Attacking with Tal – Part 3

June 15, 2018 Matthew Sadler 5 comments

I’ve been enjoying reading through Tibor Karolyi’s 3-volume Best Games Collection of Mikhail Tal and especially the 3rd volume dealing with Tal’s later games (1972-1992). I love those later games: Tal’s style is a wonderful blend of the swashbuckling attacking with which he stormed the pinnacle of the chess world with some smooth positional play and great endgame technique. In this week’s blog, we take a look at his game against the Italian IM Stefano Tatai from Las Palmas 1975. A playable version of this game is available at


Tal,Mihail – Tatai,Stefano [A61]

Las Palmas 1975


1.d4 Nf6 2.Nf3 c5 3.d5 e6 4.c4 exd5 5.cxd5 d6 6.Nc3 g6 7.Bg5



[A system against the Benoni that I played when I was very young as an extension of the Smyslov system against the King’s Indian – 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.Nf3 0–0 5.Bg5 – which my coach Steve Giddins had taught to me.]




[7…Bg7 8.e3 0–0 9.Nd2



represents White’s initial opening goal. This requires some explanation! Black’s main counterplay in the Benoni is directed against the White centre, and in particular the pawn on e4 which Black can attack easily with his rook along the half-open e-file. Black’s typical Benoni counterplay with …b5 is based on the premise that this play will overload one of the pieces defending e4 (typically a knight on c3) 1. Playing e3 instead of e4 avoids giving Black an early central target which should make counterplay with …b5 both less dangerous and more difficult to achieve. 2. Developing the dark-squared bishop outside the pawn chain with 7.Bg5 ensures that the bishop is not left passive on c1 after e2–e3. 3.Nd2 controls e4 (to prevent Black from using this square) and also removes Black’s option of netting the bishop pair with …h6, …g5 and …Nh5. Once White has completed his development, he will most likely execute one of 2 typical anti-Benoni plans:a. Lining up against the d6 pawn by bringing the dark-squared bishop to the h2–b8 diagonal and combining it with Nc4b. Advancing e3–e4 and then f2–f4 to activate White’s central pawn majority. This idea of playing first in a restrained fashion with e2–e3, developing and then only later expanding with the desired e3–e4 and f2–f4 reminds me strongly of the system I played for most of my professional career against the Benoni structures (usually transposing from a King’s Indian): 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 0–0 6.Bg5 c5 7.d5 e6 8.Qd2 exd5 9.cxd5. In this case, White plays f3 to shore up the e4 pawn to inhibit counterplay with …b5. Once White has completed his development, he looks to advance f3–f4! The thing that put me off playing this as White is the line that Tatai played: Black has a window of opportunity to win the bishop pair before White achieves his desired structure and Tatai – just like Tal did when he played the Black side – takes it!]


8.Bh4 g5 9.Bg3 Nh5 10.e3 Nxg3 11.hxg3 Bg7



[It’s worth spending a little time evaluating this position as quite a lot has happened in 11 moves! In a short space of time, Black has established a fair amount of control on the dark squares: his dark-squared bishop rules the h8–a1 diagonal and his counterpart has been exchanged. White has been given doubled pawns on the kingside and without the dark-squared bishop, Black’s pawn weakness on d6 is harder to attack. The light squares are the downside of Black’s position. Winning the bishop pair with …h6 and …g5 has cost Black some control of the f5 square, which would be a very tempting square for a White knight. Moreover, playing e3 instead of e4 has given White the ability to attack this square with a bishop on d3 and a queen on c2. In general, I think that Black should be fine, but Tal himself got into trouble with Black against Furman so some care is required.]


