15 Qf4 ...

Option 1
15 ... Nxd4
16 Qh4 h6 (forced)
17 Bxh6 Nf5
18 exf5 ... and now black cannot stop mate...
19 ... gxf6 20 Qxf6+ mating
19 ... gxh6 20 Qxh6#
19 ... g6 20 Bxf8#
20 ... Bxf5 20 Bg5+ mating

Option 2
15 ... Bxd4
16 Rxd4 Nxd4 translates into the above variation one move later...

Option 3
15 ... Ne7
16 Qh4 h6 (forced)
17 Bxh6 Ng6
18 Qh5 gxf6 (19 Bg5# is coming! Bxd4 is met by Rxd4)
19 Bg5+ Kg8 (Kg7 loses quicker)
20 Bxf6 and the next two moves are unstoppable without sacrificing the black queen 21 Qh6
22 Qg7#

Option 4
15 ... Bxd4
16 Rxd4 Ne7 translates into the above variation one move later...

Spielmann vs Tarrasch, 1912 (13.?)

White to play and win.

Material: N for B. The Black Kg8 has 1 legal move and is vulnerable to 13.Ne7+ or 13.Nf6+. White has a battery Rd1 and Qd2 and generally good development, particularly with his centralized Nd4 and Nd5. White also has a second battery, Qd2 and Be3, supporting Nd5 in controlling critical dark squares around the Black Kg8. The Black Pf7 must support Be6, which Nd4 attacks, so Pf7 is effectively immobile, suggesting the candidate 13.Bg5. The White Kc1 is secured from check.

Candidates (13.): Bg5

13.Bg5

(1) Black cannot advance Pf7:

13…f6 14.Nxe6 Q any 15.Nxg7

There might be better moves, but White wins at least a clear P with a better position with this variation. Because of the immobile Pf7, the Black Qd8 must run.

(2) 13…Qb8 14.Nxc6 bxc6 15.<Ne7+> Kh8 16.Nxc6 wins a P

<[Toga evaluates 15.Ne7+ as even and prefers 14.Nf6+ a move earlier. After 15.Ne7+, Black regains the P with 16…Qb7, and the threat on Pe4 costs White a P.>

(3) 13…Qc8 14.Nxc6 (threatening 15.Nce7+ or 15.Nde7+ forking Qc8 and Kh8)

The fork threat forces Black to move Qc8, giving White a tempo to rescue Nc6, winning a N.

(4) 13…Qd7 [Qe8 is similar] 14.Nf6+ gxf6 [else, 15.Nxd7 wins Q for N]

15.Bxf6 (threatening 15.Qg5+ [or Qh6] 16.Qg7#)

Black must sacrifice excesses of material to prevent mate.

< <dzechiel> wrote: [snip] OK, does my line work as well? >

Whew! Too many candidates and a long time spent investigating each. Perhaps it would've been more efficient to consider more powerful threats first: 1) Nf6+ (a check and fails quickly); 2) Nxe6 (a capture and fork of Q & R); 3) Nxc6 (a capture hitting the powerful Q); 4) Bg5 (hitting the Q). And if those seem a dead end then consider Nxb6. That's the order I would recommend as best, in terms of examining the candidates. But it's also important not to go too deep before examining each (which was another mistake I made). But I did come up with the correct 13.Bg5 and saw most of the critical lines with exception to the game continuation 13...Qb8.

<dzechiel: ...

13 Bg5 ...

13...Qc8 14 Ne7+ Kh8>

After 13 of 17 rounds, Rubinstein headed the field with 9.5 points. He was followed closely by Tarrasch and Teichmann with 9 points. Duras, Marshall and Schlechter were next with 8.5 points, and Spielmann, the eventual 7th place finisher, trailed far back, with only 5 points.

An unexpected loss in the 14th round, Rubinstein vs M Lowtzky, 1912, knocked Rubinstein out of first place. Tarrasch and Teichmann both won, and moved into first place with 10 points. Duras and Schlechter also won, moving them into a tie with Rubinstein at 9.5 points. Marshall lost and Spielmann won, which started Spielmann's strong finish of four consecutive wins.

In the 15th round, Rubinstein won while Teichmann and Duras drew their game. Rubinstein and Teichmann headed the field with 10.5 points.

The other tournament leader Tarrasch, faced off against Spielmann in the 15th round. No doubt Tarrasch was intent on winning, but disaster struck on his 12th move. Analysis by Fritz indicates that Tarrasch had a number of reasonable alternatives to 12...Be6??.

