What's the best move? (Knights and half-open files)
Best highway sign (Photo by Nick Youngson, Alpha Stock Images, CC 3.0)

What's the best move? (Knights and half-open files)


This is a quick post about a position that popped up in a 3/2 blitz game I just played where my opponent and I both missed a key tactic.

While this position is so far out of book that you can't even see the book anymore, it showcases a tactic that often sneaks up on me. I figure that if I overlook it, others at my level might miss it too and so am sharing it in this post.

What's Black's best move?

Black's best move is one of those moves that's immediately obvious as soon as you see it but, if you're like me, you won't see it unless you're primed to look out for it in the first place.

So if you spot it right away, props and congrats!

If you don't, and I didn't, both during the game and afterwards when I worked through the post-game analysis, here's a quick series of problems to prime the pump.

Here's a mate-in-1 problem that arose in a daily game I played late last year.

And here's another mate-in-1 problem from a 3/2 blitz game I played earlier this month.

As you picked up, in both cases the knight delivered mate and in both cases the knight was able to do so because a rook controlled the half-open file preventing the knight from being captured.

Now let's get back to tonight's blitz game and look at the same position but from my opponent's point of view after I made an incorrect move. What's White's best move here?

White's best move in tonight's blitz game is the same as my mate-in-1 problems from my earlier daily  and blitz games — move the knight to attack.

In the earlier games, the knight delivered check, which was mate because the knight couldn't be taken.

In tonight's game, the knight should fork the rook and the queen. While I would have been able to take the knight, I could have done so only at the cost of being on the losing end of a knight for rook exchange and would have to face the additional disadvantages of my opponent controlling the open file and my having to deal with doubled pawns and a rook pinning my knight on the back rank. 

So with this background, Black's best move in the original problem becomes clear. 

Why did I miss it? I can think of a few reasons.

First, it's a prophylactic move. I often don't see those moves because I'm just not looking for them. This position is a helpful reminder to myself that it's as important (and perhaps more important) to inhibit my opponent's actions.

Second, the move is out of the ordinary. Even though I've argued for playing the knight to Na6 as a defense to the English opening, I still don't routinely think about shifting the knight to the edge of the board.

Third, I wasn't primed to look out for the threat against my rook. While I'm a believer in CCT, I didn't put 2 + 2 together to get 4. I look for checks but this wasn't one. While I thought I had looked for threats, I didn't see the a file as an open one because White's knight was "blocking" the file and so I missed the tactic. Of course, in reality the a file is already half-open, a fact that becomes abundantly clear as soon as White plays Nb6. 

As it happens, this is a motif that appears from time to time. As @Silman suggests, a chess pattern "can be mastered by simply finding examples of one of them, study that pattern over and over (look for dozens of examples or more), and once you understand its nuances you should step to another pattern and master that too." Given that advice, here are some examples of this pattern in action.





So if you see a half-open file controlled by a rook (or a queen) with a knight lurking nearby, start paying attention to possible tactics, especially if there's a possible check/mate or fork and in particular if the fork threatens a rook that's hanging out there on its own without support.



6/22/19 Update #1 — With a hat tip to @simaginfan, here's Taimanov's victory over Karpov. 



7/22/19 Update #2

I’ve been thinking further about my blitz game to identify the root cause of my error and to figure out how to fix it. So that you can follow along, here’s my annotation of the game.

I now realize that I didn’t make a mental error when I played 12...e6??, overlooking the threat of 13.Nb6!

Instead, when I played 12...e6??, I merely compounded the same mental mistake I had been making for the last six moves.

I should have been constantly assessing both sides’ pawn structures to mentally flag pawn islands, open files, half-open files, weak squares, strong squares, backwards pawns, doubled pawns, hooks, etc.

In my defense, at 7.axb3, I mentally noted White’s doubled pawns on the b file.

At 8.h3 I also paid attention to the hook at h3 in the pawns protecting White’s king and was planning to attack it.

But I never mentally bookmarked the fact that the a file was half-open.

As a result, I didn’t appreciate that simply by playing 7.axb3, White had developed its rook without having to move it anywhere. This is similar to the Benko gambit where Black can develop its rook at a8 merely by leaving it in place.

So my first mental mistake was at 7.axb3 when I didn’t notice that the warning light for a half-open file on the a file changed to yellow.

My second mental mistake was at 9...c6 when I didn’t realize that by advancing my c pawn, I had created a weak square at b6. At that point the warning light for a half-open file on the a file changed to red but I paid it no attention.

My third mental mistake was at 10.Qc8 when I moved my queen to a new square where it was no longer guarding b6. At that point the warning light for a half-open file on the a file started blinking on and off to try to catch my attention but I paid it no heed.