12.Qc2 Nd7 13.Bd3



[These moves look very natural for White, but they are actually quite subtle. Both against Botvinnik in 1960 and Germek in 1961, Tal had to face the early manoeuvre Nd2–c4 from White. In both cases, Tal reacted with …Nd7–e5, exchanging off the knight which has taken so much trouble to reach c4! In both cases, Tal had a comfortable game:


7…Bg7 8.e3 h6 9.Bh4 g5 10.Bg3 Nh5 11.Nd2 Nxg3 12.hxg3 0–0 13.Nc4 Qe7 14.Bd3 (14.Be2 Rd8 15.0–0 Nd7 16.a4 Ne5 17.Nxe5 Qxe5 18.a5 Rb8 19.Ra2 Bd7 20.Nb5 Bxb5 21.Bxb5 b6 22.a6 Rbc8 23.Qd3 Rc7 24.b3 Qc3 25.Qxc3 Bxc3 26.Rc2 Bf6 27.g4 Re7 28.Rc4 Rc8 29.g3 Bg7 30.Rd1 Rf8 31.Rd3 Kh7 32.Kg2 Kg6 33.Rd1 h5 34.gxh5+ Kxh5 35.g4+ Kg6 36.Rc2 Rh8 37.Bd3+ Kf6 38.Kg3 Ree8 39.Bb5 Re4 40.Rc4 Rxc4 41.bxc4 Ke7 42.Ba4 Be5+ 43.Kf3 Rh4 44.Rg1 f5 ½–½ (44) Botvinnik,M-Tal,M Moscow (Russia) 1960) 14…Nd7 15.0–0 Ne5 16.Nxe5 Qxe5 17.Re1 Bd7 18.Rb1 g4 19.Ne2 b5 20.b3 Rfc8 21.Nf4 c4 22.bxc4 bxc4 23.Bc2 Rab8 24.Qe2 Qc3 25.Rec1 Qa3 26.Kh2 a5 27.Re1 Rb4 28.Ne6 fxe6 29.Qxg4 Re8 30.Bg6 Re7 31.Qh4 Bf8 0–1 (31) Germek,M-Tal,M Bled 1961)


In this game, Tal follows Furman’s plan of leaving the knight on f3 and aiming at the f5 square instead. In that way, if Black does play …Ne5 to exchange knights, then White has at least not wasted time moving his knight on f3. Furman also had an excellent plan in mind which he implemented to perfection against Tal.]




[A natural developing move which had been played already by Vladimirov (although the position on the 23rd move in which a draw was agreed was pretty poor for Black!) but now the plan of exchanging knights with …Ne5 is much further away. I wonder whether Tatai was worried White might castle queenside if he castled kingside too early!]


[13…a6 14.a4 Rb8 15.0–0 0–0 16.Rab1 Qc7 17.Rfc1 Re8 18.b4



A great idea, expanding on the queenside where Black is typically strongest. 18…cxb4 19.Rxb4 Nc5 20.Bh7+ Kf8 21.Nd4



and this is the key point, justifying White’s entire opening strategy. The pawn on e3 provides support to a White knight on d4 which then clamps down on the f5 square weakened by Black’s early …h6 and …g5. Tal struggles to free himself, but has absolutely no chance as White consolidates his grip. 21…Bd7 22.Bf5 Rbc8 23.Rc4 Kg8 24.g4 White manages to make use of an advantage of his doubled g-pawns: he can clamp down on the f5 square. 24…Qa5 25.Nce2



Another great move, linking all White’s minor pieces to the attack and occupation of the weak f5 square. 25…Rcd8 26.Bxd7 Nxd7 27.Nf5 Nf6 28.Rc7 Qxd5 29.Ne7+ Rxe7 30.Rxe7 winning the exchange after which the result was never in doubt. 1–0 (68) Furman,S-Tal,M Tallinn 1971]




[Since …Ne5 is no longer on the table, Tal switches back to Botvinnik’s plan of Nd2–c4.]


[14.Rb1 Qa5 15.Nd2 a6 16.Nc4 Qc7 17.a4 0–0 18.0–0 Bd7 19.a5 Bb5 20.Nb6 Bxd3 21.Qxd3 Rae8 22.Rfc1 Re5 23.b4 ½–½ (23) Yuferov,S-Vladimirov,Y Minsk (Belarus) 1974]


14…0–0 15.Nc4 Qe7 16.a4 Bd7 17.0–0



[Black has developed his pieces to reasonable squares, but it isn’t too clear what he is intending from here on. I would be tempted to play …Ng4–e5 anyway, even though it has cost some extra time. Tatai’s slow plan doesn’t really hit the mark.]


