Chess tactics

[Christopher Kreuzer]

This doesn't quite fit in the 'Best of Chess' section which is for the very best games and spectacular tactics, but is a curious case of chess 'blindness' or oversight, and may be of interest.

While playing a game outdoors recently (not on a chess table!) I had reached the following position a pawn down:

r2r2k1/pb3pbp/1q2p1p1/2p5/Q1P1P3/8/PP1N1PPP/R1BR2K1 b - - 3 18

Black to move. I played Bxe4 (better moves are available) thinking that if the knight took the bishop, I would be able to mate by taking the rook on d1. To my horror, I realised that this rook was protected by the queen on a4, so after knight takes e4 I played Bd4 and duly ended up in a lost position.

Can you see what I had missed and why?

[Christopher Kreuzer]

Here is another test of chess tactics. Simple once you see the key move (the first move is obvious), but surprisingly difficult to see if you are not aware of the right way to play the most forcing continuation.

r1bq1r2/1p2ppk1/p2p1n2/7p/P3P1p1/2N1B3/1PPQ1P2/R3K2R

This is also a presumably rare example of a position not that far out of the opening, with both armies still mostly in their own halves of the board, and where the position is deceptively decisive. It would probably be difficult to construct a similar position where it is a forced mate in less than 10 moves. In this case, with White to move, it is forced mate in 9 moves.

Can you find the mating sequence?

[Roger de Coverly]

Can you see what I had missed and why?

If there's something spectacular, I'm not seeing it either. In compensation for the pawn, you have a Bishop pair and more activity, therefore no immediate need to worry about beiong a pawn down. The "no thought" move would be .. Rd4 with twin threats of being able to take on e4 and doubling or tripling on the d file with .. Rad8 and maybe .. Qd6.

[Joey Stewart]

I think you can get your knight back with Qc6! after he takes it - he can't take the queen because of rxd1mate, and if he tries to retreat to c2 you can capture on E4 because the same tactic is still on

[Christopher Kreuzer]

Joey got it. Sorry, Roger, I should have been clearer where you needed to start finding alternative moves, but that was maybe part of the puzzle!

When I saw this, I thought it did not work because White can throw in Rxd8 with check. I thought to myself, if only I had moved my a-rook to d8 instead of the f-rook, then the tactic would work. I had missed that the mate threat is still on after the exchange of rooks on d8. You don't have to capture a piece to deliver mate... oh well.

[Roger de Coverly]

When I saw this, I thought it did not work because White can throw in Rxd8 with check. I thought to myself, if only I had moved my a-rook to d8 instead of the f-rook, then the tactic would work. I had missed that the mate threat is still on after the exchange of rooks on d8. You don't have to capture a piece to deliver mate... oh well.

Sometimes when you analyse too far, you can talk yourself out of a decent move. In this case you just have to have seen .. Qc6.

I have a theory that when you have the pieces on good squares it doesn't matter so much if you don't see the tactics arising that far in advance. The theory loops on itself as quite often the best squares depend on the available tactics.

[LawrenceCooper]

Here is another test of chess tactics. Simple once you see the key move (the first move is obvious), but surprisingly difficult to see if you are not aware of the right way to play the most forcing continuation.

r1bq1r2/1p2ppk1/p2p1n2/7p/P3P1p1/2N1B3/1PPQ1P2/R3K2R

This is also a presumably rare example of a position not that far out of the opening, with both armies still mostly in their own halves of the board, and where the position is deceptively decisive. It would probably be difficult to construct a similar position where it is a forced mate in less than 10 moves. In this case, with White to move, it is forced mate in 9 moves.

Can you find the mating sequence?

It may not be forced mate if black gives up his queen but it's hard to resist 1 Bb6 Qxb6 2 Qg5+ Kh8 3 Rxh5+ Nxh5 4 Qxh5+ Kg7 5 Qg5+ Kh7 6 0-0-0 with Rh1 to follow.

[Christopher Kreuzer]

Oh dear. Yes. I am looking at the wrong side of the board. If it helps, the mate in 9 starts with 1.Bh6+. You are right that most people would play 1.Bb6 so not that good a test of tactics. Shall we pretend there is a Black pawn on c7 or c5?

