Is there a way to make a breakthrough from this position?

Study by Kling/Horwitz 1853. White to play.

Over the last few weeks we've been looking at endgames and, in particular, pawn promotion. After a queen exchange there's little chance of a checkmating attack, so getting a pawn to the other side of the board has to be the ultimate aim in most endings.

RB Somewhere in the recesses of my memory is a basic position similar to this one, though with three pawns rather than four facing each other, that allows the player with the move to create a passed pawn. So I'm going to try to drag up that position and try to use it as a starting point. Let's imagine there are no e-pawns. White would then play 1 g5 and if 1...hxg5 White can play 2 f5, and one of his two surviving pawns will queen before Black's.

Now, let's see if the same trick holds good for four pawns: 1 g5 hxg5 2 f5 gxh4 3 fxg6 (or e6) and White's pawn will queen first. So we've established that, yes, White can queen first. But we also have to take into account the king position and – here's the problem – the white king is on the back rank and opposed by Black's king, which means that queening first doesn't help because when Black promotes it is mate. The only way I can see to stop this from happening is for White to promote the f-pawn, queening with check. But I can't see how to do this and I'm also beginning to think that my whole approach may be wrong.

DK Here White has to go for an immediate breakthrough or Black's king will march over and take the pawns, but Ronan has hit the snag. It's simple to break through with a pawn sacrifice but when Black queens it is checkmate on the back rank. He has also spotted how White might win: queening with check on the f8 square. There's only one sequence that does it: 1 h5 gxh5 2 e5 fxe5 3 f5 hxg4 4 f6 and mission accomplished. After 1 h5 Black could try 1...g5, but 2 e5 is again the move.

chess@guardian.co.uk

How does white get a win from this position?

Endgame study by Prokes, 1944. White to play and win.

RB Is there anything more headache-inducing than an endgame study? And the simpler the position, the more intense the pain. I think, after almost an hour, that I've found the first three moves of the winning sequence, but I can't seem to find the knockout blow.

1 Kf6 seems obvious and good – go for the diagonal opposition! Since 1...g5 loses immediately to 2 Kg6, Black's only move is 1...Kh7. Now 2 g4 is equally obvious (after 2 h4 h5 White can make no further progress). Black has only two options, and since 2...h5 allows 3 g5, when the black king will be forced to retreat to the back rank and White will gobble up the pawns, 2...g5 is the sensible continuation. Now, I thought, I have him: 3 Kf7, gaining the opposition and forcing the king away from the defence of the pawns. Except there's a flaw in the plan: Black doesn't have to retreat but can play 3...h5 when White has nothing more than a draw after 4 gxh5 Kh6.

DK I don't share Ronan's feelings about so-called "simple positions". With so few pieces on the board I always feel I have a chance to find the truth in the position, just by the mechanical process of elimination.

Actually Ronan has almost cracked it. His logic is impeccable and has brought him to 1 Kf6 Kh7 2 g4 g5 3 Kf7 h5 – but this is no draw. White wins using a similar breakthrough to Magnus Carlsen's in last week's game. It's extremely rare that two chess games are identical, but patterns do recur. 4 h4! is the very visual winning move. The main variation runs 4...gxh4 5 g5 h3 6 g6+ Kh6 7 g7 h2 8 g8(Q) h1(Q) 9 Qg6 checkmate. After 4 h4 Black could try 4...Kh6 but White wins with 5 Kf6 gxh4 6 g5+ Kh7 7 Kf7 transposing to the line above.

Incidentally, in the starting position 1 Kf7 would only draw: 1...Kh7 2 g4 g5 3 Kf6 h5! 4 Kxg5 (in this case 4 h4 gxh4 5 g5+ Kg8 even wins for Black) 4...hxg4 5 hxg4 Kg7 gaining the opposition with a standard drawing position.

chess@guardian.co.uk

A superb lesson from Magnus Carlsen in endgame technique

Carlsen-Wang Yue, Medias 2010

It's been a while since we looked at the endgame, so this week we begin a short series on this often neglected area with a terrific demonstration of endgame technique from the ever-improving world No 1 Magnus Carlsen. Although White has the advantage of rook against bishop, Black has an extra pawn and a solid position. How did Carlsen make progress?

RB I'm pretty sure the answer lies in g4. The question is whether to nudge the pawn forward immediately or whether some preparation is needed first, perhaps exchanging on d5 or advancing the king to e3. The only way is concrete calculation, so let's see what happens after the immediate 1 g4. After 1...gxh4 2 gxh5 what would Black play? Maybe 2...Be4, threatening to block the pawn's advance. But White seems to have a whole host of good moves available – the rook could go to c7 or c4, the king could go to e3, or even g1 when it threatens to get to the black h-pawn via h2 and h3.

Does it make any difference if Black responds to 1 g4 with 1 hxg4...? Not that I can see. After 2 h5 Be4 3 Rc7 Black faces the same problem as he does in the first variation. In both cases White is completely winning, unless I'm missing something.

DK 1 g4 is the move. With this finesse White creates a distant passed pawn – quite an achievement considering that Black started with a pawn majority on this side of the board. The game continued 1 g4 hxg4 2 h5 Be4 3 Rc7. That's a really important move, not only preparing to push the pawn through, but preventing Black's king returning to block. Black tried to get counterplay with his own pawns, but White won the race with ease: 3...f5 4 h6 f4 5 h7 g3+ 6 Ke1 f3 7 h8(Q) f2+ 8 Ke2 Bd3+ 9 Ke3, and here Black resigned. He can also get a new queen, but White gets in first with the checks. For example: 9...f1(Q) 10 Qe8+ Kf5 11 Rf7+ winning the queen.

chess@guardian.co.uk

How did the late Russian grandmaster deal with this situation?

Smyslov-Ribli, London 1983.

White to move.

The life of a professional chess-player is mentally and physically challenging. After the age of 40, for most professionals, there is a seemingly inevitable slide down the rankings. Garry Kasparov retired at 42, rated No 1 in the world, but I suspect he felt that the chasing pack was getting too close. That's why Vassily Smyslov's career was so remarkable. Smyslov, who died earlier this year, became world champion in 1957 at the age of 36, but continued playing at the highest levels, qualifying for the final of the world championship at the age of 63. This was the victory that got him there and shows his ability to calculate incisively.

RB In sequences involving the calculation of more than three or four moves ahead I am always nervous about making that one false move. But this seems quite clear: 1 Qh8+ Ke7 2 Rxe6+ fxe6 3 Qxg7+ Kd6, and now White can win the enemy queen with 4 Ne4+ Kxd5 5 Qxb2. Unless I'm seriously deluding myself, this looks like curtains for Black.

DK When calculating, it's nice to have an obvious and reasonable move at your disposal. Here 1 dxe6 is possible. A check is threatened on e7, so Black must recapture – 1... fxe6 – and the exposed king should give White the better chances, even if there is no clear way to continue the attack.

Returning to the starting position, let's try the spectacular. It's always worth considering forcing moves and the most forcing are checks, so 1 Qh8+ Ke7, and going further: 2 Rxe6+ fxe6 3 Qxg7+. With the queens on the same diagonal, a tactic appears. Ronan has pointed out one way to win, but Smyslov had to show a little more ingenuity after the game continuation, 3...Nf7. White needs to deliver a check with the knight to win Black's queen, and this he achieves with the subtle 4 d6+, clearing the d5 square for the knight. After 4...Rxd6 5 Nd5+ Rxd5 6 Qxb2 the queen was in the bag and Black didn't last much longer.

chess@guardian.co.uk

How did the great Russian grandmaster win this game?

Bielicki-Smyslov, Havana 1964. Black to play

Vasily Smyslov, the former world champion who died in March, was often described as a master of the endgame and brilliant positional player, but great chess players are able to change their style according to the demands of the position. I've been going through his games and found strategic masterpieces but also blistering attacking play (I use ChessBase MegaBase but you'll find plenty of Smyslov's games at chessgames.com). Earlier in this game he sacrificed a piece to activate his forces and followed up by advancing his h-pawn – one of my favourite attacking methods. But it takes skill to finish the game.

RB This may be a very famous position but it's not one that I ever remember seeing and so I was very pleased with myself when I found 1…Rc4 almost at once, which is crushing for Black, or so I thought. After 2 Qxc4 Rd2, depriving the black king of his flight square, the queen will go to h1 and that will be that. I was so pleased, in fact, that I immediately started to write my side of the column. But then, just to be safe, I had another look and found what most of you will already have seen: Black can simply ignore the offer of the rook sacrifice and play 2 Qb2, keeping the second rank defended against the incursion of Black's rook.

Back to the drawing board. Neither 1…Qg2+ nor 1…Qh1 lead to anything, and 1…Qxg3, 1…Qxe3 and 1…Rxe3 are all obviously mad. So what about 1…Rg4, with the idea of 2…Rxg3. The only trouble here is that it is not a forcing sequence – White can play 2 Red1 (important to play the right rook in this variation) and save himself. Time to surrender.

DK Right ideas but wrong order! 1...Rc4 is correct, and after 2 Qb2 Rg4! now wins as the defence with Rd1 is impossible – the queen no longer covers the d1 square. Black threatens Rxg3 followed by pushing the h-pawn through to the queening square. There is no decent defence.

chess@guardian.co.uk

The Bulgarian is an aggressive player. But the Catalan doesn't bring about a quick victory

Topalov-Leko, Amber Blindfold 2008. Concluding our look at the Catalan, White to play.

RB When you see that the player with the move is Topalov you immediately start thinking tactics – quiet, careful improvement is not the Bulgarian's style. So, using the player's identity as a clue, and with an additional hint provided by the absence of a knight on f6, my eye is immediately drawn to 1 Ng5, with the twin threats of 2 Qxh7 mate and 2 Bxb7.

Let's have a look at how Black might counter: 1...g6 might not be as bad as it first appears. After 2 Bxb7 Rb8 both the bishop and knight are en prise. Here I would be tempted to play 3 Bc6, rather than retreat the bishop all the way back to its defensive post at g2, and try to maintain it there, cramping Black's queenside pawns. Black might be better off playing 1...Bxg5, in which case 2 Bxb7 Rb8 also allows the same idea of taking advantage of Black's backward c-pawn and light-square weakness with 3 Bc6.

DK Unmasking the Catalan bishop is a common tactic. Here it does not force the win of material, but allows White to take a firm hold of the position. 1 Ng5 Bxg5 (Ronan is wrong about 1...g6 because 2 Bxb7 Rb8 3 Ne4 Rxb7 4 Qc6 Rb6 5 Qxd7 wins) 2 Bxb7 Rb8 3 Bc6 was the game continuation. Black's backward c-pawn is a long-term problem: he could easily find himself condemned to passive defence. That's why Leko countered with 3...Nb6 4 Rd1 Qd6 5 Bg2 e5, but after 6 Nc3 (threatening Ne4) 6...f5 (preventing Ne4 but weakening the king) 7 dxe5 Qxe5 8 Nd5 Nxd5 9 Rxd5 White was still in control. See online for the rest of the game.

If you play the Catalan with White, don't expect a quick victory, but enjoy the space advantage, the solid king position and the sound pawn structure. Last week we recommended Play the Catalan by Nigel Davies (Everyman Chess, £14.99). For more advanced players, Grandmaster Repertoire 1 d4 Volume One by Boris Avrukh (Quality Chess, €24.99) has an excellent section on the Catalan.

chess@guardian.co.uk

Black has decided not to capture on c4. What is white's best option?

Continuing our brief survey of the fashionable Catalan Opening: instead of capturing on c4, Black holds firm in the centre. How should White develop?

RB Staring reproachfully at me from my desk is a newly bought but unopened copy of Play the Catalan by Nigel Davies (Everyman Chess). I haven't had the time to make even a start and am still as innocent of the theory as I was when we began this series of columns. I've resisted the temptation to cheat by looking to see what Davies recommends and am going to go on general principles: what looks like the most useful developing move?

Two possibilities suggest themselves: Qc2 and Nc3. So which one? Over the board I'd probably go for Qc2, only because it's a little more flexible: on the next turn I might not play the queen's knight to c3 but to d2. As for the dark-squared bishop, I'd be looking to develop it to b2 or, possibly, f4.

DK This position arises after the opening moves 1 d4 Nf6 2 c4 e6 3 Nf3 d5 4 g3 Be7 5 Bg2 0-0 6 0-0 c6. Bolstering the d-pawn means the Catalan bishop on g2 is locked out of play. Therefore there is some logic in White opening up the middle of the board with the pawn break e4. I like 7 Qc2, supporting the c-pawn and bringing the e4 square under control. Black might reply 7... Nbd7. Here Ronan wants to play the queen's bishop into the game on b2 or f4 – both possible - but I think it is more to the point to get in e4. I favour 8 Nbd2. Black has to bring his queen's bishop into the game, so 8...b6, and now 9 e4. Black has a choice. If everything is exchanged on e4 White is left with easy development, so in practice most go for 9... Bb7. Now White should push with 10 e5, forcing the knight to retreat to the back rank. I've played both sides of this position and it's a lot more fun to have White's space advantage than Black's cramped pieces. I'd strongly recommend that Ronan opens Play the Catalan: Davies explains these positions well.

chess@guardian.co.uk

In this opening, the action is often on the queenside. How best to make use of that knowledge?

