: 88 have a good game
]
;B[qp]BL[3523]
;W[dq]WL[3554]C[weird to 3-4 for me, but I wanted to try something different from all my London open losses]
;B[oq]BL[3519]
(;W[ck]WL[3528]C[going back to a fuseki style I had over a year ago.

50.59 to white]
(;B[qj]BL[3453]C[a bit low, my expectation was that leela would say this is too small a moyo, especially it normally thinks both the 3-4 and the enclosure is slow, so that I should take the centre - you take low means I get more high.

52.08 to W]
;W[nc]WL[3524]
(;B[pf]BL[3441]
(;W[jd]WL[3491]
(;B[cf]BL[3385]
(;W[ce]WL[3470]
;B[df]BL[3381]
(;W[fd]WL[3463]C[51.79 to W]
;B[ci]BL[3379]
(;W[ek]WL[3379]C[unsurprsingly leela agrees]
(;B[ei]BL[3329]
(;W[jp]WL[3351]C[I was very happy to get this, but the main uncertainty is how to respond to C17]
(;B[lq]BL[3307]
(;W[ne]WL[3195]C[
 Uberdude 3d : and daniel said he doesn't like moyo games...

I guess I don't like the uncertainty, but after leelabot, I get more scared of losing centre control. Maybe that means I actually dislike moyos less than most.

50.83 here is down 2.2% compared to optimal
]
;B[do]BL[3214]
(;W[fp]WL[3139]C[50.62]
;B[cq]BL[3212]
;W[cr]WL[3135]
(;B[fo]BL[3211]C[
 afar 2d+: Maybe he doesn't like territory games either
 afar 2d+: I could sympathise
]
;W[go]WL[2973]
;B[ep]BL[3197]
(;W[eq]WL[2970]C[51.8 for W]
;B[gp]BL[3196]
;W[fq]WL[2968]
;B[fn]BL[3195]
(;W[ho]WL[2963]C[52.01]
(;B[cp]BL[3194]
;W[gk]WL[2940]C[as usual from my bot learning experience, especially when W so solid, there is no point play 2-2, as B C7 is very big to help the upper group, so this one point jump was obvious.

leela strongly approves

52.57]
(;B[br]BL[3116]C[Now it is very hard to choose where to attack B, as W has a lot of weaknesses to fix, and the G9 group isn't that solid if B comes out due to H7, so I thought I should fix it first despite the lack of territory (I was trying to fix my london open issue of not protecting my weak groups in the centre)]
(;W[gi]WL[2786]C[52.22]
(;B[gh]BL[3081]C[I was very happy to see such a crude move, as I know how little leela thinks of this sort of thing]
;W[hh]WL[2782]
;B[gg]BL[3079]
;W[fi]WL[2778]
;B[eh]BL[3078]
(;W[hg]WL[2767]C[53.37

a little crazy, though no.1 on the leela's neural net at 18%.

too small, -2% from optimal]
(;B[ge]BL[3077]C[a little crude]
;W[gd]WL[2737]
;B[hf]BL[3048]
;W[if]WL[2697]C[a forceful sealing in, just because I read the that top solidified and the centre seemed ok, but if he just reduced with expanding with P10 it would have been painful

even so, it is 53.88 to W]
;B[ie]BL[3047]
;W[je]WL[2690]
;B[ig]BL[3046]
;W[jf]WL[2689]
;B[hi]BL[3043]
;W[ih]WL[2688]
;B[ii]BL[3041]
;W[jh]WL[2686]
;B[ji]BL[3041]
(;W[kh]WL[2669]C[53.72]
(;B[ik]BL[3039]C[55.88]
;W[hl]WL[2647]
;B[il]BL[3038]
;W[hm]WL[2638]
(;B[lj]BL[3035]C[too slow]
(;W[nj]WL[2333]C[I also wasn't sure if this should be one to the right. leela prefers P10

I knew the proverb don't play too close to weak groups, but I was worried I didn't have a follow up on the right

