[Event "USSR Championship"] [Site "Moscow URS"] [Date "1957.02.21"] [Round "21"] [White "Mikhail Tal"] [Black "Alexander Kazimirovich Tolush"] [Result "1-0"] [Annotator "USER"] [Variant "Standard"] [ECO "E80"] [Opening "King's Indian Defense: Sämisch Variation"] [StudyName "Mikhail Tal Best Games"] [ChapterName "Mikhail Tal - Alexander Kazimirovich Tolush"] [ChapterURL "https://lichess.org/study/KJ5Vxrwu/LAPY0sPt"] [Orientation "white"] { Let's start our journey through the imaginative, wonderful world of the games of Mikhail Tal by examining his game against the old Soviet Master Alexander Kazimirovich Tolush from the 21st USSR Championship, held in 1957. --- Tolush, who would later become very famous as one of the trainers of the future World Champion Boris Spassky, was very known for his sharp and attacking style - so it is not surprising that his games against Mikhail Tal were often a spectacle in their own right! --- The two players have already created a mind-boggling masterpiece in the 1956 USSR Chess Championship (which will not be analyzed in this article, although it was analyzed by Garry Kasparov in his book On My Great Predecessors in the chapter about Tal). } 1. c4 { Throughout this career, Tal was playing all four main moves by White, although he was mainly 1.e4 player - especially in his youth. Still, I guess he had his reasons for the opening choice in this game - although it is not unfair to say that back in 1957, the significance of the opening stage at the top level was much less significant compared to today. } 1... Nf6 2. Nc3 g6 3. e4 d6 4. d4 Bg7 { The King's Indian was brought to the forefront at the end of 1940s and beginning of 1950s (particularly through the effors of Boleslavsky and Bronstein, among others) so it is not a surprising opening choice for Tolush - especially when we consider the traits of his style! } 5. f3 { The Saemisch variations is one of the sharpest and arguably most critical ways of meeting the King's Indian defence. The idea is to keep the center fluid, castle queenside and go for the sharp g4-h4 attack. It has remained quite popular to this day. and was tested extensively in one of the Karpov-Kasparov matches, although nowadays the neural networks are suggesting some other ways as more critical. } 5... e5 { This typical central counterstrike, characteristic of the King's Indian defence, is not the most popular move by today's standards as it reveals Black's intentions prematurely, but we shouldn't judge the state of the opening theory from 1957 too harshly. It has to be said that this move was played by some strong players like Ivanchuk and Sindarov, so it has every right to exist. } 6. Nge2 { A standard move in the Saemisch, defending the d4 point and intending the Be3/Bg5-Qd2-O-O-O developing scheme. Once again - other moves are possible, but I wouldn't want to delve too deeply in the opening stage of this game. } 6... Nbd7 { Also a bit strange as the knight is passive here and it blocks the light-squared bishop, although some strong players have played this even in 2020s. } 7. Bg5 { Developing the bishop on a more active square than e3 as the d4 pawn is not under the attack here. White is trying to provoke h6. } { [%cal Gh7h6,Gd1d2] } (7. Be3 { Is an equally valid alternative. White's plan remains the same. } { [%cal Gd1d2] }) 7... c6 { Taking the d5 square under control and opening the queen's diagonal. Very standard procedure. } { [%cal Gc6d5,Gd8a5] } (7... h6 8. Be3 { And now the h6 pawn can become an object of the attack. } { [%cal Re3h6] }) 8. Qd2 O-O 9. d5 { A very typical idea, closing the center before starting the kingside attack. White is not yet hurrying with castling and intends to keep the king in the center for a bit, as it turns out it is more difficult for Black to create counterplay with the king in the center compared to the king on the queenside. } { [%csl Gc4,Gd5,Ge4] } 9... c5 { This closing of the center was relatively a standard proceudre in the past, but it is just bad for Black as now White has a free hand on the kingside. } { [%csl Gc4,Rc5,Gd5,Rd6,Ge4,Re5] } (9... Nc5 { It was much better to keep the tension between the pawns and, say, move the knight and keep the d6 