Le Quang Liem-Azmaiparashvili, Vung Tau 2008. Continuing our survey of the Catalan, White to play.

RB The themes of the Catalan are coming through loud and clear even for those as innocent of the opening as I am. Last week I was struck by the duel of the bishops along the h1-a8 diagonal and here again the bishops are squaring up. A second theme seems to be recurring: activity on the queenside. Does the Catalan tend to drag play to the queenside, or is it just a quirk of the games we've been looking at? So, armed with this general knowledge, let's see if we can find a way forward for White.

The answer, depressingly, is no. Black's position looks pretty solid. The only weakness I can see is the knight on a6. He'd probably like to play it to c5 but obviously can't just yet. However, he will be able to get it into the game with ...Nc7 and possibly ...Nd5, where it will be very nicely placed. Then, under cover of the bishop on c6, he can double his rooks on the c-file and he'll be fine. Can I stop that, assuming this is the best way for Black to go? I don't see how.

DK Yes, White's activity often takes place on the queenside in the Catalan: that's where the fianchettoed bishop points, where White made his initial pawn moves, and where the queen can easily emerge. As we have seen over the last couple of weeks, defence can be depressing for Black as it is difficult to counter-attack White's solid king position. This game is no different.

Black looks comfortable, but White has a typical Catalan manoeuvre that secures the advantage: 1 Ne1! Bxg2 2 Kxg2. Black must do something about the attacked b-pawn: 2...b6 and now 3 Nd3! prevents Black's knight from entering the game and kills counterplay on the d-file. White is ready to probe on a4 with his queen, so Black tried 3...Qd4, but White took control over the c-file with 4 Qc4 Qxc4 5 Rxc4 and won a beautiful ending. More on the Catalan next week.

chess@guardian.co.uk

How can White best exploit the potential of his fianchettoed kingside bishop?

Kramnik-Carlsen, Dortmund 2007. Continuing our look at the Catalan, White to play.

RB I'm a Catalan virgin, but I'm beginning to see some themes, in particular the power of the bishop on g2. Unfortunately the bishop's dominance of the h1-a8 diagonal is contested by the black bishop on b7. If I could somehow block its scope, my own bishop's potential would be enhanced. That's why I'm fixating on the c6 square.

Ideally, I'd like to land a knight on c6, and maintain it there or force Black to swap off his bishop. The trouble is that Black is threatening to win a pawn by exchanging on b4. I don't want to play 1 bxa5 because of 1...Ra8 or, possibly stronger, 1...c5. So what to do? Occupy c6 with the knight immediately. I'm convinced I'm on the right lines, even if 1 Nc6 Bxc6 2 Qxc6 axb4 3 axb4 Bxb4 doesn't look that appealing.

DK This is a typical Catalan position in which White's d4 pawn gives him a space advantage. Carlsen was hoping for 1 bxa5 when 1...c5 liberates his pieces and hits out at the centre. Usually, if White can prevent the c-pawn moving forward, he will hold a long-term advantage. Unfortunately, as Ronan notes, 1 Nc6 just ends up losing the b-pawn.

1 e4 is a tempting move, to expel the knight from its central position, but after it retreats White's d and b pawns would both be under fire.

Instead, Kramnik found 1 Nb3, temporarily sacrificing a pawn, but the first step in setting up a massive clamp. Over the next few moves Black could only roll with the punches: 1...axb4 2 Na5! Ba8 3 Nac6 Bxc6 4 Nxc6 Qd7 5 Bxd5 exd5 6 axb4. Mission accomplished. The c-pawn is nailed to the spot and the knight dominates, splitting Black's queenside pawns. Even though White has given up his Catalan bishop, there have been so many exchanges that his king remains safe. Kramnik soon won a pawn by playing Ra1-a5 and, shortly after, the game. More on the Catalan next week.

chess@guardian.co.uk

The classic opening served Anand well in his chess world championship matches. But how best to exploit its strengths?

Kramnik-Morozevich, Mexico City 2007. White to play.

The Catalan opening gave Anand two wins in the world championship, and has long been popular among the world's elite. It is basically a Queen's Gambit in which White fianchettos. It can be reached via several different move orders, for example: 1 d4 d5 2 c4 e6 3 Nf3 Nf6 4 g3, or 1 d4 Nf6 2 c4 e6 3 g3 d5 4 Bg2. The bishop is tricky on the long diagonal, and the fianchetto provides a safe king position after castling.

RB I've always tended to get the Catalan confused with the Bogo-Indian, various Benoni systems, and even the fianchetto variation of the King's Indian, and so given it a wide berth. The themes and ideas behind it are, to me, very murky. Perhaps this kind of position is characteristic of the Catalan – who knows? Not me anyway. All I can see is that the retreat 1 Nf3 dxe4 2 Qxe4 Re8 gives Black a pleasant position. After 3 Qc2 (or perhaps 3 Qb1) Black could play 3... Be6 and suddenly the phalanx of black queenside pawns looks very worrying.

But if the retreat is passive, what's the alternative? Is there a sacrifice on c6? It looks as if there should be, but I can't see how to make it work. Likewise it's tempting to leave the knight on e5 and look for some way to unbalance Black – except that I can't find it either. Over to the grandmaster.

DK Black's last move – pawn to f6 – provokes a crisis. If White backs down, as Ronan demonstrates, he will give Black time to develop. Kramnik realised he had to strike: 1 exd5, giving up the knight, but opening up the diagonal of the Catalan bishop. After 1...fxe5 2 bxc4 the position had switched round – White suddenly had the dangerous mass of pawns, and that, combined with the weakness of Black's king, led to victory.

Kramnik has been the greatest exponent of the Catalan in recent times, playing it in a far more aggressive fashion than was characteristic of games in the 1980s. More from him next week.

chess@guardian.co.uk

Topalov's aggressive style made for a classic match. But in the final game, Anand was just too good for him

Topalov-Anand, world championship (game 12), Sofia 2010. Black to move.

World championship matches were once truly epic in scale. Alekhine and Capablanca (1927) slugged it out in 34 games over 10 weeks, while Kasparov and Karpov made the first move of their first extraordinary encounter on 10 September 1984. It ended when Karpov turned over his king on 9 February the following year, after 48 games.

Things are different today. Rule changes, the emergence of computers – putting an end to adjournments – and the struggle to find sponsorship have led to a more compressed format. In Sofia, Anand and Topalov played 12 games over a little more than two weeks. And while the games were watched by thousands online, the event was barely mentioned in the non-chess media.

The compensation for the diminished scale and interest was in the chess itself, which was thrilling. Topalov is an uncompromisingly aggressive player. He promised to bring the fight to Anand and this he did. The match went to the wire, with the two contestants level as they went into the 12th game.

White seemed to be doing fine until Black started to open up the centre in this position with 1...f5. Now Topalov made an extraordinary blunder: 2 exf5 (2 Nd2 was necessary). Anand played 2...e4 and Topalov committed a second and even graver error: 3 fxe4 (in a post-match interview Anand said, "I honestly thought he had gone nuts"). Play continued 3...Qxe4+ 4 Kh3 Rd4 5 Ne3. And now Anand played his masterstroke 5…Qe8, which Topalov must have overlooked when he took the f-pawn (backwards moves are easily overlooked) with the threat of 6 …Qh5 mate. Topalov fended off the mate, but he lost first his queen, then the game and lastly what could well be his last shot at the title.

The Catalan featured prominently in the early phase of the match, and, being nothing if not trendy, next week we'll be starting a short survey of this newly fashionable opening.

chess@guardian.co.uk

The chess World Championship is on a knife-edge. But Anand could have made a mark in this game

Anand-Topalov, game 9, World Championship 2010. White to play.

Anand and Topalov go into the 12th game of their match tomorrow with scores level. If drawn, tie-break rapid games will be played on Thursday. Topalov's declaration before the match that he would neither offer nor accept a draw clearly had a liberating effect on both players, resulting in one of the most entertaining world championship finals we have seen in years. All the games have been long and hard-fought, so it's hardly surprising that tiredness is affecting play. Anand missed several chances to win in game 9, Topalov had a promising endgame in game 10 but allowed Anand to slip out, and game 11 was looking like the dullest of the match, but suddenly exploded into life after inaccuracies from both players.

If Anand loses this match, it is the ninth game that will haunt him. He prepared an excellent opening, achieved a winning position out of the complications but, unusually, ran short of time with Topalov rushing his moves to try to unsettle the Indian. Anand had a couple of minutes to play his last move before the time control and the position was complicated, but two rooks should be superior to the queen, particularly with Black's king in such an exposed position. Unable to find a clear continuation, Anand played 1 Rh8+, thinking that he would be able to repeat the position after 1...Ke7 2 Rh7+ forcing the king back to e8. With the time control reached he would have an extra hour to work out the win. Instead, Topalov played 1...Kd7! 2 Rh7+ (2 Rd1+ Nd3! is probably what Anand missed, and the rook on h8 is inconveniently threatened by the queen) 2...Kc6, slipping away to the queenside and putting the win in doubt.

Instead, in the starting position Anand should have moved his threatened rook to a safe square: 1 Re4. With the king still trapped on the back rank, Black will not be able to survive: 1...b4 2 Rxa7 b3 3 Rb7 b2 4 Kh2 Qc1 5 Ra4, and the mate threat on a8 is too much.

chess@guardian.co.uk

A classic opening has had an unexpected revival at the world championship

Anand-Topalov, world championship game 4, Sofia 2010

At the halfway stage, defending world champion Vishy Anand led challenger Veselin Topalov 3.5-2.5. It has been an exciting contest so far, full of fighting chess. After Topalov stormed out of the traps to score a stunning win with the white pieces in the first game, the champion struck back to win games two and four, also with white (with three draws).

Many world championship encounters involve theoretical opening duels. When Alekhine and Capablanca slugged it out in Buenos Aires in 1927 it wasn't just the title that was at stake but the reputation of the Orthodox Defence to the Queen's Gambit, which was played in almost every one of their 34 games. In London in 2000, defending champion Kasparov's inability to break down Kramnik's Berlin Defence to the Ruy Lopez was critical in deciding the outcome.

In Sofia, Anand's two wins so far have been achieved with the Catalan – 1 d4 Nf6 2 c4 e6 3 g3 d5. The opening was invented to order by the Franco-Polish player and chess writer Savielly Tartakower (1887-1956) who was asked by the organisers of the 1929 Barcelona tournament to come up with an opening to name after the city's region. Anyone interested will find useful chapters on the Catalan in Boris Avurkh's Grandmaster Repertoire 1 d4, vol 1 (Quality Chess).

In the position, Anand has broken up Topalov's king defence with an earlier knight sacrifice on h6. With the black pieces on the wrong side of the board, White now pressed his advantage with 1 e5, against which there is no defence. After 1...Bxg2 (1... Qg7 2 Qxg7+ Kxg7 2 Bxc6), White continued 2 exf6 Rxd6 (2...Qh7 would not have helped: 3 Qg5+ Kh8 4 Rc4 Rg8 5 Nf7+ Qxf7 and mate will follow) 3 Rxd6 Be4 4 Rxe6 Nd3 5 Rc2 Qh7 6 f7+ Qxf7 7 Rxe4 Qf5 8 Re7 and Black resigned.

The remaining games can be watched live on anand-topalov.com.

chess@guardian.co.uk

The Bulgarian challenger finished the first game in brutal style

Topalov-Anand, World Championship, game 1. White to play.

After two games, the score in the Anand-Topalov match world championship match is even. Bulgarian Topalov won the first, but Anand struck back in the second. As is often the case, the moves off the board play a role too.

Anand's flight to Sofia was cancelled due to the ash cloud, so he was forced to make a 40-hour car journey from Frankfurt. He asked for the first game to be postponed by three days, but the organising committee – which includes Topalov's manager and Bulgaria's PM – rejected the request. The match only started after the International Chess Federation insisted on a compromise of one day's postponement.

Topalov's win came as a result of excellent home preparation in a tense opening variation – he rattled off his first 20 moves in a matter of minutes – before delivering a crushing sacrifice and hounding Anand's king into submission. Anand also moves quickly, but he does occasionally succumb to this kind of blitzkrieg tactic. In game 2 the Indian played a position more to his taste, sacrificing a pawn for long-term pressure and exchanging queens, reducing the chance of a Topalov attack. Although the Bulgarian's position wasn't bad, Anand outplayed him strategically and forced resignation after 43 moves. Judging by these games, the battle lines have been clearly drawn: Topalov will be looking for attacking knockouts while Anand will aim for positions where his strategic sense plays a greater role.