[Christopher Kreuzer]

Here is another test of chess tactics. Simple once you see the key move (the first move is obvious), but surprisingly difficult to see if you are not aware of the right way to play the most forcing continuation.

r1bq1r2/1p2ppk1/p2p1n2/7p/P3P1p1/2N1B3/1PPQ1P2/R3K2R

In this case, with White to move, it is forced mate in 9 moves.

Can you find the mating sequence?

There are two main lines:

[Variant "From Position"] [FEN "r1bq1r2/1p2ppk1/p2p1n2/7p/P3P1p1/2N1B3/1PPQ1P2/R3K2R w - - 0 1"] 1. Bh6+ Kh7 2. Bg7 Kxg7 3. Qg5+ Kh8 4. Rxh5+ Nxh5 5. Qxh5+ Kg7 6. Qg5+ Kh8 7. Ke2 g3 8. Rh1+ Bh3 9. Rxh3#

[Variant "From Position"] [FEN "r1bq1r2/1p2ppk1/p2p1n2/7p/P3P1p1/2N1B3/1PPQ1P2/R3K2R w - - 0 1"] 1. Bh6+ Kh8 2. Bg7+ Kg8 3. Qh6 Nh7 4. Bh8 Kxh8 5. Rxh5 Bf5 6. exf5 g3 7. Qxh7#

Both those lines feature quiet moves by the bishop to an empty square (g7 and h8) in order to force mate in the quickest number of moves.

Not particularly difficult to understand and not critical as White is easily winning in other ways, but not that easy to spot, and I thought it was a good lesson in how powerful quiet moves can be.

[John Clarke]

From an evening league match many years ago, here's another instructive example of what Christopher was writing about. The position below arose after Black’s 14th in a Smith-Morra Gambit that had not been very well handled by either side. White’s superiority is clear (as confirmed by StockFish), but how best to continue? Two pawns down, he can’t afford to hang about. His next move is fairly obvious, but the only subsequent one which maintains a decisive advantage is pretty counter-intuitive IMHO.

r3kbnr/1p1b1ppp/pq2p3/4P1B1/2B2p2/2N5/PP2Q1PP/R2R3K

[Christopher Kreuzer]

The only thing I can think of is 1.Rxd7 Kxd7 2.Bb3, which is certainly not the first move I would have thought of, but makes sense in that Ba4+ is hard to counter, and it helps to clear the c-file. Not that anything decisive is obvious.

PS. How did the Black pawn end up on f4. Presumably it was originally on d7?

PPS. I was being too clever. It is not 2.Bb3. Interesting to see why the winning move works though. It is the only move that prevents the king scurrying back to e8.

[Christopher Kreuzer]

Did no-one else try the above puzzle from John Clarke?

Here is a pretty mate in 7:

r2q2kr/pp5p/2pN4/2p1Q3/8/5P2/PPn2P1P/5RK1 w - - 16 27

White to move. Guess which square the Black king gets mated on.

[IM Jack Rudd]

I think it's d4.

[Christopher Kreuzer]

I think it's d4.

Yes, that is indeed the square that the Black king got mated on (via g7, h6, g5, f4 and e5). Well, I say mated, it was in fact a nugget of chess tactics I unearthed after an anonymous 3+2 game on Lichess. I find looking at interesting games afterwards, to see what was missed, in many ways more interesting than the games themselves!

There was an alternative square. I am pondering whether this could be phrased as a chess puzzle.

The best I can come up with is to show the position where it is a mate in 3 (i.e. 4 moves on from this position: 27. Qe6 Kg7 28. Qf7 Kh6 29. Nf5 Kg5 30. h4 Kf4). You have a discovered knight check. In fact, there are seven possible discovered checks with the knight.

But only 5 of them lead to mate in 2. One leads to mate in 8. One leads to mate in 10. Four of the mates in 2 end up with the Black king being mated on d4. The other discovered check with the knight leads to mate on a different square. Which knight move is this and which square?

r2q3r/pp3Q1p/2p5/2p2N2/5k1P/5P2/PPn2P2/5RK1 w - - 1 31

Seven knight moves by White. Five are mate in 2. One is mate in 8. One is mate in 10. Name the five knight moves that lead to mate in 2. In addition to this, name the knight move that leads to a mate in 2 that is NOT on d4. And name that square.

[Mick Norris]

Is 30 f4 an alternative to h4?