pawn defended. }) (9... h6 { The modern engine also suggests this move, which is based on a nice tactical shot } 10. Bxh6 (10. Be3 { Should instead be played when Black can go } 10... cxd5 11. cxd5 h5 { And now the fluidity of the kingside attack is prevented so Black has a much better position than in the game, although White still remains better due to the massive space advantage. }) 10... Nxe4 11. Nxe4 Qh4+) 10. g4 { The beginning of the standard procedure in the Saemisch. White wants to follow things up with moves such as Ng3-h4-h5 and so on. } { [%csl Ge4,Gf3,Gg4][%cal Ge2g3,Gh2h4,Gh4h5] } (10. h4 { Is also possible as now the blockading } 10... h5 { Doesn't work due to } 11. g4 hxg4 12. fxg4 { When the g4 pawn is immune due to the pin and Black is in mortal danger. } { [%cal Rf6g4,Rg5d8,Gh4h5,Gf1h3] }) 10... a6 { Trying to create counterplay on the queenside somehow, but Black is much, much slower. } 11. Ng3 { The standard attacking scheme in Saemisch. The knight goes to g3 and h4-h5 is next on the agenda, while the knight can also contemplate jumping on f5 in the future - as happens in this game. } 11... Re8 { This is just pointless, though. } (11... b5 { For better or for worse, Black should have started counterplay on the queenside. }) 12. h4 { White's play is very easy and straightforward while Black is struggling to create counterplay. } 12... Qa5 { Black tries to play for the b5 push now, but it is all very slow and shouldn't work. } 13. Bh6 { This logical move - exchanging the bishop that defends the king, seems to be a bit premature as White is not yet fully developed and is ignoring the play on the queenside. Although this is purely from the engine perspective - it doesn't yet relinquish White's advantage - it just reduces it somewhat. } (13. a4 { Is a move that computer really likes, simply nipping the b5 idea in the bud. }) 13... Nf8 (13... Bxh6 14. Qxh6 Kh8 { And now the king has prepared the move Ng8 if required, say after } { [%cal Gf6g8,Gd7f8] } 15. h5 Ng8 16. Qd2 Rb8 { And even though White has the advantage, it is not yet clear how the attack is going to continue and Black has more counterplay than in the game! Of course, this is purely computer defence which is incredibly difficult to see even in 2022, let alone in 1957. }) 14. h5 { Now we again have a one-way street as Black has no counterplay. } 14... Qc7 { Intending to play b5 and then c4 before recapturing on b5. } { [%cal Gc5c4,Gb7b5] } 15. Bd3 { Correctly judging that White shouldn't really fear the b5 advance. } (15. a4 { Was also possible, preventing the counterplay in the bud. However, since White is intending to castle queenside, it is probably not necessary to weaken the queenside further and provide Black with a hook on a4 they can use to play b5 in the future. }) 15... b5 { Finally, but this now comes a bit late in the day. } 16. O-O-O { Ignoring the demonstration on the kingside seems like the best approach. There are some other possibilities like sacking the knight on f5 at once, but this move seems very human and logical. } (16. cxb5 { would give Black some play after } 16... c4 17. Be2 (17. Bc2 axb5 18. Bxg7 Kxg7 19. Nxb5 Qb6 20. a4 Kh8) 17... Bxh6 18. Qxh6 axb5 19. Nxb5 Qb6 { [%cal Rb6b2] } 20. a4 Bd7 { Although even here White retains the superiority with the thematic tactical shot } 21. Nf5 { Threatening mate on g7 and forcing } 21... Bxf5 22. gxf5 { when White is firmly in the driver's seat. This Nf5 idea is very thematic in this position so it is definitely something worth noting. }) 16... bxc4 { Black opens the b-file, at least - but it turns out there is little they can do along it as White has many pieces defending the king and the bishop on g7 is not participating in the attack. } 17. Bb1 (17. Bc2 { Seems more precise, but I guess Tal wanted to keep the 2nd rank clear in case some lateral defence of the b2 pawn is required. }) (17. Bxg7 { Btw, it also seems it was better to include the preliminary exchange on g7 in order to avoid having the queen dragged to h6 and avoid that whole Kh8 business. } 17... Kxg7 18. Bb1 { and now if Black tries } 18... Kh8 19. g5 { is very strong, as the queen is not getting trapped on h6. }) 17... Bh8 { This is just