Topalov's finish in game 1 was brutal. He sacrificed with 1 Nxf6 Kxf6 and followed with 2 Rh3. Swinging the rook over to the h-file after advancing the f-pawn is a theme we examined a few weeks ago. The threat is to check on h6 and play Qxg5. Anand defended the g-pawn with 2...Rg8, but his poorly placed pieces could not prevent invasion: 3 Rh6+ Kf7 4 Rh7+ Ke8 5 Rcc7 Kd8 6 Bb5 Qxe4 7 Rxc8+. Anand resigned. If 7...Kxc8 8 Qc1+ is the end.

Game 3 starts today at 1pm (UK time) and you can follow it live online.

chess@guardian.co.uk

At 35, this may be the Bulgarian grandmaster's last chance to take the title

Topalov – Anand, Sofia 2005. White to play.

On Friday, in Sofia, Veselin Topalov will begin his challenge to Vishy Anand for the unified Fide/classical chess world championship title. The high point of Topalov's career to date was in 2005 when he won the Fide world championship tournament at San Luis ahead of Anand, Svidler, Leko and other super grandmasters. It may have been that the format suited Topalov – he has always been considered a stronger tournament than match player.

The following year, at Elista, he lost his bid to unseat Anand's predecessor Vladimir Kramnik as classical chess world champion in a match marred by gamesmanship and (unproven) allegations of cheating. Topalov, now 35, will be aware that this is probably his last chance to capture the title before the pack of hungry young lions led by Magnus Carlsen see him and his generation off for good.

We know very little about Topalov the man. He rarely gives interviews and, when he does so, restricts himself to chess matters. At the board, however, he is an aggressive, attacking player. He once said. "I'm not afraid to lose. This makes the difference between me and [other players]." His play is all about heightening the tension and upsetting the balance.

The position demonstrates Topalov's characteristics: the deep opening preparation (the novelty on move 11, followed by the knight sacrifice on move 14, came after analysing – with Anand! – an earlier game against Leko), the willingness to gamble, and his liking for fiendishly complicated tactical play. Like Tal's combinations, Topalov's may not always be sound, but they are not easy to refute. In this game Anand failed to find the correct continuation and went on to lose.

But even Topalov lost track of his own creation when he played 1 Rxe7+. Winning by force was 1 Re5, when 1...Bb7 (if 1...Be6 2 Nxh8+ Qxh8 3 Rxe6 Kxe6 4 Qc6+ and White scoops up practically every remaining enemy piece) 2 Qf5+ Kg8 3 Bc4 is crushing for White.

chess@guardian.co.uk

Age . . . playing style . . . even gamesmanship. There are many vital factors in the upcoming clash

Anand-Leko, Amber (blindfold) 2009. White to play

Vishy Anand defends his world title against Veselin Topalov in Sofia in a 12-game match starting on 23 April. Most pundits reckon Anand is the better player, but the stats tell us that in their classical chess encounters Topalov has one more win than Anand. It's only in games with a faster time control that Anand has a big plus score. If the match ends 6-6, Anand will be a clear favourite in the tie-break rapid games.

Age could be a factor. Anand has turned 40. Among top chess players that practically qualifies for a free bus pass. Topalov is five years younger and will be a highly motivated challenger: he has yet to win a "classical" world championship match and he is also playing in front of his home crowd.

Topalov's team have not been above gamesmanship in past matches. You may recall the infamous "toiletgate" scandal of Elista 2006 when they accused Kramnik of using computer assistance during visits to the restroom. In the runup to this match there has already been an attempt to needle Anand. Topalov has declared that, over the course of the entire match, if Anand wishes to communicate with him, he should do so through the arbiter. Anand and his team have reacted to this posturing with typical discretion and diplomacy. During a recent inspection of the playing site, Anand's wife and manager, Aruna, stated: "We've visited Bulgaria three times and we've always been well accepted, which leaves us greatly impressed. I am convinced that after this match we'll become even better friends." We'll see.

Anand has a pragmatic and fluent middlegame style, preferring to play a move that looks good rather than sweating to try to find the best move. But when it comes to the critical phase of the game his tactics are deadly, as here. Anand finished with 1 Bxf7+ Rxf7 2 Qh8+ and Black resigned. If 2...Kxh8 3 Nxf7+ picks up Black's queen with a simple winning ending.

chess@guardian.co.uk

It's never easy to face a brace of queens. But can black find an answer here?

Kramnik-Gashimov, Melody Amber 2010. Black to play.

RB It's never exactly a comfort to find yourself facing a brace of queens, but at least they're as far from the defence of their own king as it's possible to get, and that king is exposed. Also, as long as the black queen can keep an eye on f8 neither enemy queen can give check on the next move. So there may be a chance for a draw.

Three possible moves suggest themselves: 1...Qxg3, Ne1 and 1...Qf1+. 1...Qxg3 looks tempting – the threat is 2...Nf2 mate – but it has the rather obvious and slightly fatal drawback of 2 Qf8+ Kh5 (2...Kg5 is no better) 3 Qxh7+ with mate to follow. What about my second option, 1...Ne1...? This looks promising. White still can't give check on f8, and the threat is 2...Qf1+ 3 Kh2 Qf2+, with a perpetual. Okay, let's test this. After 1...Ne1 I want my queen to go to f1. Can White prevent this? Yes, he can – 2 Qb5 or 2 Qba6.

That leaves 1...Qf1+. Let's see where this goes after 2 Kh2. Obviously not 2...f2 because of 3 Qg2. Wait, I think I've got it – 2...Ne1, when White can't block the queen on f2.

DK It's often said that queen and knight complement each other well and this is a case that supports the rule. Here they even match the power of two combined queens. The only way to approach this position is with precise calculation and Ronan has reasoned correctly. After 1...Qf1+ 2 Kh2 Ne1 White is unable to prevent Black giving perpetual check on f2 and f1 with a draw. However, such reasoning takes time, and in this rapid-play game Gashimov didn't have much to spare. He played 1...Ne1, Kramnik defended the f1 square with 2 Qb5 and Black was lost. If 2...Nf3, White mates first with 3 Qf8.

Vishy Anand and Veselin Topalov begin their world championship match in Sofia on 23 April. Over the next few weeks we'll be profiling the players and giving the latest results. Tou can follow the build-up on the official match site: anand-topalov.com

chess@guardian.co.uk

It can be addictive following the Amber tournament online - but what should Black do here?

Carlsen-Smeets, Amber (Blindfold) 2010. How should Black play?

RB Every now and then, when a deadline looms, I go and hole up in an attic for a week or two. With no – or at least fewer – distractions, I can get a lot of writing done. However, the attic does have an internet connection, and one distraction I've been unable to resist is the live coverage of the Melody Amber tournament. The rapid and blindfold games are the perfect internet spectator sport – it's great when you can keep pace, even just for a move or two, with the world's elite.

I watched Carlsen-Smeets live, but such is my chess memory that by the time Dan sent me this position I'd forgotten how the game went. Obviously not 1...Kxe5 because of the bishop fork 2 Bd4+ and catastrophe. Let's see, does White have any threats? Yes, he does – a huge one in 1 Rd6+, which looks so serious after 1...Ke7 2 e6 that I'm not even going to waste time going there. What about 1...Rxc4, which at least avoids the bishop fork on d4? Except then 2 Bg5 looks fatal.

Although White is a pawn up, Black has the better pawn structure. If Black could swap off a pair of rooks it would not only put paid to White's mating threats along the d-file but should increase the chances of snaffling his isolated pawns. So I'm going for 1...Rd7.

DK 1...Rd7 is Black's best move – and I can't fault Ronan's reasoning. After 2 Rxd7 Kxd7 3 e6+! Ke8 4 Bd4, attacking rook and g-pawn, keeps the initiative. But with so many pawns exchanged Black has reasonable drawing chances.

In the game, Black fatally took the bait: 1...Rxc4 and was swiftly dispatched: 2 Bg5 threatened Rd6 mate, Black parried with 2...Rd7, but 3 Rxd7 Kxd7 4 e6+ Ke8 5 Rd1 and Rd8 mate was unstoppable.

All major tournaments can be followed online, often with live commentary. The two best sites are (playchess.com) run by ChessBase and the Internet Chess Club (chessclub.com)

chess@guardian.co.uk

In this world-championship game, Black has a very big decision to make

Kasparov-Anand, World Championship 1995. Black to play.

RB What better way to finish our survey of the f-pawn push than with a world championship game? A Scandinavian defence too, which we also took a look at in a recent series of columns, though I don't remember this variation.

White's knight looks to have found a powerful and well-supported post, and an exchange of knights on e5 – 1…Nxe5 2 fxe5 – would leave White with a very menacing pawn wedge and the semi-open f-file for his rooks. So we won't be trying that. What else have we got? Say Black castles, which seems like a sensible step. Would White want to push the f-pawn? I don't think so: 1…0-0-0 (or 1…0-0) 2 f5 just loses a pawn after 2…Bxf5 (3Bxf5?? Nxe5).

OK, I don't see any immediate threats, so castling it is. Kingside or queenside? I'm going to go for queenside, bringing the queen's rook into the game at once, but, more important, to have my king on the other side of the board should White find a way to force open the f-file.

DK By advancing the f-pawn Kasparov signals that he would like to attack on the kingside, so at the moment I would be reluctant to play 1...0-0. With 2 Qe1 White transfers the queen to h4 and looks to break with f5 in standard fashion. No thanks.

It's understandable that Ronan wishes to play the king to the other side of the board, but 1...0-0-0 looks too draughty. I like 2 Be3 to support the centre followed by advancing the b-pawn to open a file.

In the game, Anand kept his options open and went for 1...Bf5, physically blocking the f-pawn, and in turn blocking in the bishop on c1 and rook on f1. 2 Bxf5 exf5 would only give Black a potential outpost for the knight on e4, but this might have been better than 2 Bc4 e6 3 Be2 – Kasparov is attempting to trap the bishop – but 3...h5! prevented g4 and the f-pawn's blockade cannot be lifted. A few moves later Anand castled safely on the kingside and stood well.

chess@guardian.co.uk

It it time for black to use his f-pawn as a battering ram?

Balashov-Vaganian, Odessa 1989. Black to play.

We saw last week that advancing the f-pawn can leave you horribly exposed at the back if the attack stalls. Here Black has removed the knight from f6, clearing the way for the battering ram. Is it wise to advance the pawn to f5?

RB When playing against 1 d4 I will often go for a Nimzo-Indian (as in the present game) or King's Indian. I have a liking for the kind of kingside attack that can arise – with certain key differences obviously – from both defences. With the centre safely closed, Black has the option of exploiting the counter-attacking potential of his setup, and it looks to me that this is the option Vaganian is going for here: he's clearly been preparing the f-pawn push – what else would the knight be doing on h7? – so making the advance now looks logical.

However, I don't see any kind of forcing continuation after 1…f5. White, I assume, will not allow 1… f4, and play would probably go 2 exf5 Bxf5. White might then go 3 Bxf5 or 3 Ng3 – either way, Black's strategy is clear enough: double the rooks on the f-file, get the knights into more threatening posts, and perhaps, when the time is right, advance the g- and h-pawns. It's not winning, but Black looks like he has a healthy initiative.

DK 1... f5 is a good move: the pawn break liberates Black's pieces and there is the potential to attack on the kingside. Does the advance have any drawbacks? Not really. After the exchanges, 2 exf5 Bxf5 (eliminating White's bishop pair is an added bonus) 3 Ng3 Bxd3 4 Qxd3, White could play the knight into e4 to attack the d6 pawn, but that can always be exchanged off by Black's knight moving to f6.

What about White's usual queenside counterplay? Here it is limited to the b-file, and Black can cope with that easily by playing the queen to d7 and rook to f7 to defend the b-pawn. If White has no entry, Black can build on the kingside with confidence.

chess@guardian.co.uk

Black has taken White's f-pawn, leaving White exposed. Is there a way to deal with it?

Stefansson-Djurhuus, Munkebo 1998. White to play.

RB Over the last few columns we have been looking at the f-pawn as battering ram – an early, bullish advance to break open the enemy king's position with sheer aggression. But here the f-pawn has been exchanged very early – on the seventh move! There is no ram with which to batter, and the rook on f1, the usual beneficiary of any successful f-pawn push, is staring down at a solid phalanx of enemy pawns. Manoeuvring rather than force seems to be called for. But manoeuvre what where?

