Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

Jouni
Posts: 3281
Joined: Wed Mar 08, 2006 8:15 pm

Re: Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Post by Jouni »

FINALLY they stopped whole tournament :!: .
Jouni
jp
Posts: 1470
Joined: Mon Apr 23, 2018 7:54 am

Re: Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Post by jp »

kiroje wrote: Thu Mar 26, 2020 3:18 pm Xiphos 0.6 no popcount

After Nxe5
<snip>
20.58: -0.73/41 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxg2 15.Kxg2 a5 16.c3 Ra6 17.a4 Rg6+ 18.Kh2 Bd6 19.d4 Bxe5+ 20.dxe5 Qf5 21.Bc2 Qh5 22.Bf4 Bxf3 23.Bxg6 hxg6 24.Qg1 bxa4 25.Rxa4 Qf5 26.Qg3 Bb7 27.Rxa5 Qb1 28.Qg1 Qxb2 29.Rc5 Qb6 30.Be3 Ba8 31.Qd1 Qb7 32.f3 Re8 33.Kg2 Qa6 34.Bd4 Bc6 35.Kh2 Ra8 36.Kg3 Qb7 37.Be3
<snip>
And before Nxe5, does Xiphos at depths 40+ avoid playing it?
kiroje
Posts: 60
Joined: Wed Jul 25, 2012 10:12 am
Location: Copenhagen, Denmark
Full name: Kim Jensen

Re: Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Post by kiroje »

Ethereal 12.00

After Nxe5

Code: Select all

12.20 12...Nxe5 (depth 41, 1/48), 3,5G nodes (4,8M nps)

0.53: -0.01/31 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Be3 Nd5 17.Bxd5 Bxd5 18.Rc1 Bb7 19.Nh4 Be7 20.Nf3 Bd6
1.33: +0.01/32 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rae8 15.Ne4 Bxe4 16.dxe4 Qxd1 17.Rxd1 Bc5 18.Bf4 Nxe4 19.Kh2 Bxf2 20.Bxc7 Rc8 21.Ba5 Bg3+ 22.Kg1 Bf2+ 23.Kh1 Bg3 24.Kg1
2.35: +0.11/33 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rae8 15.Ne4 Nxe4 16.dxe4 Qxd1 17.Rxd1 Bxe4 18.Be3 c5 19.Bd5 Bxc2 20.Rdc1 Bg6 21.Bxc5 Bxc5 22.Rxc5 Be4 23.b4 Bxd5 24.Rxd5 h6 25.Rd6 Ra8 26.a4 Rfd8 27.Rxd8+ Rxd8 28.axb5 axb5 29.Ra5 Rd5 30.Kf1 Kh7 31.Ra8 Kg6 32.Ke2 Kf6
4.30: +0.01/34 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Rd5 Qg6+ 19.Kf1 Bf6 20.Be3 Bxd5 21.Bxd5 Qh5 22.Be4 Qxh3+ 23.Kg1 Qg4+ 24.Kf1 Bxb2 25.Rb1 Bf6 26.Bxc5 Qh3+ 27.Kg1 Qg4+ 28.Kf1
4.42: -0.01/35 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Rd5 Qg6+ 19.Kf1 Bf6 20.Be3 Bxd5 21.Bxd5 Qh5 22.Be4 Qxh3+ 23.Kg1 Qg4+ 24.Kf1 Bxb2 25.Rb1 Rxe4 26.dxe4 Qh3+ 27.Kg1 Qg4+ 28.Kf1
6.02: -0.01/36 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Bf6 17.Rxc5 Rfe8 18.Be3 Qxh3 19.Bd5 Qg4+ 20.Kf1 Qh3+ 21.Kg1
8.08: +0.01/37 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Rd5 Qg6+ 19.Kf1 Bf6 20.Be3 Bxd5 21.Bxd5 Qf5 22.Be4 Qxh3+ 23.Kg1 Qd7 24.Bxc5 Qg4+ 25.Kf1 Qh3+ 26.Kg1 Qg4+
9.30: -0.01/38 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Bf6 18.Be3 g6 19.Rxc5 Qxh3 20.Bd5 Qg4+ 21.Kf1 Qh3+ 22.Kg1
10.40: -0.01/39 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Rd5 Qg6+ 19.Kf1 Bf6 20.Be3 Bxd5 21.Bxd5 Bxb2 22.Rb1 Qf5 23.Be4 Qxh3+ 24.Kg1 Qg4+ 25.Kf1 Bf6 26.Bxc5 Qh3+ 27.Kg1 Qg4+ 28.Kf1
12.20: -0.01/40 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Rd5 Qg6+ 19.Kf1 Bf6 20.Be3 Bxd5 21.Bxd5 Bxb2 22.Rb1 Qf5 23.Be4 Qxh3+ 24.Kg1 Qg4+ 25.Kf1 Bf6 26.Bxc5 Qh3+ 27.Kg1 Qg4+ 28.Kf1
“Modern chess is too much concerned with things like pawn structure. Forget it, checkmate ends the game.” – Nigel Short
kiroje
Posts: 60
Joined: Wed Jul 25, 2012 10:12 am
Location: Copenhagen, Denmark
Full name: Kim Jensen