bad, although very understandable. Tolush tries to keep the bishop that covers the dark-squares, but it loses some more time. } (17... Bxh6 18. Qxh6 Kh8 { was once again most resilient. The point is that now } 19. g5 (19. Qd2 { is what White should play instead, when g5 is coming and White is nevertheless in the driver's seat, but Black has gained a few tempi compared to the game. }) 19... Ng8 { Traps the queen. }) 18. Rdg1 { Now all pieces are participating in the attack and sacrifice on f5 is in the air. } { [%cal Rg1g8,Gg3f5] } 18... Rb8 { This is why Tal put the bishop on b1 - the b2 pawn is protected by the queen. } { [%cal Gd2b2] } 19. Nf5 { Here we go. This thematic pseudosacrifice is very tough for Black as the knight is untouchable on this square. } (19. g5 { The prosaic advance of the g-pawn was also very strong as after } 19... N6d7 20. Nf5 { Comes under arguably even stronger circumstances. At first glance it is hard to understand why given that now after } 20... gxf5 { The g-file is closed. But it transpires that the sinple recapture on f5 is good enough - but also that there is } 21. g6 hxg6 22. hxg6 fxg6 23. exf5 { With total destruction of Black's kingside. Black can't survive this position. }) 19... N6d7 { It is hard to suggest a move for Black, but this undeveloping of the knight is surely not the most resillient attempt. } (19... gxf5 { The knight, of course, can't be taken due to } 20. gxf5+) (19... Rb7 { or some other similar waiting move, might be a bit more resistant, although Black's position is very much lost all the same. }) 20. Bg5 { Preparing some Nh6-Nxf7 ideas. } { [%cal Rf5h6,Rh6f7] } (20. hxg6 { The prosaic opening of the h-file was also sufficient - and arguably even simpler, as after } 20... fxg6 21. Bxf8 Nxf8 22. Nh6+ Kg7 23. f4 { Black is helpless against the opening of the kingside, e. g. } 23... exf4 24. Qxf4 Qe7 25. Rf1 Rb7 26. g5 { And Qf6 and Qf7 threats are impossible to deal with. }) 20... Bg7 { Black tries to deal with the threat of Nh6 but ends up giving the bishop away. } (20... Qb6 { Black could have also ignored the threat, but after } 21. hxg6 fxg6 22. Nh6+ Kg7 23. f4 { Comes with a great effect once again - just like in the previous variation } 23... exf4 { And here there are several wins, the most brutal of them being } 24. Qxf4 Qxb2+ 25. Kd1) 21. Nxg7 Kxg7 22. Bh6+ Kg8 { Now Black's dark squares are very much weakened, although the position is much more resillient than it looks like at the first glance. } 23. f4 { Now with the knight gone from h6, this move seems to be not as powerful as in the previous variations, although White still keeps a big chunk of the advantage. } (23. Rh2 { Some patient move like Rh2, overprotecting b2 and maybe intending to double on the h-file, was probably better. White has all the time in the world to improve the pieces. }) (23. Bc2 { With the idea of playing Ba4 and exchanging the passive bishop for one of the defenders of the king, was also an interesting idea worth examining. } { [%cal Gc2a4] }) 23... exf4 24. Qxf4 Qd8 { Now Black is kinda covering some dark-squares and his position does have some degree of the resillience, although White is objectively completely winning. } { [%cal Gd8f6] } 25. hxg6 { Opening the h-file seems very sensible. } (25. Qxd6 { Is exactly what Black was hoping for, as then there are some crazy defensive ideas such as } 25... g5 { With the idea of Rb6. } { [%cal Gb8b6] } (25... Rb6 26. Qf4 (26. Qg3 g5 27. Bxf8 Kxf8 { And Black is actually the one to be preferred due to the control of the dark-squares, all of a sudden. }) 26... Ne5 { Is also not as clear cut as Black now has some nice dark-squares to work with. })) (25. Bg5 f6 26. hxg6 { Was also possible, weakening the king even further and profiting from the fact that the bishop on g5 is immune due to } 26... fxg5 27. Qf7+) 25... Nxg6 (25... fxg6 26. Qxd6 { Is now much stronger for White as the h- and f-files are opened, so after } 26... Rb6 27. Qg3 g5 (27... Ne5 28. Bf4 Qc7 29. g5 { Is also very strong for White. Yes, the knight on e5 is great, but we have the open h-file and a strong dark-squared bishop of our own. }) 28. Bxf8 Kxf8 { Now doesn't work due to numerous reasons - but mostly