Looking at White's own setup, there's an obvious target for Black in the e3 pawn. Can I advance it to e4 and, with exchanges, try to rid myself of the weakness? 1 e4 Nxe4 (not 1…fxe4 because of 2 Bg5) 2 Nxe4 – oh, oh, I can already see this is not going to work out well – 2…fxe4 3 Bxe4 and now 3…Rxf1+ 4 Kxf1 (4 Qxf1 loses the knight) Qf6+ and it's horrible for White.

One way to protect the e-pawn would be to plant a knight on e6, the only problem being that e6 is covered by Black's light-squared bishop. This is as far as my thinking goes. I'm in a hole – I'll let the grandmaster dig me out.

DK Black has made an early f-pawn advance but White has it under control: it cannot move on, and that means the bishop on c8 and rook on f8 have limited scope. That's why I would not consider 1 e4 at all – why free those pieces?

As Ronan mentions, the e6 square is weak, and that's where I would focus my thoughts. The knight on d4 is perfectly placed, so why not bring a second knight into play? White can do that with 1 Nce2, threatening to pirouette via f4 into e6, and from there a knight would paralyse Black's entire position. If only Black could retreat his pawn back to f7. Now we are seeing the downside to kick-and-run chess: you leave yourself exposed at the back if the attack peters out. I hope the England football team are taking note.

chess@guardian.co.uk

Punting the f-pawn down the board may be desperate but it sometimes gets results

The Grand Prix Attack. White has just offered the f-pawn. How should Black reply?

RB We've been using the metaphor of route-one football in our survey of the f-pawn advance, and while we all know from experience that the long, hopeful punt down the park can produce the occasional goal, this looks to me just a little too early and a little too crude. The team is barely on the pitch and the ball has already been booted up the other end. On principle alone it seems wrong.

I can't decide whether Black's best response is 1…exf5 or simply to continue with a move like 1…Nge7, developing a piece and at the same time putting White on the spot over his decision to push the f-pawn so early. Does it matter? Dan will undoubtedly tell me it does. I may not be taking this seriously enough, but White's advance leaves me very unimpressed.

DK Crude, yes, but not to be taken lightly. This is the dreaded Grand Prix Attack, developed on the weekend tournament circuit in the mid-1970s, and played at one time or another by most of England's leading players. It certainly does matter how Black reacts here, and that is one of the opening's merits: at such an early stage it is hard to imagine that this choice is critical. If 1…exf5, White's simple reply 2 d3 often comes as a shock. Black might be a pawn up, but the bishop on c4 has an open diagonal towards the king, and the rook on f1 and bishop on c1 can easily enter play. Let's continue with a few more developing moves: 2…Nge7 3 0-0 0-0 4 Qe1. Suddenly Black is faced with the prospect of Qh4 followed by Bh6 or Bg5 with a huge attack.

In the initial position, the most sensible course is to ignore the sacrifice and keep developing: 1…Nge7. Then White plays 2 fxe6 and now I like the solid 2…dxe6. The bishop on c4 hits a rock on e6. The game might continue: 3 d3 0-0 4 0-0 Nd4. Black follows with Nec6 and White's queen cannot transfer to the kingside. There's no attack and no danger.

chess@guardian.co.uk

Does White have a better option than moving the f-pawn?

McShane-Cheparinov, Novi Sad 2009. White to play.

Continuing our look at the chess equivalent of route one football – the f-pawn advance . . .

RB Well, let's see what happens when we push – 1 f5. The answer comes back faster than Manchester United on the break, not much. Black is under no compulsion to take the pawn and can instead centralise with 1…Nd4 or even 1…Qd4, or start getting the queen's rook into play with 1…Rac8, and 1…Nb4, hitting the d-pawn, would be irritating.

All right, let's try to be logical. What's wrong here is that even if we could swap off pawns on the f-file, the f1-rook would remain blocked by the bishop. Let's move the bishop and clear some lines – 1 Bg4, with the idea that after 1… Nxg4 2 Qxg4 White can make the f-pawn advance with some muscle behind it.

DK 1 f5 doesn't work well here. Black should reply 1…Ne5, hitting bishop and d-pawn. After 2 Be2 Rac8 White has been pegged back and it's not clear to me how he makes further progress. I'm also not that keen on Ronan's 1 Bg4. Black should swipe the bishop, instantly reducing White's attacking potential and opening the diagonal for the bishop on g7. With 1…Nxg4 2 Qxg4 Nb4 3 Qe2 Qa6 Black takes the initiative.

But removing the bishop from the f-file is indeed the key. McShane played 1 Bd1 – a subtle idea that prefaces a brutal execution. The retreat allows the rook to emerge on the third rank, guarding the sensitive d3 pawn and preparing to swing to the h-file. The bishop will soon appear on b3, slicing down to Black's king. Black no longer has an equivalent light-squared bishop, so an exchange is impossible, making this piece – potentially – a deadly attacker.

The game only lasted a few more moves: 1…Qa6 2 Rf3 Rfc8 3 Rh3 h5 4 f5 Ne5 5 Bg5 Kf8 6 fxg6 fxg6 7 Bb3 Nxd3 8 Rf3. The switch back to the f-file is devastating. Black resigned here, unable to prevent a crushing sacrifice on f6.

chess@guardian.co.uk

Is an f-pawn advance the game's equivalent of route-one football?

Botvinnik-Vidmar, Nottingham 1936. White to play.

RB For some time I've been nagging Dan to do a series of columns on the f-pawn advance. He's been resistant, partly, I suspect, because pushing the f-pawn is a bit like route-one football, the long ball punted down the park in hopes of a quick goal. The advance can sometimes be crude, obvious and speculative, but, like the infamous route-one game, it can also produce results, and, as we will see over the next few weeks, some very classy players have used it.

We start with one of the classiest of all time, Botvinnik, in a game played in the decade before he became world champion. Having announced our topic, no points if you've guessed that White now played 1 f4. There followed 1...Rc8 2 f5 exf5 3 Rxf5. Here Vidmar went badly wrong with 3...Qd6, and, after the sacrifice 4 Nxf7, Botvinnik quickly won. 3...Rc7 would have been better but it wouldn't have helped much: White's knight on e5, bishop on b3 and rooks ready to double on the f-file will give him irresistible pressure.

Had Vidmar played 1...g6 to try to block the advance of the f-pawn, White had 2 Bh6 Re8 3 Ba4, winning the exchange, while 1...Ne4 fails to 2 Nxf7 Kxf7 3 Rde1.

DK To my eyes, the f-pawn attack looks unsophisticated and cliched. That said, I'm not above using it myself – playing the Grand Prix Attack against the Sicilian is a guilty pleasure. Here the move's merits are clear: the rook on f1 can enter the game without moving and the f-pawn acts as a battering ram to breach the enemy kingside.

When we are caught up in attacking, we tend to ignore the drawbacks of the f-pawn's advance, but they can be serious: the king is exposed along the a7-g1 diagonal, the second rank opens and the e4 and e3 squares are also weaker. In other words, if you lose control a counter-attack can be deadly, but there's no chance of that happening here.

chess@guardian.co.uk

How does White do something with this passive position?

Howell-Werle, London 2009. White to play.

Concluding our brief survey of the Scotch (1 e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Nxd4), we look at the kind of dynamic endgame struggle that can arise from this old but still important opening.

RB White's position after 1 exf5 Nxf5 2 Re1 (or 2 e4, or 2 Kd2) is so passive – and the e-pawn likely to be a long-term weakness – that White's best chance is to push with 1 e5. Assuming Black takes (1...d5 2 Na4 followed by Nc5 would give the white knight a great post ) with 1...dxe5, White has 2 Bc4, developing his bishop to another great square. If 2...exf4 3 exf4, OK White is a pawn down, but it hardly matters. With the white pieces all ready for action and Black's queenside pawns in a dreadful state, this has to be the best move.

DK 1 e5, as Howell played, is an excellent pawn sacrifice. To find it one needs to appreciate that the alternative, 1 exf5, leaves Black comfortably placed. Here's a typical scenario: 1...Nxf5 2 e4 Nh4. The f-pawn must advance to save itself, and then the knight hops via f3 into e5. Compare this knight with White's, unable to move to its outpost on d5. With the pawns blockaded, Black has stabilised the position and can begin to take control.

After 1 e5 dxe5, it would be wrong to recapture with 2 fxe5 as 2... Ng6 followed by Nxe5 gives Black more freedom than in the game. 2 Bc4 is correct, preventing Black's bishop from emerging on e6, and now 2...exf4 3 exf4 Ng6? would be met by 4 Rxg6! hxg6 5 Rd3, and there is no defence to the checkmate on h3.

If you'd like to investigate the Scotch, then Starting Out: The Scotch Game by John Emms (Everyman Chess, £13.99) will provide you with the basics, while The Scotch Game for White by Vladimir Barsky (Chess Stars, £17.99) – just published – will provide more up-to-date theory. If you prefer to study with a DVD, I recommend the The Scotch Game by Nigel Davies (ChessBase, €26.99).

chess@guardian.co.uk

Fashions come and go in chess. And this is is one that may well be on the way back

4... Bc5 is one of Black's main options against the Scotch (1 e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Nxd4). How should White continue?

RB I've done a tiny bit of study on the Scotch and so am not quite the total novice I was when we began this survey. In fact, I've reached this position in a couple of recent casual games. I've tried both of White's two principal continuations here, 5 Nxc6 and 5 Be3 (5 Nb3 is less often seen).

The first of these looks appealing for White: 5...dxc6 6 Qxd8+ Kxd8 puts an end to Black's castling rights, while 5...bxc6 messes up his queenside pawns. But Black, as I discovered to my cost in one game, has a nice move in 5...Qf6, threatening mate and forcing White into an awkward defence of f2 (both 6 Qd2 and 6 Qe2 block a bishop). I prefer the second alternative 5 Be3, developing a piece and maintaining the tension. After 5...Qf6 6 c3 Nge7 I would play 7 Bc4, at which point my knowledge of the theory runs out.

DK How one continues here is largely a matter of taste. At the London Chess Classic, McShane-Short continued 5 Nxc6 Qf6 6 Qf3 (the latest trend), and the players quickly reached an endgame in which White had a little more space and therefore a theoretical advantage. In the same tournament Howell-Adams continued instead 5 Be3 Qf6 6 c3 Nge7 7 Bc4 b6, and with all the pieces still on the board, a complex struggle lay ahead. Although this line is relatively common, I must admit to feeling uncomfortable with White: the knight can no longer move to c3 to suppress Black's ...d5 break, and I prefer to have more control.

5 Nb3 is rarely played at the top levels, but perhaps that's just fashion. I have a book from 1977 that indicates it as the strongest move. Magnus Carlsen played it four times at the World Blitz Championship in November – winning three. After 5...Bb6, he tried 6 Qe2 followed by Be3, Nc3 and queenside castling. I bet we will see more of that in the future.

chess@guardian.co.uk

The former world champion was responsible for a revival of interest in the opening. But how does he exploit it here

Kasparov-Karpov, World Championship (game 14), 1990.

More on the Scotch. Garry Kasparov was responsible for its revival. Here he finds his e-pawn under pressure. What should he play?

RB There are so many possibilities that I'm going to have to go through a process of elimination. We can discard the obvious hara-kiri options of 1 cxd5 and 1 exf6 Qb4+. Nor does 1 f4 fxe5 2 fxe5 Qg5 look particularly appealing. 1 Bg2 looks like it just loses a pawn and gives Black all the play after 1...fxe5 2 0-0, and 1 e6 is just bad. Nor do 1 Nd2, 1 Kd1 and 1 Ba3 solve the problem.

That leaves the queen. Where to? More elimination. 1 Qg4 and 1 Qh5 seem pretty pointless. 1 Qf3 looks better. Ah, wait. Black will play 1...Nb4, threatening to fork the king and rook. 1 Qd2 Nb4 2 a3 would result in Black losing the knight, and if 1...Nb6 I'd probably opt for 2 Bg2, giving up the e-pawn in return for getting my king to safety. Black gets a formidable pawn centre but at least the present crisis has been dealt with.

DK This position arises from 1 e4 e5 2 Nf3 Nc6 3 d4 cxd4 4 Nxd4 Nf6 5 Nxc6 bxc6 6 e5, the so-called Mieses Variation, one of the sharpest and most demanding lines of the Scotch. The number of options that Ronan has considered already tells us how complicated the position is.

It is tempting to try to force Black's centralised knight to move, with 1 Qd2, but after 1...Nb6 2 Bg2, instead of capturing the e-pawn, Black should play 2...Bxc4! sacrificing a piece for three pawns: 3 bxc4 Nxc4 4 Qe2 Nxb2 5 Qxb2 Qxe5+. Even in the endgame, Black has the initiative.

Instead, Kasparov developed and castled straight away: 1 Bg2! fxe5 2 0-0 h5 3 Qd2 Nf6 4 Qa5. A highly unusual position. White has given up a pawn and the centre, but his king is safe and he is banking on his attack against Black's weakened king position. Karpov did well to survive from here.

chess@guardian.co.uk

Black has made an incredibly aggressive move with his queen. What does White do?