55.56]
(;B[ko]BL[3015]
(;W[nl]WL[2221]C[55.33]
;B[ok]BL[2990]
;W[nk]WL[2186]
;B[jo]BL[2988]
(;W[nh]WL[2072]C[very slow - I got nervous 54.63]
;B[oi]BL[2978]
;W[ni]WL[2046]
;B[oc]BL[2976]
;W[nb]WL[1990]
;B[be]BL[2867]
;W[bd]WL[1987]
;B[bf]BL[2866]C[slow - J4 says leela

]
;W[bj]WL[1983]
(;B[bi]BL[2844]C[slow says leela]
;W[cn]WL[1970]C[leela approves]
;B[dr]BL[2842]
;W[er]WL[1964]
;B[cs]BL[2841]
(;W[jq]WL[1938]C[60.62

I only noticed the H3 issue later and the liberty shortage due to the H6 cut]
;B[cc]BL[2726]
;W[cd]WL[1926]
;B[bc]BL[2724]C[I thought for some time whether this was killable but decided it wasn't]
(;W[dc]WL[1601]
;B[ad]BL[2722]C[this is huge, but this timing is strange and I thought there must be issues around M4.

This sequence is worth 21 points, or 19.5 if I had spotted the C18 tesuji]
(;W[np]WL[1501]C[walking into N5 or O6, leela doesn't like it]
;B[mo]BL[2717]
;W[no]WL[1442]
(;B[mn]BL[2705]
;W[nq]WL[1321]
;B[or]BL[2683]
;W[nn]WL[1300]
(;B[mm]BL[2622]C[crude]
;W[nm]WL[1241]C[63.10]
;B[qm]BL[2607]
;W[ff]WL[1190]
;B[gf]BL[2595]
;W[qc]WL[1139]C[I wasn't too bothered to let B connect at O1 as M9 was still big, R17 etc and B cannot connect back comfortably with O1 due to M2

This move is worth over twenty points]
;B[qd]BL[2590]
;W[pc]WL[1051]
;B[rc]BL[2453]
;W[od]WL[1044]C[
 Uberdude 3d : m9 looks big for eye hassling
]
;B[cb]BL[2044]C[huge, maybe I should  have played here earlier to take over 15 points in reverse sente. B might even have responded so I could tenuki having already gained 1.5 points in sente.

Now B is threatening the terrifying monkey jump for around 7 points in sente]
;W[qo]WL[908]C[good timing as this only works before B connects at O2, this gains around 4.5 points in sente]
(;B[ro]BL[2034]
;W[pp]WL[905]
(;B[qq]BL[2028]
;W[lk]WL[847]
;B[kk]BL[2020]
;W[ll]WL[839]
;B[kl]BL[2017]C[M7 and O2 are both large, I wasn't sure, but in any case D18 is ridiculously big]
(;W[db]WL[662]C[the 7 point reverse sente]
;B[da]BL[2013]
;W[ea]WL[660]
;B[ca]BL[2013]
;W[eb]WL[644]
;B[rb]BL[2006]C[The follow up at Q/R 18 is 6 points]
(;W[nr]WL[559]C[seemed much bigger than M8, but only around 7 points at most


B should have ataried P15 at some point

67.35 for W]
;B[lm]BL[1989]C[about 3 in reverse sente, and necessary for life]
;W[po]WL[540]
;B[qn]BL[1905]
;W[of]WL[522]C[I didn't want to see the P15 pain so this seems very big]
;B[og]BL[1890]
(;W[oh]WL[514]
(;B[ph]BL[1881]
;W[ng]WL[511]
(;B[pg]BL[1880]
;W[pl]WL[509]
;B[ql]BL[1865]
;W[pk]WL[397]
(;B[pj]BL[1709]C[this is a mistake, losing 1 2/3 if W plays, gaining 1 if B plays.