Re: Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Post by kiroje »

jp wrote: Thu Mar 26, 2020 3:24 pm
kiroje wrote: Thu Mar 26, 2020 3:18 pm Xiphos 0.6 no popcount

After Nxe5
<snip>
20.58: -0.73/41 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxg2 15.Kxg2 a5 16.c3 Ra6 17.a4 Rg6+ 18.Kh2 Bd6 19.d4 Bxe5+ 20.dxe5 Qf5 21.Bc2 Qh5 22.Bf4 Bxf3 23.Bxg6 hxg6 24.Qg1 bxa4 25.Rxa4 Qf5 26.Qg3 Bb7 27.Rxa5 Qb1 28.Qg1 Qxb2 29.Rc5 Qb6 30.Be3 Ba8 31.Qd1 Qb7 32.f3 Re8 33.Kg2 Qa6 34.Bd4 Bc6 35.Kh2 Ra8 36.Kg3 Qb7 37.Be3
<snip>
And before Nxe5, does Xiphos at depths 40+ avoid playing it?
Yes it wants to play Ne4 instead of Nxe5(at least for the time i ran it)
“Modern chess is too much concerned with things like pawn structure. Forget it, checkmate ends the game.” – Nigel Short
jp
Posts: 1470
Joined: Mon Apr 23, 2018 7:54 am

Re: Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Post by jp »

kiroje wrote: Thu Mar 26, 2020 3:48 pm
jp wrote: Thu Mar 26, 2020 3:24 pm And before Nxe5, does Xiphos at depths 40+ avoid playing it?
Yes it wants to play Ne4 instead of Nxe5 (at least for the time i ran it)
Komodo 13.2 avoids Nxe5 up to depth 20 (looking at a4, Nf1,...), but chooses Nxe5 at around depths 21-24, before switching again at depth 25:

depth=25
+0.14 12.Nf1 Nb6 13.Nxe5 Nxe5 14.Rxe5 Bf6 15.Re1 a5 16.c3 a4 17.Bc2 a3 18.Ne3 axb2 19.Bxb2 Qc6 20.Bb3 Na4 21.Bxa4 Rxa4 22.d4 Qd7 23.Qb3 Bg5 24.c4 Bxe3 25.Qxe3 Qc6 26.f3 Qxc4
depth=26
+0.08 12.Nf1 Nb6 13.Nxe5 Nxe5 14.Rxe5 Bf6 15.Re1 Rae8 16.Rxe8 Rxe8 17.Ne3 a5 18.a3 a4 19.Ba2 Nd5 20.Nxd5 Bxd5 21.Bxd5 Qxd5 22.c3 Be5 23.Be3 Rd8 24.Rc1 c5 25.d4 cxd4 26.cxd4 Bxd4 27.Qxd4
depth=27
0.00 12.Ne4 Kh8 13.a4 f6 14.d4 exd4 15.Nxd4 Rad8 16.axb5 Nxd4 17.Qxd4 Qxb5 18.Ba4 Qb6 19.Qxb6 cxb6 20.Bd2 Nc7 21.Bb3 Nb5 22.Bc4 Bxe4 23.Rxe4 Rxd2 24.Bxb5 axb5 25.Rxe7 Rxc2 26.Raa7 Rxb2