White to play.

We continue our survey of the Scotch (which arises after 1 e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Nxd4) with a move that was the height of fashion in the 19th century.

RB On my shelves is a book on the 4...Qh4 variation of the Scotch. I wish I'd actually opened it because these early queen sorties never fail to unnerve me. I'm sure this is partly because when we first try our hand at chess we immediately reach for the queen. Then, as we grow more practised, we learn to delay her deployment until we've developed the minor pieces and got the king to safety. So this kind of aggressive invasion – on move 4! – can come as a shock. It presages sharp play and the danger of getting blown out of the water very early in the game.

I can see that 5 Nb5 is a possibility. The pawn-snatching spree of 5...Qxe4+ 6 Be2 Qxg2 would lead to disaster for Black after 7 Bf3. But Black doesn't have to take the g-pawn and could guard c7 by 6...Kd8. It's all sharper than I like, especially not knowing the theory. So I'm going to be boring – and, I hope, safe – and play 5 Nc3, and if 5...Bb4, I'll continue with 6 Be2, giving up the e-pawn in return for better development.

DK 4...Qh4 was a favourite of the first world champion, Wilhelm Steinitz, but it is telling that since the modern revival of the Scotch none of the elite players has ever risked it. I already dislike Black's position after the direct 5 Nb5, but in fact 5...Bb4+! 6 Bd2 Qxe4+ 7 Be2 Kd8 – as Steinitz played – isn't so clear.

Ronan's idea of 5 Nc3 Bb4 6 Be2 is stronger. You can't go too far wrong with simple development. After 6...Qxe4 7 Nb5, Black will again have to defend c7 with his king. Several games have continued 7...Bxc3+ 8 bxc3 Kd8 9 0-0. White has greater compensation for the pawn than in the variation above as he has the two bishops – as well as the sitting target of Black's uncastled king. You'd have to be a masochist to play like this with Black.

chess@guardian.co.uk

The Scotch Opening is still played at the highest level. But how does use it to gain control of the centre?

Carlsen-Leko, Nanjing 2009. White to play.

Over the next few weeks we're going to take a look at the Scotch Opening, which arises after 1 e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Nxd4. The Scotch is hardly new: it got its name after it was used by players from the Edinburgh Chess Club in a correspondence match against London (1824-28). But unlike some of its contemporaries from the romantic era – the King's Gambit, for example – the Scotch remains a solid and respected opening. Kasparov was responsible for its modern revival after he used it in world championship matches against Karpov, Short and Anand. With Kasparov now coaching Magnus Carlsen, it is no surprise that the 19-year-old Norwegian, the world's No1 according to the latest Fide ratings, is also using the Scotch.

The Scotch is a true "open game" with free lines for pieces and fluid pawn structures liable to change at any moment – as in this position.

RB I don't know what theory recommends. I do see that White is a pawn down, that he has an isolated queen's pawn and the cover around his king is a little loose. On the plus side, Black's pieces don't have anything like the scope of White's two bishops, or the central squares available for the white knight.

But what's the plan? I'm going to harry the queen with 1 Bd3 and after 1...Qf6 follow up with 2 Qb3 to try to exploit Black's relatively unguarded queenside.

DK 1 Bd3 Qf6 2 Qb3 is excellent for White. Not only is Black's queenside under threat, but the queen is close to being trapped by Ne4. But Black should reply to 1 Bd3 with ...Qg4 and he is closer to clearing his jam of pieces with ...Bf5.

In the game, Carlsen played 1 Bf3!, preparing to bring the knight to a great square on d4 via e2 and opening the c-file for the rook. After 1...Rac8 2 Qb3 b5 3 Ne2 Qh3 4 Nd4 Bg4 5 Bg2 Qh5 6 h4, White had a firm grip on the centre allowing him to attack the queenside pawns, and Black's pieces were still in each other's way.

chess@guardian.co.uk

The former world champion Vladimir Kramnik had a superb year. But how did he turn this game to his advantage?

Kramnik-Kosteniuk, World Cup Blitz Moscow 2009. White to play.

Apart from Magnus Carlsen's breakthrough, the major story of 2009 was Vladimir Kramnik's comeback. Since he lost the world title he looks more relaxed, his opening preparation is more thorough, and his play is sharper. This game is a good illustration of Kramnik's style.

RB This looks to me like a pretty evenly balanced middlegame, and I can find nothing – nothing – at all for White. I can't see any tactical strikes. 1 Nxf7, 1 Ng6, 1 Rxc7 and 1 Bxd5 are all clearly pointless. 1 Nc6 and 1Rc6 are equally self-defeating. How about improving the position of a poorly placed piece? I could manoeuvre the d2-knight via b1 to c3 to put pressure on the black b-pawn, but would it be worth the time? No, it's white flag time. I'm going to have to opt for something straightforward such as 1 Rc2 with the idea of doubling on the c-file. But with the c7-pawn well protected by the bishop on d6, I'm not at all sure what White achieves by this.

DK The position is so complex – all the pieces are on the board minus a pair of pawns – that I can imagine investing some time to find a decent continuation. As nothing much is on, Ronan's idea of 1 Rc2 and doubling on the c-file doesn't look at all bad. The black bishop might be protecting c7, but under the right circumstances an exchange sacrifice could work. It's also worth thinking about Black's plans. I would be concerned about the knight leaping into e4 followed by ...f6, driving our knight away, so I would consider 1 f3 preventing that. Later, the central pawn advance e4 can be considered.

But in the game, Kramnik's instinct was to play boldly. He offered a sacrifice with 1 Rc5. Black could not resist the material, but after 1...Bxc5 2 dxc5, White's bishop on b2 was reborn. The game continued 2...Nc4 3 Nec4 bxc4 4 Bxf6 gxf6 5 Nxc4 and White has tremendous compensation: he has a potent queenside pawn majority and Black's kingside is a mess.

chess@guardian.co.uk

Britain's David Howell ended up joint third in the prestigious London tournament - a real achievement

Ni Hua-Howell, London Chess Classic 2009. Black to play.

While Carlsen and Kramnik scrapped it out at the top of the London Chess Classic, there was an intriguing battle taking place in the middle order: who would emerge as the leading English player? Nigel Short crashed, Luke McShane was patchy, but Michael Adams and David Howell showed consistency, tying for joint third. For Howell, this was a breakthrough at the top level.

RB Black looks very comfortable : a pawn up, rook on the seventh rank, powerful bishop bearing down on what's left of the enemy king position, and the d-pawn one square away from queening. Let's try the obvious first – 1...Re1. White can't take the rook, which would allow the pawn to promote, so he'd probably play 2 Kg2. Now what? I can't find any tricks that would enable the pawn to get to d1 and I don't want to exchange rooks because the bishops of opposite colour ending would only give White chances to draw. Trouble is, I don't see what to do next.

Perhaps 1...Bd6 is better, threatening to take on h2 with check. 2 h3 doesn't save the pawn after 2...Rh2+, when Black would be two pawns up. Maybe this is the way to go. Forget about trying to force a quick promotion and mop up White's broken kingside pawns while his pieces are passively placed.

DK Grabbing another pawn isn't bad, but it does allow White to flap around. After 3 Kg1 Rxh3, I would take the chance to move the king off the back rank: 4 Kf2 followed by Ke3. Of course Black is winning, but some work is still required. In the initial position, it's worth asking, what exactly can White do? Precisely nothing if the black pieces stay as they are. In which case all Black needs to do is activate his king. Howell played 1...Kf8 with the simple plan of marching the king down the board. After 2 Bd1 Re1 3 Kg2 Ke7 4 Bc2 Kf6 5 h3 Ke5 White resigned. He can only watch while the king moves to c1 to force the pawn through.

chess@guardian.co.uk

The world No 1 found himself in danger of falling into a trap in his last game at the London Chess Classic. Would he spot it in time?

Short-Carlsen, London 2009. Short has just advanced c4-c5. Should Carlsen take the pawn?

Magnus Carlsen, provisionally rated No 1 in the world, hung on to his early lead to win the London Chess Classic last week. By the end of the tournament he was looking tired, and in the final game, he almost made a horrible blunder.

RB Is there a mental torture more acute than the queen-and-pawn endgame? With all the other pieces and most of the pawns gone, the queen's untrammelled range and manoeuvrability create so many possible lines that I, for one, will do almost anything to avoid getting into this kind of endgame. And here, to make matters worse, there is the imminent appearance of another couple of queens on the board.

Should Black take the pawn? Let's see what happens after 1...Qxc5. White doesn't get much after 2 Qg4+ Kf8 2 Qg7+ Ke8 3 Qg8+ Qf8. What happens if instead White takes the h-pawn – 2 Qxh4...? Nothing good after 2...Qb4+, forcing the exchange of queens: Black will win the f6-pawn, get his king in front of his own f-pawn and promote. Obviously, Short has a plan, but I can't see what's wrong with making the capture on c5.

DK At first, Carlsen also couldn't see what was wrong with capturing. Then it hit him: after 1...Qxc5, White has 2 Qg2+. If the king moves to the h-file, Qg7 is mate. And if 2...Kf8 3 Qa8 will also be mate – that's the difference between checking on g2 rather than g4. Carlsen was visibly shaken as he realised that the game had turned: not only is the c-pawn steaming down the board, but his king is also perilously placed on the back rank. But the mark of a great player is to remain cool. He found 1...h3 2 c6 a4 3 Ka2 Qd1! The Norwegian's last two moves have set up the possibility of perpetual check against White's king – a common resource in queen-and-pawn endings – and that gave him sufficient counterplay to save the game.

chess@guardian.co.uk

In this game, Vladimir Kramnik can exploit his positional dominance. But how?

McShane-Kramnik, London 2009. Black to play.

With two rounds to go in the London Chess Classic, the Norwegian prodigy Magnus Carlsen looks set to win the tournament. Vladimir Kramnik, his main rival, is in second place. In this game from round three, Kramnik displayed his refined positional understanding.

RB I've been following this tournament online, but I missed this particular game, and more's the pity because I can't find a good continuation for Black. Clearly Kramnik has the better game – the two centralised knights look very threatening – but how to convert Black's positional superiority into a winning position? 1...Nxd2 2 Nxd2 doesn't lead anywhere and just throws away one of Black's most actively placed pieces. I can't see any tactical tricks coming out of a sacrifice on c3 either. So I'm going for something crude – 1...Rab8. White won't be able to defend the b-pawn with 2 Ra2 because that would allow 2...Nxc3, and if 2 Rxa7 Qxb2 3 Qxb2 Rxb2 Black looks to me to have all the play.

DK After the game Kramnik revealed that he had investigated 1...Rab8 at length, focusing on the reply 2 Rxa7 Qxb2, and now, instead of exchanging queens he thought 3 Qf1 was the only move, eyeing up a counter-attack on f7. He was convinced that he was winning but couldn't see anything clear and was concerned that White was gaining too much activity. Therefore, in the diagrammed position, he played the pragmatic 1...h6. This is useful – it takes away the g5 square from White's pieces and gives the black king an escape square off the back rank – but is essentially a waiting move. Kramnik recognised that his pieces were so dominant that he did not need to complicate. White has no constructive plan and duly self-destructed in a dozen more moves.

With the London tournament galvanising interest in chess, the timing is perfect for How to Win At Chess (10pm, 21 December, BBC4), featuring archive footage of chess in the UK as well as interviews with leading players.

chess@guardian.co.uk

Carlsen-Nakamura, BNbank Blitz, 2009.

Black to play and win.

The London Chess Classic, the strongest tournament to be held in this country for 25 years, begins today at Olympia. Details at londonchessclassic.com. Magnus Carlsen, the brilliant 19-year-old Norwegian, is one of the favourites, but he was dealt a psychological blow when he lost to one of the other participants, US champion Hikaru Nakamura, in a blitz tournament in Oslo 10 days ago. Carlsen played the classier chess, but Nakamura proved to be the better hustler. This was the game that turned the match around. Carlsen is threatening to queen the pawn, but Nakamura, with just a few seconds on his clock, found a win.

RB: One of the first chess books I owned was Practical Chess Endings by Paul Keres, and I can still remember the simple rule Keres gave in the chapter on queen endings: the player with the queen always wins even if his opponent has a pawn on the seventh rank, no matter how far away the queen and king. Keeping in mind the well-known basic procedure of forcing the enemy king on to the square in front of the pawn and using the tempo to bring your own king a step nearer, 1...Qg7 is the right way to go about this. White obviously can't advance the pawn without losing it, and after 2 Ke6 Kd4 and Black succeeds in bringing his king closer to the action, leading to the capture of the pawn and mate. And if 2 Ke8 Kc6 3 f8=Q, Qd7 is mate.