But B did need to respond as there is a tesuji by cutting S6 though this is only worth about 5 points locally]
;W[fe]WL[302]C[3 point sente]
;B[eg]BL[1694]
;W[bo]WL[223]C[solving the aji locally of C7 or D8 B7 combo, or B6 directly so very big, maybe even 9 points]
;B[bp]BL[1633]C[block is a bit slow, but bigger than it looks - normally just 3 point reverse sente, but there remains aji in W's area and E1 is no longer W sente, so E1 is B's right, another 2+ points difference]
;W[dn]WL[98]C[a bit slow again, just playing solid out of confidence]
;B[co]BL[1632]
;W[bn]WL[96]C[W has a certain 8 points on the left now and B has no points about E5,6, so D6 wasn't small.]
;B[kq]BL[1611]C[crude, why not try K2? or L2]
;W[jr]WL[71]
;B[io]BL[1574]
;W[ip]WL[68]
;B[es]BL[1541]C[I kept counting but wasn't sure of victory if I gave up 4 points to avoid the ko, it felt like maybe I had less points than I previously counted, so I wasn't confident anymore. It seems by the final result that I was right to fight the ko.]
(;W[fs]WL[442]
(;B[ds]BL[1396]C[W is really lucky to be so solid, but after making sure of those extra 4 points to my count, I was sure of victory.]
;W[gq]WL[438]
(;B[pb]BL[1214]C[6 points]
;W[ob]WL[435]
;B[qb]BL[1211]
;W[rn]WL[407]C[good tesuji to play before P1, gains 1.5 points in sente]
;B[rp]BL[1198]
;W[os]WL[405]
;B[pr]BL[1191]
;W[in]WL[403]
;B[jn]BL[1186]
;W[im]WL[402]
;B[jm]BL[1185]
;W[oj]WL[400]C[huge, 3 2/3 in sente]
;B[hd]BL[1180]
;W[hc]WL[395]
(;B[pi]BL[1136]C[mistake - B lost several points here]
;W[kr]WL[362]C[too small 3 points with 3 point follow up is 4.5]
;B[fl]BL[1087]
;W[fk]WL[359]
;B[em]BL[1083]
;W[en]WL[314]
;B[eo]BL[1079]
;W[qk]WL[288]
;B[rk]BL[1077]
;W[pm]WL[288]C[6 points combo]
;B[dl]BL[1060]
;W[cl]WL[229]
;B[cj]BL[1058]
;W[aj]WL[173]
;B[ai]BL[874]
;W[bl]WL[170]
(;B[li]BL[826]
;W[lh]WL[152]
;B[lr]BL[824]
;W[dm]WL[138]
;B[el]BL[822]
;W[pn]WL[132]
;B[rm]BL[820]
;W[id]WL[122]
;B[he]BL[818]
;W[ki]WL[97]
;B[kj]BL[816]
;W[ej]WL[52]
;B[gn]BL[803]
;W[hn]WL[48]
;B[pe]BL[662]
;W[dj]WL[33]
;B[di]BL[654]
;W[oe]WL[554]
;B[hj]BL[650]
;W[gj]WL[520]
;B[ee]BL[648]
;W[ed]WL[518]
;B[oc]BL[619]
;W[ap]WL[488]
;B[aq]BL[617]
;W[ao]WL[487]
;B[oa]BL[616]
;W[na]WL[485]
;B[pa]BL[615]
;W[pq]WL[481]
;B[ps]BL[614]
;W[ns]WL[474]
;B[pc]BL[437]
;W[lp]WL[454]
;B[kp]BL[432]
;W[mp]WL[451]
;B[mi]BL[412]
;W[mh]WL[449]
;B[ks]BL[386]
;W[js]WL[447]
;B[ls]BL[385]
;W[lo]WL[442]
;B[ln]BL[383]
;W[qr]WL[436]
;B[rr]BL[382]
;W[op]WL[435]
;B[qs]BL[381]
;W[dk]WL[391]
;B[gl]BL[367]
;W[gm]WL[389]
;B[fm]BL[366]
;W[hk]WL[388]
;B[ml]BL[360]
;W[mk]WL[387]
;B[]BL[348]
;W[]WL[383]
;B[]BL[348]C[
 mathmo 4d+: mathmo dead @ P9
 mathmo 4d+: mathmo dead @ S6
 mathmo 4d+: mathmo dead @ G4
 mathmo 4d+: mathmo dead @ J13
 GGroenen done
]
;TW[fa][ga][ha][ia][ja][ka][la][ma][fb][gb][hb][ib][jb][kb][lb][mb][ec][fc][gc][ic][jc][kc][lc][mc][kd][ld][md][nd][ke][le][me][kf][lf][mf][nf][ig][jg][kg][lg][mg][fj][ak][bk][ok][al][ol][am][bm][cm][om][an][on][oo][gp][hp][hq][iq][fr][gr][hr][ir][gs][hs][is]TB[aa][ba][qa][ra][sa][ab][bb][sb][ac][qc][sc][rd][sd][ae][qe][re][se][af][qf][rf][sf][ag][bg][cg][dg][qg][rg][sg][ah][bh][ch][dh][qh][rh][sh][qi][ri][si][ij][jj][rj][sj][jk][sk][jl][rl][sl][km][sm][kn][rn][sn][so][sp][bq][rq][sq][ar][cr][qr][sr][as][bs][rs][ss]C[W+4.5