because of the open f-file } 29. Rf1+ Kg8 30. e5 { And now the weakness of the h7 is terminal. This is the difference compared to the previous line. } { [%cal Rb1h7,Rh1h7] }) 26. Qh2 { Lining up on the h-file. Despite all the inaccuracies, the position is still winning for White - which goes on to show how bad Black's position was to begin with. } { [%cal Rh1h8] } (26. Qxd6 Nge5 { Is once again very unclear and better to be avoided. }) 26... Nde5 { It is logical to bring the knight to e5 and open up the bishop, although there are other options as well. } 27. Bf4 { But this is a very strange decision, giving up the dark-squared bishop. } (27. Bd2 { Seemed like the best move - discovering an attack on h7 and not giving up the strong bishop. }) 27... Nf8 (27... Nxf4 { It was possible to take the bishop on f4 and then after } 28. Qxf4 Qe7 { Hope to establish some sort of a defense on the dark-squares with Ng6 and Qe5 next. It does look scary for a human, but the engine insists it is not at all so easy for White to demonstrate a way toward a clear advantage. }) (27... Qf6 { Was also more resillient, intending to return the pawn on h7 as after } 28. Qxh7+ Kf8 { Black has a very nice control over the dark-squares, the bishop is under the attack, g4 pawn is hanging and the king will escape to e7 where it will be very safe. White does have some advantage, but it is much less clear compared to what happened in the game. }) 28. Qh6 { Now White will get in Bg5 and maintain the pressure on the dark-squares. } { [%cal Gf4g5] } 28... Neg6 { A final mistake in an objectively difficult position. } (28... Rb7 { Or something along those lines - not moving the knight from e5 and trying to add some more protection along the 7th rank, was more resillient, although White is still very much winning. }) 29. Bg5 f6 30. e5 { A brilliant breakthrough at the most fortified point. White throws in a pawn, but opens up the bishop on b1 as well as the square on e4 for the knight. } { [%cal Rb1h7] } 30... Rxe5 (30... fxg5 31. Bxg6 Nxg6 32. Qxh7+ { Leads to a very quick mate. }) 31. Bxg6 Rb7 { Trying to add some protection along the 7th rank, but now this comes in too late. } { [%cal Gb7h7] } (31... Nxg6 32. Qxh7+ Kf8 33. Qxg6 { Once again leads to a mate very very soon. }) 32. Ne4 { Bringing the last piece that is not participating in the attack into the game ("Bring all the boys to the party") and not really caring about the fact that the bishop on g5 is hanging. } { [%cal Re4f6] } 32... fxg5 (32... Rxg5 { Is also losing after the prosaic } 33. Bf5 { When the pawn on f6 can't be defended. } { [%cal Re4f6,Rh6f6] } 33... Rf7 34. Bxc8 Qxc8 35. Nxf6+ $18) 33. Rf1 { A brilliant quiet move, threatening the capture on f8 and check on f6, among other things. } { [%cal Ge4f6,Gf1f8] } 33... Rxe4 (33... Nxg6 34. Nf6+ { Wins on the spot for White e.g. } 34... Kh8 35. Qxh7+ Rxh7 36. Rxh7#) 34. Bxe4 { Now it is all over as White is up an exchange and still has a raging attack going on. } 34... Rg7 35. Rf6 Bxg4 36. Rhf1 { A prosaic and "boring" way of ending up the game - but more than good enough. } (36. Bxh7+ { Was leading to a quick mate after } 36... Nxh7 (36... Rxh7 37. Rxf8+ Qxf8 38. Qxh7#) 37. Rg6 Rxg6 38. Qxg6+) 36... Nd7 37. Rxd6 Qe7 38. Rxa6 { Simply collecting all the pawns and threatening Ra8. } { [%cal Ga6a8] } 38... Kh8 39. Bxh7 { A flashy finish, exploiting the weakness of the back rank. } { [%cal Ga6a8] } 39... Nb8 (39... Rxh7 40. Ra8+ { Leads to a quick mate. }) 40. Bf5+ Kg8 41. Be6+ Bxe6 42. Rxe6 { And here Tolush finally decided to throw in the towel. Perhaps not the most beautiful game by Mikhail Tal, but the one I felt is a good introduction to his style and to some of the later brilliancies we will see in this article. --- Furthermore, I find these older games very instructive mainly because quite often, one side (in this case Tolush) completely misplays the opening and gets in a lost position early on - which is something that happens very often at the club level. Seeing a player of Tal caliber handling such a position - and also misplaying it at certain moments - can be regarded as more instructive and relatable than, say, recent encounters at the very top level. } 1-0