DK: Queen against pawn on the seventh is often, but not always, a win – there are notable exceptions. With a rook's pawn and a bishop's pawn it can be problematic to force the king in front of the pawn because of stalemate tricks. Most endgame books cover this – Understanding Chess Endgames (Gambit 2009) has some good examples – and as this is a relatively common endgame, it's worth knowing the details. Here Black's king is close enough to force a win and Ronan's last variation is crucial, allowing White to queen the pawn but still force checkmate.

chess@guardian.co.uk

Black has a match-winning move if White can't find a response. But what?

Kramnik-Carlsen, Moscow 2007

Black threatens … Qh3 mate. How can White defend?

Former world champion Vladimir Kramnik and 19-year-old Norwegian prodigy Magnus Carlsen will be the favourites to win the London Chess Classic, which runs from 8-15 December at the Olympia Exhibition Centre. Kramnik has just triumphed at the Tal Memorial, while Carlsen has been training with Garry Kasparov and is now No 1 on the world rating list. The top four English players – Short, Adams, Howell and McShane – will face a stern test. Visitors are welcome: there will be live commentary and side events to play in. In this encounter Kramnik and Carlsen both show fighting spirit.

RB Scary. The king is trapped in the corner and if Black gets in 1...Qh3+ it's curtains. But White has some pressure, so what about 1 Qe6+...? After 1...Kh8 (obviously not 1...Rf7?? 2 Rd8 mate), White could go 2 Qxf5 Qxf5 3 Rxf5 Rxf5, when the mating threat is over, and, though White is a piece down, he does have two extra pawns. But while that phalanx of kingside pawns looks impressive, I'd still prefer to be a piece up. Is there anything else? 1 Kg2 seems to solve the problem at a stroke, and without having to lose a piece. After 1...Nxh4+ 2 Kh2 (or 2 Kg1, but not, clearly, 2 Kh1), Black can take a draw by repetition but, such is my lack of ambition these days, that would suit me.

DK 1 Kg2 is the simplest way to draw but Kramnik gave it one last shot with 1 Qe6+, Ronan's first variation. After 1…Kh8 2 Qxf5 Qxf5 3 Rxf5 Rxf5 he played the clever 4 Rd1, threatening mate and keeping the knight trapped. But Carlsen was up to the challenge. He played 4…g5, making room for his king and simplifying: 5 hxg5 Nxg5 6 fxg5 Rxf2. White has an extra pawn, but it is impossible to make anything of it as his king is trapped on the back rank, and a draw was soon agreed.

chess@guardian.co.uk

I am delighted to tell you that one of our books has won the Guardian 2009 Chess Book of the Year Award.

The Sorcerer’s Apprentice by David Bronstein and Tom Fürstenberg has won the prestigious prize of this English daily newspaper.

The judges, GM Danny King and novelist/chessplayer Ronan Bennet explained their choice as follows: “This is an unusual, personal, inspirational and endearing book, a mix of anecdote, musing and insight”

Our special congratulations go to David Bronstein’s widow Tatiana Boleslavskaya, daughter of grandmaster Isaak Boleslavsky who met David as a young girl, who wrote a moving In Memoriam in the book.

I would also like to thank Tom Fürstenberg, who composed the book and kindly granted us the right to publish this Revised & Updated Edition. Everyone who knew Bronstein knows how difficult Tom’s task has been.

New In Chess is proud to have published this fantastic book.

Allard Hoogland
New In Chess

When defending, capturing can take the sting out of an attack. Is that the case here?

Fedorov-Maletin, St Petersburg 2009. Continuing our theme of defence, how can White, to play, save himself?

RB What's the threat here? 1... Nxg1 2 Qxg1 gives Black nothing, except a pawn down in a queen and rook endgame. Nothing there. What else? Oh,yes. 1... Qg3. This is very bad. The only thing White could do then is 2 Qxf3, but 2...Rxf3 3 gxf3?? Qxh3 is mate.

OK, we need to stop the black queen getting to g3. Is it bonkers to take with the queen? 1 Qxf3 Rxf3 2 gxf3 and White has two rooks for the queen, which is not terrible. Let's try consider a few more moves before committing to this defence. 2...Qf4, with the obvious threat against the f3-pawn, 3 Rf1 and then 3...Qg3 wins the h-pawn. The black queen and h-pawn might be able to make some mating threats, but I think would still have some work to do.

Anything else? Why not just the obvious 1 gxf3...? I can't see what's wrong with it. The black queen is now threatened – 1...Qf4, threatening mate on h2, seems obvious, but that's seen off by 2 Rg2. Am I missing something here?

DK Generally, when defending, I first look to see if I can capture something. Capturing reduces the number of attacking pieces. Two rooks are often stronger than the queen, but in this case, after 1 Qxf3 Rxf3 2 gxf3 Qf4 3 Rf1 Qg3 4 Ra2 Qxh3+ 5 Kg1, the queen has lots of targets – the open king, weak pawns on f3 and b4, and the rooks themselves. I would only go for this if there were nothing better.

So 1 gxf3 – the best move. Ronan correctly points out 1...Qf4 2 Rg2, but the danger has not passed. Black continues 2...Rxf3, threatening ...Rf1+ winning the queen. 3 Ra1 defends, and now 3...Rxh3+ 4 Kg1 Qe3+ 5 Rf2. Black has lots of checks, but it is only good enough for a draw: 5...Qg3+ 6 Rg2 Qe3+ 7 Rf2, and so on.

In the game, White played 1 Ra2 but after 1...Qg3 2 Qxf3 Rxf3 3 gxf3 Qxh3+ had to resign as 4 Rh2 Qxf3+ 5 Rgg2 g5 leaves him completely bound.

chess@guardian.co.uk

Svidler-Jobava, European team championship, 2009. Black, to move, is under attack. How does he save himself?

Our theme over the next few columns is defence – hardly a glamorous subject, and little discussed, but of course vital.

White has just played his knight into the middle of the board, heading towards Black's king, and the rook and the queen are also in powerful attacking positions. What does Black have in his favour? Not much. But he does have a material advantage, rook for knight and an extra pawn. This gives him a little hope: if Black survives the attack, he could find himself with a winning position.

So let's concentrate on survival. First we should establish what White is threatening. To do that, imagine it is your opponent's move. There are two obvious attacking continuations: 1 Nxc6+ Ka6 2 a4 looks like it will lead to mate; and 1 Qd7, threatening mate on b7, is also strong. If Black defends with 1...Rab8, 2 Nxc6+ will win the rook and probably force mate.

Worrying, but now we know what to counter – Black to play again from the diagram. How about moving the king up the board? 1...Ka6 stops Nxc6+, but 2 Qd7 Rab8 3 Nxc6 is still grim. It looks like we can't prevent Nxc6, so what about letting the king retreat into the corner? First we have to move the rook out of the way. Finding a decent square is the problem. If 1...Rac8, for example, 2 Qd7 still wins.

What if we offered an exchange of rooks? Let's try 1...Raf8. In that case 2 Qd7 can be met by 2...Rxf7 3 Qxf7, and the exchange has relieved the pressure. And what about the other threat, 2 Nxc6+? Here the king retreats into the corner, 2...Ka8 and White has no way to follow up. If the rook attempts to avoid the exchange with 3 Re7, Black has the counter-attack 3...Qf4+, with at least a draw.

In the game, after 1... Raf8 White could find nothing better than 2 Rxb7+ Kxb7 3 Qxc6+ Ka6 4 Qa4+, with a draw by perpetual check.

chess@guardian.co.uk

The revised and expanded version of the classic The Sorcerer’s Apprentice by David Bronstein and Tom Fürstenberg has won the fourth Guardian Chess Book of the Year Award. According to the British daily it’s an “unusual, personal, inspirational and endearing work that deserves a place on every chess player’s bookshelf”.

The Sorcerer’s Apprentice was originally published in 1995 by Everyman. The new, revised and expanded edition, published by New in Chess, won Guardian’s Chess Book of the Year Award.

The book “contains a lot of new material and two moving In Memoriams, one by Bronstein’s widow Tatiana Boleslavskaya and one by his co-author and friend Tom Fürstenberg. Furthermore, 32 pages of unique photographs have been included, as well as extensive new articles by Bronstein himself,” according to the publisher. You can read our review of the book here.

The Guardian is a British daily newspaper owned by the Guardian Media Group. Founded in 1821, it is unique among major British newspapers in being owned by a foundation (the Scott Trust, via the Guardian Media Group). It had a certified average daily circulation of 358,844 copies in January 2009.

Ronan Bennett and Daniel King of The Guardian wrote:

Books from Botvinnik, Bronstein and Lipnitsky, along with Kasparov’s latest addition to his series on world championship chess, gave our shortlist for the fourth Guardian chess book of the year award a distinct Soviet-era flavour. (…)

Our winner is the revised and expanded version of The Sorcerer’s Apprentice by David Bronstein and Tom Fürstenberg (New in Chess, £29.95). This is an unusual, personal, inspirational and endearing book, a mix of anecdote, musing and insight. Bronstein, who drew his 1951 world championship match with Botvinnik 12-12 (Botvinnik, as holder, retained the title), was one of the most original players of all time.

(Read the full article here)

Links

• Read our review of The Sorcerer’s Apprentice
• Get yourself a copy of The Sorcerer’s Apprentice

Our book of the year is an unusual, personal, inspirational and endearing work that deserves a place on every chess player's bookshelf

Spassky-Bronstein, Amsterdam 1956. Black to move.

Books from Botvinnik, Bronstein and Lipnitsky, along with Kasparov's latest addition to his series on world championship chess, gave our shortlist for the fourth Guardian chess book of the year award a distinct Soviet-era flavour. (By the way, an apology here to publishers Quality Chess for the mistaken suggestion that there had been an earlier English language edition of Lipnitsky's 1956 Ukrainian classic Questions of Modern Chess Theory.) There were also, of course, contemporary works in contention, from John Nunn, Ivan Sokolov, and Herman Grooten. Each of the shortlisted titles deserves a place in any chess player's library: they are well produced, clearly presented, and the quality of instruction and analysis is extremely high.

Our winner is the revised and expanded version of The Sorcerer's Apprentice by David Bronstein and Tom Fürstenberg (New in Chess, £29.95). This is an unusual, personal, inspirational and endearing book, a mix of anecdote, musing and insight. Bronstein, who drew his 1951 world championship match with Botvinnik 12-12 (Botvinnik, as holder, retained the title), was one of the most original players of all time.

The position in the diagram is characteristic of his highly individual approach to chess. In his notes to the game he explains that he had played two five-minute games against Spassky during a visit to a junior chess club some years before. Although Spassky lost both games, Bronstein was impressed by the youngster's concentration and logic. Reflecting on his overall record against the future world champion, he said: "We have played many fine games but I never managed to win and I know why. I always tried to disrupt his logical play right from the opening. This worked against others such as Botvinnik but it failed against Spassky!"

The opening of this game certainly has the "disruptive" quality Bronstein speaks of – it could also be fairly described as unorthodox. The retreat 1...Qe7 looks sensible, but Bronstein, true to his exhortation to play with fantasy and energy, opted for 1...Nxg3. After 2 Qf2 Nxf1 3 Qxh4 Nxe3 4 Kf2 Nxc4, Black has two bishops and a couple of pawns for the queen. The idea is bold and original, and you will find many more in this wonderful book.

chess@guardian.co.uk

Purdy-Crane, Sydney 1926. Black to play

DK: If we spent as much time studying endings as we did openings, we'd all be stronger players. It's one of those irritating truisms that we suspect might be right, but is difficult to act on. The latest opening schemes, packaged as conveniently as ready meals, grab our attention before the high-fibre complexities of rook and pawn endings. It doesn't help that endgame books so often look like reference works. I own a series of worthy endgame classics by esteemed Soviet players but it has been a while since I delved into them with much pleasure.

That's why I welcome Understanding Chess Endgames by John Nunn (Gambit £15.99), which completes our shortlist for book of the year. Nunn has distilled a vast amount of detailed endgame research into clear and well-presented chunks. There are 100 short sections, each with four examples, each with a diagram, fitting neatly across two pages. For the sake of clarity, Nunn has made a conscious decision to omit distracting side variations, concentrating instead on the main message of the section. This is no-nonsense practical advice.

The position is taken from the section Reserve Tempi in the chapter on king and pawn endings. If either king steps backwards a pawn will be lost and with it the game, so whoever runs out of pawn moves first will lose. Which pawn should Black play first? Nunn states: ' . . . in a battle for tempi it is generally better to make a move with a pawn that is not on the second rank.'

In keeping with this rule, the winning move is 1…h5! and Black wins the tempo battle. For example, 2 h4 g6, and now Black blocks on the queenside: 3 a3 a6 4 b4 b5, and White's king must retreat, losing the central pawn. In the game Black blundered with 1… g5? And after 2 g4 White had secured the reserve tempo h3 on the kingside, winning the game.