Here are Fritz's nine top choices for Black's 12th move: (.16) (18 ply) 12...Re8 13.f3 Ne5; (.16) (18 ply) 12...Ne7 13.Nc3 Nc6; (.19) (18 ply) 12...Ba5 13.c3 Re8; (.21) (18 ply) 12...Nxd4 13.Bxd4 Bc6; (.32) (18 ply) 12...Qh4 13.Rhe1 Nxd4; (.32) (18 ply) 12...f6 13.Kb1 Nxd4; (.32) (18 ply) 12...Ne5 13.b3 Re8; (.36) (18 ply) 12...Rc8 13.Kb1 Ne7; or (.36) (18 ply) 12...Bxd4 13.Bxd4 Be6.

Tarrasch's move 12...Be6??, was Fritz's 25th choice!: (1.76) (18 ply) 12...Be6?? 13.Bg5! Qb8 14.Nf6+ Kh8, (2.59) (21 ply) 15.Nxc6 bxc6, or (1.75) (21 ply) 15.Nxe6 fxe6.

With this loss, Tarrasch remained with 10 points, a half of point off the lead, and tied with Duras and Schlechter, who had both drawn their games.

In the 16th round, Duras, Rubinstein and Tarrasch won, while Schlechter and Teichmann had draws in their games. Rubinstein headed the field with 11.5 points, Duras, Tarrasch and Teichmann were next with 11 points, and Schlechter was fifth with 10.5 points.

The scene was set for an exciting last round. Rubinstein could only draw in a long game against Teichmann's strong resistance Teichmann vs Rubinstein, 1912, while Duras moved into a first place tie with a victory over the veteran Burn Burn vs Duras, 1912. Schlechter, who could no longer attain first place, agreed to a quick draw with Cohn.

Tarrasch with 11 points was pitted against Carls, who was to finish 15th out of 18 at Breslau. No doubt, Tarrasch was anticipating a win, and a share of first place, if Teichmann could hold Rubinstein to a draw. However, the fates were against Tarrasch that day, and he suffered a disastrous loss, and had to be satisfied with a 4th and 5th place tie with Schlechter.

The final standings for the first seven at Breslau 1912:

I probably would have played some positional move here: either 13.f4 or 13.Rhe1.

Now I see the attack with 15. Nxc6 is worth more tho as Black can blunder and lose.

Good point! It seems that 17.Qh4 wins the house and to me would be a lot more convincing it's time to resign as black.

I have a "little red (hb) book" of Spielmann's games. This is game # 44 on page # 160 of that book.

In case you are curious, the book's title is "The Chess Career of Rudolph Spielmann," (Vol I) by J. Spence. [1961]

Analysis by Rybka 3 1-cpu:

1. = (0.15): 12...Re8 13.f3 Ne5 14.Kb1 Nc4 15.Qd3 Nxe3 16.Nxe3 a5 17.g4 Bc5 18.h4 a4 19.a3 h6 20.h5

As for the game, Tarrasch makes a slight slip in the opening and Spielmann never lets him recover. 5. ... Bb6 is not the strongest move in view of the more active Qf6. Spielmann chooses not to play 6. Qf5 where he would reach an advantageous position after 6. ... Bxe3 7. Nxe3 Nf6 8. Nc3 0-0 9. Qd2, but instead sticks to his game plan with 6. Nc3.

In order to castle quickly, Black must play d6 which costs him a move and is really the flaw behind Bb6 (i.e. 5. ... Qf6, 6. ... Ne7, 7. ... 0-0).

By 8. Qd2 the dye is cast. At this point, you might notice that White's position looks very much like the setup for the Yugoslav attack that White often uses to attack Black's Sicilian defense.

This is also a critical moment for Black. Unfortunately, the tactical 8. ... Ng4? is not a good response. It causes no problems for White since it still leaves White with the initiative after 10. ... Bd7 and it deprives Black of a key king side defender. Objectively, Ne5 might be better.

White to play (13.?) "Medium"

13. Bg5! exploits Black's poorly placed pieces. Now:

A) 13...f6?? 14. Nxe6 obviously loses, as does 13...Ne7?? 14. Nxe7+ Kh8 15. Ng6+.

B) Both 13...Qd7? and 13...Qe8? lose to 14. Nf6+! gxf6 15. Bxf6 followed by 16. Qg5+ or 16. Qh6 and a quick mate.

C) 13...Qc8? also loses to 14. Nxc6 bxc6 (or 14...Bxd5 15. Ne7+) 15. Ne7+, winning the queen.

click for larger view

Rybka 3:

[+2.21] d=19 16.Qf4 Rg8