CWAP Game 4: From-Steinitz, Paris 1867

Steinitz-From, Paris 1867, isn’t an ideal game for this project, in which I want to provide instructive examples of masters defeating amateurs who are castled behind the K3-KB2-KN3-KR3 pawn structure.

For one, From was a master (his name is on the promising gambit 1. f4 e5 2. fxe5 d6). Games of master vs. master include few mistakes, and the mistakes are of relatively little consequence. From didn’t handle the sharp opening well, and never reached equal; he didn’t make any amateur mistakes. But bonus feature! Annotated move-by-move from move 7! (That’s the extra attention I wanted to give my writings about the 1931 Capablanca-Euwe match, and I just forgot which project I was working on.)

Two, the game is sort of over by the time the Common, Weak, Airy Pawn (CWAP) structure arises, though you can see how easily Steinitz tears through it after it forms.

I could only find a few games where Stenitz played against the subject pawn position. There just aren’t enough master vs. amateur games in circulation. Imagine how much richer chess literature would be if every game in a simultaneous exhibition had to be turned in; there are surely very many instructive and beautiful exhibition games we never saw because the losers didn’t want to show them.

[Site "Paris"] [Date "1867.06.20"] [White "W. Steinitz"] [Black "M.S. From"] [Result "1-0"] { Click or tap on a move in the game text for a popout display board.[#]} 1. e4 e5 2. f4 { Grandmaster Bronstein said every chessplayer should have a love affair with the King’s Gambit, and the way love affairs work is that some are very fleeting, while some are lifelong. I’ve written so extensively from this position that it’s easy to see which way I went.} 2... d5 { The first thing to understand about 2. f4 is that 3. exf5 is an unreal threat (3. exf5? Qh4+ and White’s in trouble), so Black is free to play at will at move 2. 2…d5 is the right move to play, but since it usually entails a loss of material for Black, average players will stick to their natural inclination to capture whatever isn’t nailed down and play 2…exf4. In a purely technical sense, 2…exf4 gives Black a winning position, but White shouldn’t resign right away, which is what makes things interesting.} 3. exd5 e4 { 3…exf4 is also good, and I’m fond of Nimzovich’s move 3…c6. The trouble with 3…c6 is when White doesn’t feel compelled to play dxc6, enabling Black to develop with the recapture. (See <a href="https://www.chessgames.com/perl/chessgame?gid=1075588" rel="noopener noreferrer" target="_blank">Schulten-Morphy, New York 1857.</a>.} 4. Nc3 { 3…e4 makes it difficult for White to develop his king knight, so White usually aims to knock the e4-pawn from the board. Experience shows that a ….Bf8-b4 should be avoided, so 4. d3 is the usual move.} 4... Nf6 5. Qe2 { The game that everyone should know — even if they never intend to play the King’s Gambit as White or the Falkbeer Countergambit as Black — is <a href="https://www.chessgames.com/perl/chessgame?gid=1075588" rel="noopener noreferrer" target="_blank">Schulten-Morphy, New York 1857</a>. 5. d3 Bb4 6 Bd2 e3!.} 5... Bc5 { Because the pin on the e1-a5 diagonal is promising, Black should hold this bishop in reserve if he can, and play 5…Bf5.} 6. d3 { White can breathe a little easier since Black won’t forfeit a move with …Bc5-b4.} 6... O-O 7. dxe4 { With a bishop on b4, Black would happily play …Nf6xe4.} 7... c6 { Like I said, the drawback to …c7-c6 is that if White refuses d7xc6, then the pawn is just in the way. I suggest 7…Bf5 as development with a threat, while Stockfish says it’s perfectly OK to spend an entire additional move on …Bc5-b4 (see, it was that important at move 5 to hold back this bishop).} 8. d6 { Akin to declined gambits like 1. d4 d5 2. e4 dxe4 3. Nc3 Nf6 4. f3 e3, and 1. e4 c5 2. d4 cxd4 3. c3 d3, where White captures with a developing move without loss of a pawn. This position is different, because Black runs into a pawn fork after recapturing on d6.} 8... Ng4 { Useful move, threatening to drop a piece on f2 while preventing a fork at e5.} 9. Nh3 { Best.} 9... Qxd6 { Black’s problem now is queen exposure.} 10. Bd2 { The only safe square for bishop development also screens the d-file to enable O-O-O.} 10... b5 { You can’t write this off as having no chance for the usual reasons (less center control, less development), because opposite-wing attacks turn primarily on time. Black’s lag in development won’t look so bad if the pawn rush results in clues for the pieces’ best squares. For now, the bishop can’t develop further than d7 where it blocks the knight (10…Be6 11. f5 separates bishop from knight), and neither move by the queen knight feels progressive.} 11. O-O-O { Naturally, but White can’t feel entirely comfortable for “castling into it”.} 11... a5 { In for the penny, in for the pound. Games of opposite-wing castling are scary for the side with traditional advantages, because the usual value judgments are discounted. Time is everything, so on one hand, you’re thinking ‘I should shore up my king position, then let my traditional advantages in center control and development tell’, while on the other you’re thinking ‘If I spend too much time on passive defense, then my leads in center control and development are meaningless, and boy, will it look dumb if I lose.’ } 12. e5 $1 { That’s what White needed, a method for exploiting the traditional advantages. White’s central lead grows, while e4 comes open for the knight.