So it looks like Komodo 13.2 passes this test.

kiroje wrote: Thu Mar 26, 2020 3:37 pm Ethereal 12.00

After Nxe5

4.30: +0.01/34 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Rd5 Qg6+ 19.Kf1 Bf6 20.Be3 Bxd5 21.Bxd5 Qh5 22.Be4 Qxh3+ 23.Kg1 Qg4+ 24.Kf1 Bxb2 25.Rb1 Bf6 26.Bxc5 Qh3+ 27.Kg1 Qg4+ 28.Kf1
...
12.20: -0.01/40 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Rd5 Qg6+ 19.Kf1 Bf6 20.Be3 Bxd5 21.Bxd5 Bxb2 22.Rb1 Qf5 23.Be4 Qxh3+ 24.Kg1 Qg4+ 25.Kf1 Bf6 26.Bxc5 Qh3+ 27.Kg1 Qg4+ 28.Kf1
Again that 14 ... Nxh3+ is a red herring for engines.
jp
Posts: 1470
Joined: Mon Apr 23, 2018 7:54 am

Re: Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Post by jp »

jp wrote: Thu Mar 26, 2020 7:27 pm Again that 14 ... Nxh3+ is a red herring for engines.
[d]r4rk1/1bpqbppp/p7/1p2R3/8/1B1P1N1n/PPP2PP1/R1BQ2K1 w - - 0 15

SF10 thinks 14... Nxh3+ leads to a draw:

depth=35
0.00 15.gxh3 Bf6 16.Rc5 Qxh3 17.Bd5 Qg4+ 18.Kf1 Qh3+


Komodo 13.2:

depth=24
0.00 15.gxh3 c5 16.Nd2 c4 17.Qg4 Qxg4+ 18.hxg4 Bd6 19.Re1 cxb3 20.Nxb3 f5 21.g5 Rac8 22.Nd4 Rfe8 23.Rxe8+ Rxe8 24.Be3 f4 25.Bd2 Bc5 26.Nb3 Bd6 27.Nd4
depth=25
-0.32 15.gxh3 c5 16.Bf4 Bd6 17.Ng5 Bxe5 18.Qh5 Qf5 19.Bxe5 c4 20.dxc4 h6 21.f4 hxg5 22.cxb5 axb5 23.Rf1 Be4 24.Qxg5 Qxg5+ 25.fxg5 Rad8 26.Re1 Bf5 27.h4 Rfe8 28.h5 Be6 29.Bc3 Bxb3 30.Rxe8+ Rxe8
depth=26
-0.19 15.gxh3 c5 16.Bf4 Bd6 17.Ng5 Bxe5 18.Qh5 Qf5 19.Bxe5 c4 20.dxc4 h6 21.f4 Rae8 22.Re1 hxg5 23.cxb5 axb5 24.Qxg5 Rxe5 25.Rxe5 Qxh3 26.Bd5 Bxd5 27.Rxd5 Qe3+ 28.Kh2 Qe2+ 29.Kh3 Qf3+ 30.Kh4 Qf2+ 31.Qg3 Qxc2 32.Rxb5 Re8 33.Kh3 Qh7+ 34.Kg4 Qg6+ 35.Rg5
depth=27
-0.33 15.gxh3 c5 16.Bf4 Bd6 17.Ng5 Bxe5 18.Qh5 Qf5 19.Bxe5 c4 20.dxc4 h6 21.f4 Rae8 22.Re1 hxg5 23.cxb5 axb5 24.Qxg5 Qxg5+ 25.fxg5 Kh7 26.Kf2 f6 27.gxf6 gxf6 28.Bd4 Rxe1 29.Kxe1 Kg6 30.c3 Rh8 31.Bc2+ Kf7
depth=28
0.00 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Bd6 18.Be3 g6 19.Rh4 Qf5 20.Nd2 Qe5 21.Nf1 Qxb2 22.Re1 Qf6 23.Qg4 Be5 24.f4 Bc3 25.Re2 Qc6 26.Qg2 Qd7 27.Qg5 Qc6 28.Qg2
depth=29
0.00 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Bf6 18.Be3 g6 19.Rxc5 Bxb2 20.Rb1 Qxh3 21.Bd5 Bxd5 22.Rxd5 Qg4+ 23.Kf1 Qh3+ 24.Kg1
Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Post by Vinvin »