B lost at least 1 point with Q18 instead of P10, probably should have fought the G3 ko, then it would be very close. Otherwise it was W who lost more points in the endgame out of feeling good about the game. Not C18 and then D18 instead of S18. And not invading into R5]OS[Ectoplasm][Elehat][Fede][Gelmer][JIG][KaC][Madsen99][Mukti][PaxDeorum][Rubenmc][Uberdude][afar][gigago][herminator][litcir][mumps][tkykjin][vor09])
(;B[dm]C[not sente and surprisingly doesn't seem to be more than 1 point]))
(;B[pm]C[this is 6 points!]
;W[pi]C[only 1 2/3 in sente]))
(;B[]
;W[qb]C[I think this is way to defend!]
;B[qa]
;W[ob]
;B[pa]C[this is still 5 points, W gets sente and gains 3 points compared to game]))
(;B[gq]
;W[gr]
;B[dp]C[ko of value 70+ points ish]
;W[ds]
;B[id]C[so this is worth over 70 points as well]
;W[cr]
;B[ic]
;W[bs]C[who is leading?]
;B[gb]C[it seems like B is still winning though B needs another move, as the top is bigger]
;W[jc]
;B[ib]
;W[jb]
;B[oj]
;W[qb]
;B[qa]
;W[ob]
;B[fj]
;W[gj]
;B[ej]
;W[fk]
;B[cj]
;W[bk]
;B[pa]
;W[rn]
;B[rp]
;W[qk]
;B[rk]
;W[os]
;B[pr]
;W[in]
;B[jn]
;W[im]
;B[jm]
;W[ki]
;B[kj]
;W[hj]
;B[ij]
;W[kr]
;B[pb]
;W[oc]
;B[lr]
;W[ia]
;B[ha]
;W[ja]
;B[oa]
;W[na]
;B[ra]
;W[li]
;B[mj]
;W[mi]
;B[pe]
;W[oe]
;B[dk]
;W[dl]
;B[dj]
;W[aj]
;B[ai]
;W[ls]
;B[ms]
;W[ks]
;B[mr]
;W[ns]
;B[ps]
;W[op]
;B[ol]
;W[pm]
;B[pn]
;W[om]
;B[on]
;W[oo]
;B[rl]
;W[lp]
;B[kp]
;W[mp]
;B[hk]
;W[gl]
;B[lo]
;W[pq]
;B[qr]
;W[fh]
;B[fg]TW[ka][la][ma][kb][lb][mb][kc][lc][mc][kd][ld][md][nd][ke][le][me][kf][lf][mf][nf][ig][jg][kg][lg][mg][lh][mh][ak][al][bl][cl][el][fl][am][bm][cm][dm][em][fm][gm][an][en][fn][gn][hn][ao][co][do][eo][fo][ap][bp][cp][dp][ep][gp][hp][aq][bq][cq][gq][hq][iq][ar][br][dr][fr][hr][ir][as][cs][es][gs][hs][is][js]TB[aa][ba][ea][fa][ga][sa][ab][bb][db][eb][fb][hb][sb][ac][dc][ec][fc][gc][hc][sc][bd][cd][dd][ed][fd][gd][hd][rd][sd][ae][ce][de][ee][fe][he][qe][re][se][af][ef][ff][qf][rf][sf][ag][bg][cg][dg][qg][rg][sg][ah][bh][ch][dh][qh][rh][sh][di][pi][qi][ri][si][jj][rj][sj][jk][sk][jl][sl][km][rm][sm][kn][ln][rn][sn][so][sp][rq][sq][rr][sr][qs][rs][ss]C[Very