17…Rab8 18.Rfe1


[An interesting plan from Tal, hinting at e3–e4 but hedging his bets by keeping a rook on the a-file for queenside play.]


18…b6 19.e4



[As we mentioned earlier, the fact that White played e3 earlier does not mean that he will never play e4 when he is ready for it! Black’s pieces are clumsily-placed and ill-equipped to deal with the typical e4–e5 break.]


19…Rfe8 20.Nb5


[20.e5 dxe5 21.d6 Qf8 22.Bf5 is mentioned by Karolyi and is also my engine’s strong preference. White has very strong compensation for the pawn as Nb5 and later a5 are coming. The text is also strong, though it gives Tatai time to re-establish some coordination by bringing the knight to g4.]


20…Bxb5 21.axb5 Ng4 22.Qd1 Qd7



[A very good defensive move: Black’s position stands or falls by his knight on g4! First of all, the knight illustrates the weak side of White’s early opening strategy: the doubled g-pawns make it hard to drive the knight away from its advanced post while White’s limited protection of the dark-squares (due to White allowing the exchange of his dark-squared bishop) turn the knight into a concrete threat (…Bd4 is threatened). These weaknesses are particularly evident as White begins to emerge from his cautious opening shell (for example playing e3–e4). Secondly, the knight covers the e3 square and stops White from transferring his own knight to f5. Not surprisingly, Tal decides that this monster knight is worth a rook!]


[22…Ne5 23.Ne3 Nxd3 24.Nf5 is pointed out by Karolyi, with a wonderful position for White.]


23.Rxa7 Qxa7 24.Qxg4



[An excellent practical idea, removing Black’s most active piece for the cost of the exchange. I would expect Black’s best course to be an attempt to activate his other pieces even at the cost of material: for example, 24…Qe7 as Karolyi suggests, followed by …Bd4 and …Ra8 to activate his rooks (which currently have no scope) even at the cost of a pawn. Tatai’s reaction is fairly meek.]


24…Rbd8 25.Qf5 Re5


[Giving back the exchange is one way to get rid of the powerful White knight, but now Black is fighting for a draw at most.]


26.Nxe5 Bxe5 27.f4 Bd4+ 28.Kh2 Qd7 29.Qxd7 Rxd7 30.Ra1





[The decisive mistake according to Karolyi. Defending the b6 pawn with the rook was better:]


[30…Kg7 31.Ra6 Rb7 32.Kh3 (32.b3 Be3 is the key resource pointed out by Karolyi and the reason he considered 28.Kf1 to be better than 28.Kh2.) 32…Bxb2 33.Kg4 Kf6 which Karolyi considers to be holdable for Black. I’m a bit nervous about that to be honest: White can play 34.Ra8 when he has all sorts of ways to dish out little pin-pricks to Black’s position like Rd8, Rh8, Re8 or Rc8–c6! I wouldn’t like to hold that with Black in time-trouble!]


31.Bxc4 Bxb2 32.Ra8+ Kg7 33.Be2



[A neat manoeuvre: the bishop goes to g4 and then White powers through with e5.]


33…Bf6 34.Rb8 Bd8 35.Bg4 Bc7 36.Re8 Rd8 37.Re7 Bb8 38.Rb7 Kf6 39.Bh5 Rf8 40.Kh3


[Black is paralysed!]






5 Comments on “Attacking with Tal – Part 3

  1. In the Furman – Tal game you wrote: “he can clamp down on the c5 square.” I guess it is f5 instead of c5 – KB5 instead of QB5 🙂

  2. Thank you for your commentary. Could you say a little more about what you have after move 7 about Nd2 removing Black’s option of …h6, …g5, …Nh5

      1. Yes. My bad! I guess the point is simply that by not playing 7 Bg5 the bishop is not subject to attack by …h6.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.