Try Nxg2, it should be good for black !
jp wrote: Thu Mar 26, 2020 8:29 pm
jp wrote: Thu Mar 26, 2020 7:27 pm Again that 14 ... Nxh3+ is a red herring for engines.
[d]r4rk1/1bpqbppp/p7/1p2R3/8/1B1P1N1n/PPP2PP1/R1BQ2K1 w - - 0 15

SF10 thinks 14... Nxh3+ leads to a draw:

depth=35
0.00 15.gxh3 Bf6 16.Rc5 Qxh3 17.Bd5 Qg4+ 18.Kf1 Qh3+


Komodo 13.2:

depth=24
0.00 15.gxh3 c5 16.Nd2 c4 17.Qg4 Qxg4+ 18.hxg4 Bd6 19.Re1 cxb3 20.Nxb3 f5 21.g5 Rac8 22.Nd4 Rfe8 23.Rxe8+ Rxe8 24.Be3 f4 25.Bd2 Bc5 26.Nb3 Bd6 27.Nd4
depth=25
-0.32 15.gxh3 c5 16.Bf4 Bd6 17.Ng5 Bxe5 18.Qh5 Qf5 19.Bxe5 c4 20.dxc4 h6 21.f4 hxg5 22.cxb5 axb5 23.Rf1 Be4 24.Qxg5 Qxg5+ 25.fxg5 Rad8 26.Re1 Bf5 27.h4 Rfe8 28.h5 Be6 29.Bc3 Bxb3 30.Rxe8+ Rxe8
depth=26
-0.19 15.gxh3 c5 16.Bf4 Bd6 17.Ng5 Bxe5 18.Qh5 Qf5 19.Bxe5 c4 20.dxc4 h6 21.f4 Rae8 22.Re1 hxg5 23.cxb5 axb5 24.Qxg5 Rxe5 25.Rxe5 Qxh3 26.Bd5 Bxd5 27.Rxd5 Qe3+ 28.Kh2 Qe2+ 29.Kh3 Qf3+ 30.Kh4 Qf2+ 31.Qg3 Qxc2 32.Rxb5 Re8 33.Kh3 Qh7+ 34.Kg4 Qg6+ 35.Rg5
depth=27
-0.33 15.gxh3 c5 16.Bf4 Bd6 17.Ng5 Bxe5 18.Qh5 Qf5 19.Bxe5 c4 20.dxc4 h6 21.f4 Rae8 22.Re1 hxg5 23.cxb5 axb5 24.Qxg5 Qxg5+ 25.fxg5 Kh7 26.Kf2 f6 27.gxf6 gxf6 28.Bd4 Rxe1 29.Kxe1 Kg6 30.c3 Rh8 31.Bc2+ Kf7
depth=28
0.00 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Bd6 18.Be3 g6 19.Rh4 Qf5 20.Nd2 Qe5 21.Nf1 Qxb2 22.Re1 Qf6 23.Qg4 Be5 24.f4 Bc3 25.Re2 Qc6 26.Qg2 Qd7 27.Qg5 Qc6 28.Qg2
depth=29
0.00 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Bf6 18.Be3 g6 19.Rxc5 Bxb2 20.Rb1 Qxh3 21.Bd5 Bxd5 22.Rxd5 Qg4+ 23.Kf1 Qh3+ 24.Kg1
jdart
Posts: 4366
Joined: Fri Mar 10, 2006 5:23 am
Location: http://www.arasanchess.org

Re: Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Post by jdart »

The whole line starting 12. Nxe5 and going through 15. .. a5 has actually been played in multiple games, for example Ivanchuk-So, Tata Steel-A 2015. Pavlovic annotating this says "it had been worked out by computers," so it was known previously and apparently found by at least some engines, maybe in correspondence play.