We will be announcing our book of the year next week.

chess@guardian.co.uk

Lipnitsky-Geller, Moscow 1950. White to move.

Even if you're a big fan of Soviet chess history you may still not have heard of Isaac Lipnitsky. Born in Kiev in 1923, Lipnitsky was a member of the same junior chess club as his contemporary and rival David Bronstein. During the second world war, Lipnitsky fought in the battle of Stalingrad and, after his return to Kiev, became one of the strongest players in Ukraine.

Although he never became a grandmaster, he scored notable wins against future world champions Petrosian and Smyslov, and against other greats such as Keres, Geller and Averbakh. However, he is primarily remembered as a theoretician and teacher, and for his authorship of the classic Questions of Modern Chess Theory (Quality Chess, £15.99), which has made it on to our shortlist for chess book of the year. Yes, it was first published in 1956, in Ukrainian, and has been available in earlier English language editions, but we both felt that this fresh version from Quality Chess deserves inclusion.

This is very much a back-to-basics manual, reminiscent of Nimzowitsch's My System, with chapters on such themes as the centre, the flanks, mobilising the pieces and so forth. It never hurts to revisit fundamentals, and in the hands of so incisive a guide it can be a pleasure.

That said, parts of the book do have a dated feel. You may not find the chapter on modern gambits useful (Lipnitsky's idea of modern, even in 1956, included Chigorin-Steinitz, cable match 1902, and Alekhine-Fahrni, Mannheim 1914) but there is so much timeless advice elsewhere – in, for instance, "Evaluating the position" and "The concrete approach" – that the book is well worth adding to your library.

The position is taken from the short selection of Lipnitsky's games at the back of the book. White is clearly winning, and Lipnitsky found the simplest route: 1 Rh6+. After 1...Kxh6 2 Nf5+ Kg6 3 Nxe7+ Kf7 4 Nc6, Black, who will have to give up the bishop on d8, resigned.

Bronstein-Geller, USSR Ch 1961. How can White force checkmate?

The Sorcerer's Apprentice by David Bronstein and Tom Fürstenberg, made it to our shortlist for book of the year without discussion – we both loved it. It's an unusual title for an unusual book. Bronstein, former world title challenger, is the sorcerer and you are his apprentice as he takes you through a meandering selection of his games and combinations, richly annotated with advice, anecdotes and extraordinary human stories.

This is a revised and expanded version of the book of the same title that first appeared in 1995. Bronstein died in December 2006 and this new edition is a fitting tribute. Through the pages you sense a great person, a nonconformist in a Soviet state that often demanded conformity, an innocent in life but a genius on the chessboard.

It's an inspirational book. Bronstein conjured some magical performances, yet his modest and instructive annotations suggest that we are all capable of such genius. As he says: "The main advice to the reader: always play with fantasy and energy . . ." There's a story that Bronstein once gave some coaching to a team of players in the Belgian national league – and the next day they won by a record 8-0.

You find yourself desperate to sit down at the chessboard again, and that in itself is the best recommendation for any chess book.

You need a bit of fantasy to solve this wild position. There are conflagrations all over the board, but Bronstein homes in on the flaws in Black's position: "You should not put a queen out of play on the edge of the board, and you should not allow an enemy rook to the 7th rank and you should not allow a pawn near your [castled king]. You may sometimes violate one or two of these rules but all three together is just too much, even for a grandmaster," Bronstein finished the game with 1 Qg6 and Black resigned. Mate follows after 1…fxg6 2 Rxg7+ and Nxg6.

chess@guardian.co.uk

Gligoric-Nikolic, Novi Sad 1982. Black to play. Who stands better?

RB Our third shortlisted book for the Guardian Chess Book of the Year award is Winning Chess Middlegames by Ivan Sokolov (New in Chess, €24.95). Like our first shortlisted title, Chess Strategy for Club Players by Herman Grooten, Sokolov's book focuses on the middlegame, but whereas Grooten emphasises the dynamic aspects of the game, here the stress is on pawn structure. In an introduction Michael Adams makes the point that we often learn opening lines without giving serious thought to the kind of pawn structures they create.

Sokolov arranges his material into four different types of pawn structure: doubled pawns; isolated pawns; parallel hanging pawns in the centre; and pawn majority in the centre, further dividing these into subgroups. He then analyses the structures with reference to the opening. Chapter 1, for example, deals with doubled pawns arising mainly from the Nimzo. Chapter 2, on isolated pawns, looks predominantly at lines in the Queen's Gambit Declined. The essential point – but one often overlooked – is that from the opening we should be able to anticipate the structure of the middlegame.

The diagram position arose out of a Nimzo (Hübner Variation), with the characteristic doubled pawns on c3 and c4. White has tried to exploit the semi-open b-file to create threats against the enemy king. His rooks are doubled, the bishop is on b5, and the queen lurks on a3. On the other wing, Black's pawns are advancing, supported by the rooks. Who stands better?

Fritz assesses the position as roughly equal. But Sokolov is unequivocal: "A sorry sight. On the queenside White is not able to create a single threat, while on the other side of the board the battle is lost." The computer is wrong. White is dead lost. After Black's simple defensive expedient 1...Na6-b8, the game continued 2 Nf1 g4 3 f4 exf4 4 Bxf4 Ng6 5 Rf2 h4 and the pawns quickly smashed the white king's position.

chess@guardian.co.uk

Smyslov-Botvinnik, 5th game World championship 1958. Black's king is in check. Should it move up to c5 or back to c7?

DK Two books on world championship matches made it on to our shortlist for book of the year. Kasparov vs Karpov 1986-1987 (Everyman Chess, £30) written by Garry Kasparov, is an automatic choice. This is the latest volume in the former world champion's monumental series, and this time he dissects the matches in London/Leningrad 1986 and Seville 1987.

Kasparov's detailed analysis of the games is admirable but I skimmed them and just read the story – it's gripping. Gorbachev had just come to power and was implementing his policies of glasnost and perestroika in the face of conservative opposition. In that context the result of these matches had enormous significance: Kasparov was the outspoken outsider, Karpov the loyal communist. Which image would the Soviet Union be projecting to the world? Kasparov alleges – and backs up with strong evidence – that there were spies in his camp passing information to Karpov, backed by the KGB. Some of the episodes could have come straight from the pages of a le Carré novel.

Botvinnik-Smyslov, Three World Chess Championship Matches: 1954, 1957, 1958 (New In Chess, €28.95 ) is another reminder of a great chess rivalry. The annotations are mainly by Botvinnik and are characterised by his typical "objectivity" (read harshness). These notes were, of course, written in the pre-computer era, which means fewer variations than many contemporary books. That's a relief. I'd rather have a few well-chosen words than blocks of indigestible moves.

In the position above, Botvinnik was short of time and played the reflex 1… Kc5, following the general rule that kings should be as active as possible in the endgame. But Smyslov replied with 2 Kd3, and checkmate with b4 was unavoidable. Black should have tried 1…Kc7, and he still had chances to save the game.

}chess@guardian.co.uk

Piket-Van Baarle, Amsterdam 1984. White to move.

RB Over the next six weeks Dan and I will be announcing our shortlist for the fourth Guardian Chess Book of the Year award. I start today with Chess Strategy For Club Players by Herman Grooten (New in Chess, €25.95).

According to Grooten, the book is for "ambitious club players and tournament sharks". Often such books follow a tediously familiar pattern. The author selects a number of fairly recent games from a database, annotates them – with Fritz lurking in the background – throws in a few diagrams and groups them in whatever thematic arrangement suits his ostensible purpose.

Grooten's book is different. The focus is on strategy rather than tactics. It's about what happens when the opening is over. At certain points, depending on the type of position, concrete calculation will be needed. But in a substantial part of the game there is nothing concrete to calculate. What do we do then?

Obviously, we try to improve our position while creating weaknesses in our opponent's. The beginner will usually attempt improvement by mounting an attack. But, even for more experienced players, finding a good plan once the opening phase is concluded is not easy.

When I looked at the position in the diagram – Grooten's challenge is to "find a plan for White" – my eye instantly fell on the tactical try 1 Bxh5, with the obvious idea that if 1... gxh5 2 Qd1 with an excellent attack for White. Even if Black declines the sacrifice by 1... Nf8 White simply plays 2 Be2 with a pawn to the good.

This seems to me a valid line, but in the game White found 1 Nh1. What on earth is that about? The answer is strategy. White takes advantage of Black's powerlessness to improve the position of his knight, bringing it to g5 via f2 and h3. Once there, all White's pieces are poised to mount an irresistible attack. If you like the look of 1 Nh1, this is the book for you.

chess@guardian.co.uk

Javakhishvili-Wang Yu, Ningbo 2009. Our position this week is more complex. Black has just played ...g6. Where should the White queen go?

RB The capture 1 Qxg6 is of course suicidal – 1...Rg8 and the queen comes off the board. 1 Bxg6 is not much better – 1...Re5 and, after the queen moves to safety, 1...Rg5+ and White's bishop is lost; in fact, here 1...Bg4 is probably even better for Black. Either way, taking the g6-pawn is obviously out.

So where else can the queen go? What about 1 Qd5, moving out of harm's way and simultaneously threatening the bishop? The trouble is that Black has 1...Qg4+, and already I can see the loss of the exchange after 2 Kf1 Qd1+.

OK, process of elimination: we have the squares c5, d1, e2 and f3. I don't like e2 because of the rook on e8. 1 Qf3 Qxf3 2 Bxf3 Bxb5, and Black's extra pawn and better pawn structure don't bode well for White. 1 Qd1 also gives up the b5-pawn and Black can also play 1...Re5 with threats against the exposed king. None of this looks good. The more I look at it, the more I think it essential to keep the queen on the fifth rank – 1 Qc5 protects the pawn and supports the loose rook on c2.

DK That's exactly how the player of the White pieces must have thought in the game: 1 Qc5 was played. The queen seems to hold everything together, but at the same time threatens a nasty check on d4. So Black took a safe way out: 1…Qg4+ 2 Kf1 Qd1+ 3 Kg2 Qg4+, forcing a draw. But she missed a win: 1…Rxe4! 2 dxe4 Qg4+ 3 Kf1 Qd1+ 4 Kg2 Bh3+!! (a familiar attacking motif, but no less beautiful for that) 5 Kxh3 Qf3+ 6 Kh4 g5+, and White has to give up the queen to prevent mate.

The lesser evil was to head for an ending. 1 Qf3 is White's best move, keeping the king protected. Although Black wins a pawn with 1… Qxf3 2 Bxf3 Bxb5, White's active pieces compensate. After 3 Be4, threatening Rc7, White should draw.

chess@guardian.co.uk

Rematch could attract 10 million viewers online

Silver hair and sharp suits have replaced the fresh faces and Soviet tailoring of 1984 but there was no mistaking the aura surrounding two of the world's greatest chess players, Garry Kasparov and Anatoly Karpov, as they prepared for an historic rematch today.

The two grand masters, who fought an epic psychological war when they battled for the world chess crown 25 years ago, will start their rematch in the Spanish city of Valencia tomorrow.

Both promised to do their best to match the intensity and drama of the gruelling 1984 encounter that was controversially called to a halt without a clear winner after five months of play.

While both men admit they are now past their playing prime, they have been preparing their encounter for weeks. "Their prestige it at stake," explained the organiser, Basilio López.

They travelled to Valencia with support teams of four or five people for a match that organisers said would have an internet audience of as many as 10 million.

The two players were holed up tonight in separate hotels in the Spanish city preparing for a series of blitz and semi-rapid games to be played over four days.

"We are both still capable of playing high-quality chess," said Kasparov, now aged 46, who stopped playing competitive matches nearly five years ago.

"If Kasparov still played competitively he would be the best or close to it," agreeed Karpov who, at 58, has continued to play despite dropping out of the top 100. Bookies were backing Kasparov, believing his opponent's age would slow him down.

If their previous encounters were portrayed as a battle between the old Soviet Union and thrusting young reformists their meeting has already been coloured by Kasparov's political battle with Vladimir Putin's Russia.

Kasparov lashed out today at a journalist from a Russian broadcaster, which he accused of not reporting the moment he was arrested and jailed in 2007 for leading an anti-Putin demonstration in Moscow.

"I'll answer in English first so that you can see how they twist my words later," Kasparov snapped in words reported by the Spanish press. More than two dozen Russian journalists were in Valencia.

Kasparov had praise, however, for his opponent – who was seen during their encounters in the early 1980s as the face of Soviet communism. "Many of my friends forget about me when I was jailed," said Kasparov. "Karpov tried to visit me."

Kasparov had alreadyturned up the heat with declarations to El Pais in which he accused modern Russia of being "a corrupt dictatorship disguised as a democracy".