interesting, this is B+0.5]))
(;W[gr]))
(;B[pi]C[should be here])
(;B[]
;W[rn]
;B[rp]
;W[qk]
;B[rk]
;W[rj]
;B[rl]
;W[qi]))
(;B[pi]))
(;B[pe]C[note that this loses 2 points if sente]
;W[ob]C[this is correct]
;B[ph]
;W[ng]
;B[pi]C[this is actually the same as the game for points]))
(;W[pg]C[leela says cut, this is an exchange, perhaps it is right]
;B[pe]
;W[oh]
;B[oe]
;W[nd]
;B[nf]
;W[qg]C[but requires some thought]))
(;W[pg]C[this looks like a nice tesuji, but it doesn't seem to work as P15 is sente]))
(;W[rb]C[actually bigger - about 18 points, but I was just playing solidly by this point]
;B[hq]C[there is still this issue though]
(;W[hp]
;B[jr]
;W[ir]
;B[kr]
;W[hr])
(;W[ir]C[actually this seems to work!]
(;B[gq]
(;W[fr]
;B[gr])
(;W[gr]
(;B[fr]
;W[hr])
(;B[hr]
;W[fr]
;B[gs]
;W[iq])))
(;B[gr]C[but there is this ko issue]
;W[gq]
(;B[es])
(;B[hp]C[actually this is the same issue]
;W[hr]
;B[io]
;W[ip]
;B[hn]
;W[iq]
;B[gn])))))
(;B[po]
;W[qq]
;B[qn]
;W[op]
;B[pq]
;W[nr]C[now that this is sente, the centre group is in trouble]
;B[qr]
;W[os]
;B[rq]
;W[lk]
;B[kk]
;W[ll]
;B[kl]
;W[lm]
;B[io]
;W[ip]
;B[in]
;W[hn]
;B[im]
;W[km]
;B[jm]
;W[kn]
;B[hj]
;W[jj]
;B[jk]
;W[kj]))
(;B[po]
;W[pp])
(;B[pp]
;W[po]
;B[mr]
(;W[qk])
(;W[pg])))
(;B[ll]C[says leela]))
(;B[nn]
;W[on]
;B[om]))
(;W[cb]C[I wouldn't have noticed this move if leela hadn't pointed it out]
;B[bb]C[This exchange gains W 1.5 points in sente as B can no longer get everthing in sente. If W plays, W gets an extra 8, if B plays, B gets an extra 7, so the next move here is worth 15 points, probably not yet the biggest point.