This is why you need an opening book.
jp
Posts: 1470
Joined: Mon Apr 23, 2018 7:54 am

Re: Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Post by jp »

Vinvin wrote: Fri Mar 27, 2020 1:51 am Try Nxg2, it should be good for black !
Komodo 13.2 did find Nxg2 already (from the position after Nxe5), but only after liking (the apparently drawing) Nxh3 first at depth 27 ( http://talkchess.com/forum3/viewtopic.p ... 05#p835805 ).

kiroje's run of Ethereal 12.00 above (from after Nxe5) shows that Ethereal at depths 34-40 also goes for Nxh3, and he did not run it beyond that to see if it switches to Nxg2 at depths >=41.

Hence my checking that Nxh3 does not also give Black an advantage, and my speculation it "distracts" the engines after they've found the previous Nf4. As well, as I said previously, the engine needs to see at the start all of ... Nf4; ... Nxg2; ... a5; ... Ra6, i.e. a5 is an only move, and at the end of it all Black has an advantage but not an obviously (to engines) decisive one. Maybe those are the factors that make the PV difficult to find.


jdart wrote: Fri Mar 27, 2020 3:49 am The whole line starting 12. Nxe5 and going through 15. .. a5 has actually been played in multiple games, for example Ivanchuk-So, Tata Steel-A 2015. Pavlovic annotating this says "it had been worked out by computers," so it was known previously and apparently found by at least some engines, maybe in correspondence play.
2015 engines were much weaker (though this could be a position where engine elo is not necessarily the best guide), so that Pavlovic comment may be giving too much credit to computers and not enough to humans working with them.

In that game, Ivanchuk went 16. Re7 and lost in 26 moves:

[pgn]1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 O-O 8. h3 Bb7 9. d3 d5 10. exd5 Nxd5 11. Nbd2 Qd7 12. Nxe5 Nxe5 13. Rxe5 Nf4 14. Nf3 Nxg2 15. Kxg2 a5 16. Rxe7 Qxe7 17. c3 Ra6 18. d4 Rf6 19. d5 a4 20. Bc2 Rd8 21. Qe1 Qd7 22. Ng5 h6 23. Ne4 Rg6+ 24. Kh2 f5 25. Ng3 Qxd5 26. Qg1 Qf3 [/pgn]
kiroje
Posts: 60
Joined: Wed Jul 25, 2012 10:12 am
Location: Copenhagen, Denmark
Full name: Kim Jensen

Re: Tactical mistake by Stockfish (Grischuk-Ding at candidates)

Post by kiroje »

Long run with Ethereal 12.00 4 threads and 8192Gb it actually finds Nf4 but get another move for white than Nf3