Karpov admitted that the two men disagreed politically and remained different in many other ways. "As human beings we are totally opposite in every area of life," he told Valencia's Super newspaper.

Karpov has been in training with his coaching team for the past two weeks. Kasparov has turned to youth to invigorate him. He has been training with Magnus Carlsen, an 18-year-old chess phenomenon from Norway.

Kasparov said he had age on his side but admitted that the lack of match practice might work against him. "That makes me wonder whether I will be able to recover my instinct for the game," he said.

"Normally you start playing more slowly once you have passed the age of 50," Karpov said. "But I won when we last played in New York seven years ago."

Both were critical of modern chess, blaming the people who run the sport internationally for its decline. "In my day we were more concerned about the beauty of the game," said Karpov, who hoped computer chess and the internet would return the game to its glory days.

Valencia is where the modern rules of chess were born in the 15th century when the queen – reputedly modelled on Isabel of Castile – was introduced to the board.

Strategic Kremlin man and his younger, louder foe

Small and slightly built, hesitant in public, Anatoly Karpov always seemed an unlikely champion. He got off to a difficult start, in 1975, when he inherited the chess crown after Bobby Fischer refused to defend his title. Karpov's riposte to those who sneered that he was champion in name only was to play in almost every top-flight tournament over the next 10 years, thrashing his rivals and setting a new record for cons ecutive tournament victories.

In the west, Karpov was seen as the Kremlin's man, a colourless, loyal communist. Rivals accused him of enjoying an unfair advantage, and certainly he benefited from the Soviet chess machine. His style is quiet and positional. He is primarily a strategist.

In 1984, Karpov's challenger for the world title was the 21-year-old Garry Kasparov. For one so young his self-belief was astonishing, and he was possessed of a truly ferocious competitive spirit. Brash, opinionated and impatient, Kasparov did not come into this world to suffer fools, delay or defeat. Born in Baku, Azerbaijan, he portrayed himself as an outsider in the Soviet system. Cold war warriors claimed him as the west's fighter in the ring, even if their man was a Soviet citizen and member of the communist party.

In their epic first match – KK1, as it is called, ran to 48 games – Kasparov's attacks foundered on Karpov's fine positional understanding. When they sat down to play their 32nd game, Karpov was leading 5-0 (the winner was the first to win six). Kasparov scored his first win. At 5-3 FIDE, the chess world's governing body, stopped the match. Kasparov, scenting victory, was outraged. He won KK2 the following year.

Like Wenger and Ferguson, the two Ks have traded their fair share of insults over the years. But they have also acknowledged each other's greatness, and when Kasparov was imprisoned in 2007 for his anti-Putin campaigning, Karpov attempted to visit him in jail.

Kasparov is acknowledged to be the greatest player of all time. Karpov cannot be far behind. For six years after their first encounter, these two great gladiators played more than 140 games in their world championship battles: Karpov won a total of 19, Kasparov 21, with the remainder drawn.

The games in Valencia will not match the brilliance of their previous encounters. They are in any case a mix of "rapid" and "blitz" (great chess needs time), and chessplayers are past their best by 40. Kasparov (born 1963) retired from competitive play four years ago, while Karpov (born 1951) has fallen out of the top 100. Even so, fans will relish the prospect of seeing two great foes once more facing each other over the board. Ronan Bennett

We thought that with our latest relocation it was time to reintroduce ourselves and to remind readers that this is a different kind of chess column. We don't bring you the latest tournaments (the internet does that a lot better), or annotate games by the greats (better left to anthologies and magazines). Instead, we try to provide the enthusiast and the club player, those for whom chess is a hobby rather than a profession, with useful advice and exercises in the form of a "master-student" dialogue between an average player (RB) and a grandmaster (DK).

Some of the positions we look at involve fairly high levels of chess understanding, but we also like to explore the kind of things ordinary players might encounter. Today's position falls into the latter category, and kicks off a series of columns themed around the question of finding the best square for a threatened piece. So where should the queen move to?

RB This looks rash, but since the queen is out we might as well go for it: 1 Qxc5. I'm expecting either 1...Nxe4 or 1...e6.

DK If you play the queen out so early, you are either very good or very bad. It's a beginner's ploy, vainly hoping for a quick checkmate. But if your opponent has an ounce of nous, the queen will be beaten back and you will have merely lost time. Ronan decides that he may as well grab a pawn, but Black recaptures, 1…Nxe4, and attacks the queen again. If 2 Qe3, Black plays 2…d5, staking a claim in the centre, and already has the more promising position.

However, if one is very careful, it is possible to play so outlandishly. The rising American star Hikaru Nakamura has made a speciality out of this shock tactic. Instead of taking the pawn, he has tried 1 Qh4. Now it is harder for Black to push the queen around, and if he castles on the kingside, the king could come directly under fire. But perhaps one needs to have the talent of a prodigy to make this work. The old rule of developing knights and bishops before anything else should still apply to most of us.

chess@guardian.co.uk

This position arose after 1 e4 d5 2 exd5 c6 3 dxc6 e5. Should White, to play, grab a second pawn or play more prudently?

RB I've never seen this before. It looks reminiscent of the Danish Gambit, with colours reversed (and a lost tempo for Black). It can't be good, but how to proceed? Over the board I'm sure I'd be satisfied with a one pawn advantage and play 4 Nc3, but, looking at it more closely, why not just take the pawn, 4 cxb7…? After 4…Bxb7, Black's bishop is on the long diagonal but White is two pawns up. Now, of course, White needs to develop quickly and get his king to safety, so I'd probably go 5 Bb5+. It doesn't make any sense for Black to block with the bishop he's spent so much material on developing, so 5…Nd7 (or 5…Nc6). Now 6 Nf3 and if 6…e4 7 Ne5 when it looks to me like White is fine.

DK I had this position – with White – in a recent online blitz game. What a pain. I generally play online when I want to chill at the end of a working day, playing on strategic auto-pilot, so an unbalanced situation like this is not to my taste. Of course, this is one of the most useless gambits one could ever hope to face, and I would expect to destroy it in a serious tournament game. But all moves had to be made in three minutes, and there was no time for calm reflection. I ducked the challenge: 4 Nc3 Nxc6 5 Nf3 Bc5 6 d3 f5 7 Be2 Nf6. Black had good development and my position was passive – and in blitz the extra pawn is irrelevant.

With hindsight, as Ronan indicates, 4 cxb7 Bxb7 is the best way to proceed, even though it brings out Black's bishop. After 5 Bb5+ Nd7 6 Nf3, I think I was concerned that Black would try 6…Qc7, looking at castling queenside, but 7 d4 opens the position in White's favour. If Black blocks the check with 5…Nc6, the same recipe can be followed: 6 Nf3 Qc7 7 d4 and if 7…0-0-0 White can hold firm with 8 c3. In both cases, White has free development – and that's perhaps the most important thing in blitz chess.

chess@guardian.co.uk

Lutz-Wahls, Biel 1992.

The Scandinavian Defence does not enjoy a high reputation in top-level chess, but below that it is worth a shot. There are good reasons for trying it: after 1 e4 d5 there's no avoiding the opening, and you will invariably find yourself with a familiar pawn structure – like the one above. And if all goes according to plan, you could reach an endgame as good is this. There are no weaknesses in Black's pawn structure, the knights are menacing, and the rooks aim like cannons at the d-pawn. How should Black, to play, proceed?

RB The more I look at the position, the less convinced I am that c4 was a good move for White. Obviously he wants to shift the knight from its dominating central position, but by chasing it from the centre he drives it to another equally threatening post. So where to put the knight? The move that appeals to me is 1… Nde7 (1… Nb6? runs into 2 Ba5), from where the knight will jump to f5, adding to the pressure on d4 (I don't see anything wrong with 1… Ndf4 , except that I'd rather have a knight on f5 than f4). Since White doesn't have enough have support behind the d-pawn to force its advance to d5, his pieces are going to be tied down to passive defence.

DK 1… Nde7 is certainly the best move: from f5 the knight hits both weak pawns – h4 and d4. The game continued 2 Rd1 Nf5 3 Be3 e5! exploiting the pin on the d-file. White had nothing better than 4 R3d2 exd4 5 Nxd4 Nxd4 6 Bxd4 c5 7 Be3 Rxd2 8 Bxd2 Rd4. Black wins at least two pawns. This is a typical scenario in the Scandinavian: White felt the need to drive away the knight from d5, but it left the d-pawn fatally weak.

There isn't a huge amount of theory on the Scandinavian. Everyman Chess is publishing The Scandinavian by Jovanka Houska in the Starting Out series in October. And I would recommend John Emms's book on the opening by the same publisher. Scandinavian expert Curt Hansen has authored an excellent CD containing a database and annotated games (chessbase.com).

chess@guardian.co.uk

Hjartason-Hansen, Reykjavik 1995. Black to play.

RB I'm looking for possible threats against my queen, always something to watch out for in the Scandinavian, and I don't like that its retreat squares on the short a5-d8 diagonal are blocked by the knight on b6. That said, I can't actually see a concrete threat. For example, if I play safe with 1... 0-0, what has White got? 2 a3 Bxc3 3 Bxc3 Qb5 and the queens come off. But so what? I might not be winning but I don't think I'm losing. 1... Nbd5, facilitating the queen's retreat and blockading the d4-pawn, looks okay: after 2 Nxd5 Nxd5, White's light-squared bishop is blocked in (or has to take). I can't really see anything else. I'd probably opt for castling and follow it up with Nbd5.

DK This is a complex position. Black has a number of options, and it isn't clear which is best. Like Ronan, my first thought here would be to bring my king to safety: 1…0-0. However, my second thought would be, what about 2 Nh4…? Exchanging off this bishop is a significant step forward for White: he can then play with a freer hand in the centre. 1… Nbd5 runs into something similar: 2 Nxd5 Nxd5 3 Nh4, with the idea 3… Bg6 4 Nxg6 hxg6 5 c3 Be7 6 c4. Once Black's knight is pushed from its central position, White takes the initiative.

To address this problem, Curt Hansen, a Scandinavian specialist, played 1…Bxc3. Black gives up bishop for knight voluntarily, but after 2 Bxc3 Qb5 he was fighting for the key squares c4, d5 and e4. Black's queen also makes way for the a-pawn to hassle White's bishop. The game continued 3 Qxb5 cxb5. The pawns might be doubled, but Black has taken control over the c4 square, and that means he controls the d5 square. The game continued 4 Ne5 a5 5 a3 Be4 6 Rhe1 0-0 7 f3 Bd5 8 Bxd5 Nfxd5. With the knight settled on d5, Black was ready for a queenside pawn storm. This position shows all the advantages of the Scandinavian: Black's pawns form a barrier that protects the king, and prevents any incursion from the enemy pieces. White has no counterplay while Black attacks on the queenside.

chess@guardian.co.uk

After the typical Scandinavian opening moves 1 e4 d5 2 exd5 Qxd5 3 Nc3 Qa5 4 d4 Nf6 5 Nf3 c6 6 Bc4 Bf5 7 Bd2 e6 8 Qe2, Black has snatched a pawn with 8… Bxc2. Can he get away with this?

RB This looks very suspect. On general principles alone – king still in the centre, enemy queen on the e-file – I would never try this pawn grab. But there are also compelling concrete reasons for not doing so. 9 d5 exploits the pin on e6 and the exposed position of the black king. The discovered threat against the queen may not be very subtle, but it severely embarrasses Black. If 9... cxd5, White continues 10 Nxd5 and even without spending a lot of time calculating the variations I can already see that the bishop on c2 is in serious trouble: 10... Qc5 11 Nxf6 gxf6 12 Rc1 threatens another nasty discovered attack against the queen if the bishop moves, so 12...Qf5 is the last try to save the threatened piece. But this is met by 13 Nd4, when White is winning. Dan may tell me that Black has something better than 9...cxd5, but even if he does it wouldn't warrant the pawn grab, in my mind at least.

DK In the Scandinavian defence, with the queen already in play and lagging slightly in development, Black must play cautiously. 8…Bxc2 would be reckless in the extreme and Ronan has pointed out the clear refutation with 9 d5. Let's just complete the variations. Black has to play 9…cxd5 or White crashes through on e6. After 10 Nxd5 Black could try 10…Qd8, but White wins easily with a direct attack: 10 Bb5+ Nc6 11 Nxf6+ Qxf6 12 Qc4 hits the bishop and the knight; or 11…gxf6 12 Rc1 Bg6 13 Rxc6 wins.

Instead of grabbing the pawn, Black should play 8…Bb4, shielding the queen from a discovered attack and preparing to castle on the kingside. At some point over the next few moves it is highly likely that Black will have to give up the bishop for the knight on c3 and retreat his queen, but getting rid of this knight can give Black increased control over the d5 and e4 squares.

chess@guardian.co.uk