B doesn't have to block, but it is worth a bit more than 14.5 points (11.5 if B responds to A18, 17.5+ if B doesn't), it is very important for both sides to continue playing here])
(;W[qo]C[61.68]
;B[po]
;W[qn]
;B[rp]
;W[qh]C[leelabot says]))
(;W[ad]
;B[ec]
;W[ed]
(;B[dc])
(;B[ac]
;W[eb]
;B[db])
(;B[fb]
;W[dc]
;B[db]
;W[fc]
;B[eb]
;W[bb]
;B[cb]
;W[ac]
;B[hc])))
(;W[iq]C[61.57]))
(;B[cc])
(;B[cm]C[(W has weaknesses too)]))
(;W[iq]C[56.42]))
(;W[pk]C[57.02 to W.

I didn't think for a bit about this, but was nervous of my own cutting points])
(;W[qo]C[56.10]))
(;B[ml]C[leela expects]
;W[nl]
;B[nm]
;W[ol]
;B[om]
;W[pl]
;B[pi]
;W[mk]
;B[lk]
;W[nh])
(;B[oc]C[or leela expects]
;W[nb]
;B[nf]
;W[nh]
;B[oe]
;W[nd]
;B[mi]
;W[ni]
;B[ml]
;W[nl]
;B[nm]
;W[ol]
;B[mm]
;W[qi]
;B[pi]
;W[pj]))
(;W[oj]C[56.10]))
(;B[mk]))
(;B[hl]C[53.81])
(;B[ki]C[53.65]))
(;W[ik]C[key point 54.01])
(;W[ki]C[53.91]))
(;B[hf]C[this is tricky]
;W[gf]
;B[ge]
;W[ff]
;B[fe]
;W[he]
;B[ee]
;W[fc]
;B[gd]
;W[if]
;B[gc]
;W[cc]C[50.85 for W]))
(;W[oj]C[55.33])
(;W[rp]C[54.17]))
(;B[gf]
;W[hd]
;B[oc]
;W[nb]
;B[nf]C[leela says]))
(;W[ff]C[53.19])
(;W[gh]C[53.96]))
(;B[cm]C[I thought this was big and leela agrees]))
(;B[cm]C[leela preference]))
(;W[gn]C[leela still prefers active moves 52.38]))
(;W[fn]C[54.25 should atari]))
(;B[cp]
;W[br]C[50.58]
;B[fo]
;W[go]
;B[fn]
;W[gk]
;B[ep]
;W[fq]))
(;W[co]C[50.91])
(;W[cp]C[51.04]))
(;W[do]C[52.48

its quite poor that I missed how important this is.])
(;W[gj]C[53.05])
(;W[fp]C[52.99]))
(;B[do]
;W[cp]C[leela says this, 51.26 to W])
(;B[cc]
(;W[be]
;B[ec]
;W[ed]
;B[fc]
;W[gc]
;B[gb]
;W[dc]
;B[db]
;W[hc]
;B[hb]
;W[ic]
;B[bb]
(;W[gj]C[53.61]
;B[gg])
(;W[gg]C[54.08]
;B[gi]
;W[dh]
;B[di]
;W[gk]
;B[ii]
;W[fh]
;B[fi]
;W[ik]))
(;W[dc]
(;B[be]
;W[do]C[52.92 to W])
(;B[do]
;W[co]C[52.09 for W]))))
(;W[fp]C[leela prefers by 1%])
(;W[gj]C[leela thinks this isn't bad, I guess it is harder to play C17]
;B[ip]C[though leela fears this]
;W[fp]
(;B[gh])
(;B[in])))
(;B[gp]C[I thought this might happen.

Leela prefers]
;W[jp]))
(;W[ej]C[this was the main other option I was considering, but I wanted solid])
(;W[jp]
;B[do]C[I didn't see where to play here though]))
(;W[ee]C[leela says this is slightly better!]))
(;W[ch]C[considered this momentarily. I am a little surprised leela prefers this by 0.4 though as the shape is weird with C9]))
(;B[gp]C[leela says])
(;B[ip]C[leela number 1 preference]))
(;W[jp]C[It's fascinating how leela prefers this by 0.4])
(;W[kd]C[leela prefers by 0.4]))
(;B[lc]C[I was a little nervous whether I would copy leelabot's style here or not, ]
(;W[ne]C[leela loves this for W]
;B[pf]
;W[kd])
(;W[qc]
;B[qd]
;W[pc]
;B[od]
;W[nd]C[or leelabot likes this stuff])))
(;B[fc]C[expecting this]))
(;W[qf]C[normal, and leela prefers it by 0.2% (It likes active play)]))