Code: Select all

42.42 12...Nxe5 (depth 41, 1/48), 14,5G nodes (5,6M nps)
0.00: +1.00/3 12...Nxe5 13.Rxe5 Bf6
0.00: +0.48/4 12...Nxe5 13.Rxe5 Bf6 14.Re1
0.00: +0.47/5 12...Nxe5 13.Rxe5 Bf6 14.Bxd5 Bxd5
0.00: +0.65/6 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1
0.00: +0.65/7 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8
0.00: +0.74/8 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Bd2
0.00: +0.61/9 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Be3 c5
0.00: +0.50/10 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Rxe8 Rxe8 17.Bd2 c5 18.a4
0.00: +0.59/11 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rae8 15.Nf3 Bd6 16.Bd2 Rxe1+ 17.Bxe1 c5 18.Bc3 Qf5 19.a4 Rd8 20.axb5 axb5
0.00: +0.38/12 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rae8 15.a4 Bd6 16.Rxe8 Rxe8 17.axb5 axb5 18.Nf3
0.00: +0.35/13 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rfe8 15.Nf3 Bd6 16.Rxe8+ Rxe8 17.Bd2 h5 18.a4 b4 19.Bc4 c5
0.00: +0.36/14 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rfe8 15.Nf3 Bd6 16.Rxe8+ Rxe8 17.Bd2 h5 18.a4 b4 19.a5 Qf5 20.Ba4 Re7 21.Bb3
0.00: +0.44/15 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rfe8 15.Nf3 Bd6 16.Rxe8+ Rxe8 17.Bd2 Re7 18.a4 b4 19.c3 Qf5 20.Bc4 c5 21.cxb4 cxb4
0.00: +0.07/16 12...Nxe5 13.Rxe5 Nf6 14.Re1 Bd6 15.Nf3 Rfe8 16.Bd2 Rxe1+ 17.Nxe1 h6 18.Nf3 c5 19.a4 b4 20.c3 Re8
0.00: 0.00/17 12...Nxe5 13.Rxe5 Nf6 14.Re1 Bd6 15.Nf3 Rae8 16.Be3 Nd5 17.Bxd5 Bxd5 18.a4 b4 19.c4 bxc3 20.bxc3 c5 21.c4 Bc6 22.d4 cxd4 23.Bxd4
0.00: +0.07/18 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rfe8 16.Be3 Qf5 17.a4 Nd5 18.Bxd5 Qxd5 19.Qe2 h6 20.axb5 axb5 21.c4 Qf5 22.Rxa8 Rxa8
0.00: +0.16/19 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Be3 Nd5 17.Bxd5 Bxd5 18.a4 b4 19.c4 bxc3 20.bxc3 c5 21.c4 Bc6 22.d4 cxd4 23.Nxd4 Be4 24.Qg4 Qxg4 25.hxg4 Bd3
0.01: +0.15/20 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Be3 Nd5 17.Bxd5 Bxd5 18.a4 b4 19.c4 bxc3 20.bxc3 c5 21.c4 Bc6 22.d4 cxd4 23.Nxd4 Bb4 24.Re2 Bb7 25.Rc1 Qd6 26.Rb2 a5 27.Nf3 Qxd1+ 28.Rxd1 Rd8 29.Rxd8 Rxd8
0.01: +0.05/21 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Be3 Nd5 17.Bd2 Rxe1+ 18.Nxe1 Nf6 19.Nf3 c5 20.c4 Qf5 21.Be3 Rd8 22.d4 Bc7 23.Bc2 Ne4 24.cxb5 axb5 25.Qd3 cxd4 26.Nxd4
0.03: +0.18/22 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Be3 Nd5 17.Bd2 Nf4 18.Rxe8 Rxe8 19.Bxf4 Bxf4 20.d4 Be4 21.c3 Qf5 22.a4 Bd6 23.c4 bxc4 24.Bxc4 a5 25.Qb3 Bxf3 26.Qxf3 Qxf3 27.gxf3
0.04: -0.04/23 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Be3 Nd5 17.Bxd5 Bxd5 18.a4 h6 19.c3 f5 20.axb5 axb5 21.b4 f4 22.Bc5 Rxe1+ 23.Nxe1 f3 24.Bxd6 cxd6 25.c4 bxc4 26.Nxf3
0.06: +0.01/24 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Be3 Nd5 17.Bxd5 Bxd5 18.a4 c5 19.axb5 axb5 20.Nh4 Be7 21.Nf3 Bd6
0.08: +0.01/25 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Be3 Nd5 17.Bxd5 Bxd5 18.a4 c5 19.axb5 axb5 20.Nh4 Be7 21.Nf3 Bd6