} 12... Qc7 { Leaving e7 open for the bishop. …Qd6-c7 oughtn’t be considered for watching the a5-pawn, because Black wants to lose that pawn.} 13. Ne4 { A commanding move. The knight has many promising choices from here: g5, f6, d6, c5 all have merit.} 13... Be7 { Purdy said: Prefer moving an attacked unit to guarding it. 13…Na6 protects the bishop, but where’s the knight going from a6 if White doesn’t capture c5?). } 14. Nhg5 { 15. e6 next would cut off the knight’s guard, and threaten e6xf7. Then …Ng4-h6 would be forced.} 14... Nh6 { Purdy said: If you know you’re going to make some move next, or soon, it almost never hurts to play it now. } 15. Qd3 { One, you saw the threat 16. Nf6+ with mate next. Two, you recognized the famous piece configuration from Reti-Tartakover New York 1910.} 15... Bf5 { Black can resign following} (15... Nd7 16. Nf6+ Nxf6 17. exf6 Bf5 18. fxe7 Bxd3 19. exf8=Q+ Rxf8 20. Bxd3 { , while } ) (15... Bxg5 16. Nxg5 Bf5 { is a little better.} ) 16. h3 { The time element that governs opposite-wing attacks has turned wholly in White’s favor, given g2-g4 is coming.} 16... Na6 { Black is at a loss for constructive development, and is reduced to the only development he can make. A rook is must go next to d8 to chase the queen from the d3-h7 diagonal, and 16…Na6 was the only knight move that keeps the file clear.} 17. g4 { Development is equal, and the pawn structures somewhat mirror each other, but you can see how far ahead White is in the race. Compare the knights, and the piece coordination. White’s material advantage is as slight as can be — one pawn — but it’s the e5-pawn, which rules the board.} 17... Rfd8 { Expected.} 18. Qf3 { The queen move that watches the knight, and doesn’t block a bishop.} 18... Bxe4 { The positions that you can’t resign, but in which you have no good moves are the worst. 18…Bxe4 lessens Black’s possible resistance on the white squares, but is dropping back to d7 or c8 at all productive? } 19. Qxe4 { On a board with many pieces on it, you don’t expect one queen to be so much mobile than the other.} 19... g6 { The first crack in the castled position.} 20. Bg2 { When we’re beginners, we’re told that the development should be complete by move 10 or 12 or something. Then we learn that this applies to positions in which nothing more important is going on. It took 20 moves for White to connect the rooks, but there was always a matter more pressing than moving that bishop. Now it’s the only light-squared bishop left, on the longest diagonal. } 20... Rac8 { If Black anything better to do, he would’ve left the c6-pawn to be captured, to trade his inactive queen for the active white queen, and make use of the c-file.} 21. Qf3 { Tidy little move. …Na6-c5 is coming, so it doesn’t hurt to move the queen. From f3, she watches g4, so the h3-pawn is free to advance.} 21... Nb4 { Not a real threat.} 22. h4 { I don’t care what Stockfish says about 22. h4 as one-half a pawn better than 22. Kb1. It’s the idea that counts; sacrifice anything but time in games of opposite-wing castling. 22. h4 saves time, while 22. Kb1 means one-half pawn to a computer that bases its evaluation on a position 13 moves down the road. } 22... Nxa2+ { Amazingly, Stockfish changed its mind. Now Stockfish goes for 22…c5, and I’d want a student to play that. If Black has any hope, it’s for his attack to involve the heavy pieces on the c-file, while 22…Nxa2+ is still a one-piece attack. The funny thing is that if Stockfish were a human, you could say: “What’s the deal? One-half move ago, you thought White should guard that pawn with 22. Kb1, and now you’re saying Black shouldn’t capture it?!”. Then your friend would rattle off 13 moves of analysis to show the difference, but it’s silly to think that the other player would follow that 13-move analysis. Humans today are really dumb about debating computer analyses that differ by thousandths of a pawn dozens of moves ahead. We truly only need to see the length of a double threat: Our double threat, their reply to one threat, our execution of the other, and then ask if that’s really what we want. This is why the 1951 book How to Think Ahead in Chess had the subtitle “How to See Three Moves Ahead”, because the masters who wrote it knew that’s all practical, human chess requires. } 23. Kb1 { The only move.} 23... Nb4 { If Black had genuine counterplay, he’d think about leaving the knight to be captured, to save time for his advance, while White would lose time capturing and retreating.} 24. h5 { To attack the enemy king, open files for the rooks.} 24... Bf8 { Otherwise 25. hxg6 makes a double attack. } 25. hxg6 { The only consistent move.} 25... hxg6 { Finally, the pawn structure we want to talk about! It took 25 moves for White to coerce it. Let’s say it was for master-class resistance.There are some positions in which the defender might consider …fxg6 to open the second rank for defense, but in this case, it invites Nxh7 plus g5, or e5-e6.} 26. Ne4 { Menacing 27. Nf6+ Kg7 28. Rxh6.} 26... f5 { Making luft.} 27. Nf6+ { Flushing the king into the open.} 27... Kf7 { Otherwise 28. Rxh6.} 28. Bxb4 { That’s an alert right hook, left uppercut. The knight is captured so it can’t hop back to interpose at d5.} 28... Rxd1+ { And White doesn’t mind shifting his rook away from the h-file because that file is relatively more secure for Black. } 29. Rxd1 { Forced.} 29... Rd8 { Also} ( 29... axb4 30. Qb3+ Kg7 31. g5 Nf7 32. Qh3 {+- . }) 30. Qb3+ Kg7 { Or } (30... Rd5 31. Bxd5+ cxd5 32. Qxd5+ Kg7 33. Ne8+{ +- . }) 31. Bxf8+ 1-0 * You must activate JavaScript to enhance chess game visualization.