0.18: +0.26/26 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rae8 15.Ne4 Nxe4 16.dxe4 Qxd1 17.Rxd1 Bxe4 18.a4 Bf6 19.axb5 axb5 20.Be3 Ra8 21.Rxa8 Rxa8 22.c3 c6 23.Rd7 Ra1+ 24.Kh2 Bg6 25.Rd1 Rb1 26.Rxb1 Bxb1 27.Bc5 Be4
0.21: +0.01/27 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rae8 15.Nf3 Bd6 16.Be3 Nd5 17.Bd2 Qf5 18.Rxe8 Rxe8 19.Ng5 Nf4 20.Qg4 Ne2+ 21.Kf1 Ng3+ 22.Kg1 Ne2+
0.29: +0.01/28 12...Nxe5 13.Rxe5 Nf6 14.Nf3 Bd6 15.Re1 Rae8 16.Be3 Nd5 17.Bd2 Qf5 18.Rxe8 Rxe8 19.Ng5 Nf4 20.Qg4 Ne2+ 21.Kf1 Ng3+ 22.Kg1 Ne2+
0.44: +0.09/29 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rae8 15.Ne4 Nxe4 16.dxe4 Qxd1 17.Rxd1 Bxe4 18.Be3 c5 19.Bd5 Bxc2 20.Rdc1 Bg6 21.Bxc5 Bxc5 22.Rxc5 Be4 23.b3 Bxd5 24.Rxd5 Re2 25.a4 bxa4 26.bxa4 Rfe8 27.Rd6 R2e6 28.Rad1 h5 29.h4 Re1+ 30.Rxe1 Rxe1+ 31.Kh2 Re2 32.Kg3
0.57: +0.14/30 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rae8 15.Ne4 Nxe4 16.dxe4 Qxd1 17.Rxd1 Bxe4 18.Be3 c5 19.Bd5 Bxc2 20.Rdc1 Bg6 21.Bxc5 Bxc5 22.Rxc5 Be4 23.b3 Bxd5 24.Rxd5 Re2 25.Rd6 Ra8 26.a4 b4 27.Rad1 h6 28.R1d3 h5 29.R3d4 a5 30.Rd8+ Rxd8 31.Rxd8+ Kh7 32.Ra8 Rb2 33.Rxa5 Rb1+ 34.Kh2 Rxb3 35.Rxh5+
1.18: +0.13/31 12...Nxe5 13.Rxe5 Nf6 14.Re1 Rae8 15.Ne4 Nxe4 16.dxe4 Qxd1 17.Rxd1 Bxe4 18.Be3 Bd6 19.a4 Rb8 20.axb5 axb5 21.Ra5 Rfe8 22.Kf1 h6 23.Rd4 g5 24.h4 Re5 25.hxg5 hxg5 26.Rd1 Ra8 27.Rxa8+ Bxa8 28.c3 Be4 29.Kg1 Bb7 30.Bc2 f6 31.Bd3 Be4 32.Bxe4 Rxe4
2.28: +0.01/32 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Re5 Qd7
2.40: +0.01/33 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Re5 Qd7
3.03: -0.01/34 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Re5 Qd7
3.42: +0.01/35 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Qc6 18.Re5 Qd7
4.24: -0.01/36 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Bf6 18.Be3 g6 19.Rxc5 Qxh3 20.Bd5 Qg4+ 21.Kf1 Qh3+ 22.Kg1
6.35: +0.01/37 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Bf6 18.Be3 g6 19.Rxc5 Qxh3 20.Bd5 Qg4+ 21.Kf1 Qh3+ 22.Kg1
8.15: -0.01/38 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxh3+ 15.gxh3 c5 16.Qe2 Rae8 17.Rh5 Bf6 18.Be3 g6 19.Rxc5 Qxh3 20.Bd5 Qg4+ 21.Kf1 Qh3+ 22.Kg1
13.45: -0.81/39 12...Nxe5 13.Rxe5 Nf4 14.Nf3 Nxg2 15.Kxg2 a5 16.c3 Ra6 17.d4 Rg6+ 18.Kh2 Bd6 19.Nh4 a4 20.Nxg6 hxg6 21.Qg4 Qc6 22.Bf4 axb3 23.axb3 Bxe5 24.dxe5 Qe4 25.Rg1 Bc8 26.Qg3 Re8 27.b4 Qe2 28.b3 Be6 29.Be3 Bxb3 30.Qg4 Qxg4 31.hxg4 Rxe5 32.Ra1 Bc4 33.Ra7 c6 34.Bd4
37.07: -0.53/40 12...Nxe5 13.Rxe5 Nf4 14.Qg4 Qxg4 15.hxg4 Bd6 16.Re3 Nxg2 17.Re2 Nh4 18.Ne4 Nf3+ 19.Kf1 Be7 20.Bf4 Bc8 21.Nd2 Nxd2+ 22.Rxd2 Bxg4 23.Bxc7 h5 24.c3 Rac8 25.Bf4 h4 26.Re1 Rfe8 27.Bd5 g5 28.Bh2 Bf8 29.Re3 Kg7 30.Kg2 h3+ 31.Kg1 Rcd8 32.Rxe8 Rxe8 33.f3 Bf5 34.Bb7 Re3 35.Kf2 Rxd3 36.Rxd3 Bxd3 37.Bxa6 Kg6 38.Bb7 f6
40.05: -0.67/41 12...Nxe5
“Modern chess is too much concerned with things like pawn structure. Forget it, checkmate ends the game.” – Nigel Short