So when White played 12.Na4 and I overlooked the threat of 13.Nb6!, that was merely my fourth mental mistake, compounding the same error I had been making repeatedly since the 7th move. By this stage, not only was the warning light for the half-open a file flashing on and off furiously but an alarm was sounding, announcing “DANGER DANGER” yet I still gave it no attention.

The lesson learned is to start paying attention to half-open files as soon as they open, not merely when danger arrives on my doorstep. In fact, if I were paying any attention to what @Silman says, I might want to start thinking about half-open files even before they open up.

To that end, here’s some basic yet helpful language from @BrucePandolfini’s Kasparov and Deep Blue: The Historic Chess Match Between Man and Machine:

Rooks, pawns, and files have an intimate relationship. From a rook's perspective (or a queen's), there are three kinds of files: a closed file, an open file, and a half-open file. A closed file contains pawns of both colors, an open file has no pawns on it at all, and a half-open file has pawns of only one color. 

Neither player’s rooks are able to use a closed file effectively. Either player’s rooks can use an open file. And the rooks of only one player may use a half-open file (the player with no friendly pawn in the way). Thus, the attacker should seek open and half-open files for his rooks (and queen). By occupying these files, his major pieces can attack the opponent’s position.

The key point is to appreciate that half-open files favor the player without a friendly pawn blocking its own rook. That’s why I should have mentally noted the half-open a file as soon as White played 7.axb3. This is exactly the same as paying attention when one side or the other fianchettos a bishop on an open or half-open long diagonal.

While the knight-bishop exchange favored Black, it created an asymmetric position. Although White got the worse of the exchange, White walked away with a half-open file for its queenside rook, creating an opportunity for White at some point to convert that slight positional plus factor into a more tangible advantage if Black played inaccurately. That came close to happening in this blitz game and almost certainly would have happened at higher ratings or longer time controls. So one way both sides can progress in the game is to start paying attention to these possibilities from the outset.

So how can I remember to pay attention to pawn structures, especially open and half-open files, and the positional and tactical possibilities that they create?

One way is making a point to use overt conscious thought. Since I still have problems with watching my time and dropping pieces, this wholly rational approach just adds to the cognitive burdens I feel most acutely when I’m playing blitz games or I suddenly realize I have 8 hours to make a move in 34 daily games.

Talk with your pieces

Here’s a totally different approach. In The Seven Deadly Chess Sins Jonathan Rowson offers this crazy suggestion:

I want to show how pattern recognition can lead to improvements in your play in a conventional manner, but in the process I aim to introduce the jovial but entirely serious notion of talking with your pieces. I have come to think of it as an exemplar of the type of thinking that can help to override existing patterns….

When analyzing a Karpov game, Rowson writes:

Assuming you didn't feel up to calculating all the way, you could simply 'talk with your a1-rook'. This sounds wacky, I know, but try it. These 'conversations' needn't involve any small talk or lead you to think of the pieces as conscious beings; it's just a question of making full use of your forces. When you think of yourself as a commander of an army you want to use all your soldiers and give then opportunities to exercise their particular talents. In this case you'd be plunging into a complex line with less than a full army because the a1-rook doesn't participate.

Rowson adds:

[M]y first suggestion [to cultivate your intuition] is to take the idea of 'talking with your pieces' seriously. When you talk to your pieces, you can bring unconscious value judgements to the surface of your thoughts and let your pieces 'tell' you what they 'think' about how to proceed…. Once you realize that evaluating well is the key to playing good moves, you need to learn to evaluate. Simple in one sense, but there's a grave danger that if you evaluate in the conventional sense (stop and weigh imbalances, etc.) you will slip into your old conscious 'thinking' patterns with all its rules, memories and painful impressions. Sometimes it is better to take a fresh look at the position, and talking to your pieces can be a very effective way to do it.

If this sounds way too wacky for a 'serious' game like chess, maybe you'd believe a more 'serious' writer. How about Nimzowitsch? Pawn-chains, blockaders, prophylaxis, whatever; Nimzowitsch did something very much like talking with his pieces too: "It may seem strange, but to me the chess pieces have living souls; they have wishes and desires, slumbering in their subconscious, to be understood only by me. They want something without understanding why. I don't understand either, but I know what they want."

Rowson quotes Tiger Perrson:

“When you ask your knights what they want, asking for the general direction of their yearnings, you will get a different view of the situation on the board; it especially helps me when I'm getting too - how do you say it - exact; when calculation takes up too much of my thinking. If you play a very sharp opening and then suddenly find yourself in a strategically difficult position where the tempo is no longer of the greatest importance, then it's a very good idea to change gear by asking your pieces where they'd like to be. I use this trick a lot. In endgames I talk continually to my pieces. I guess this is what all strong players do, to some extent; at least it's a very good explanation of what one is really trying to do in the endgame when finding the optimal constellation of the pieces.

"As well in the middlegame as in the endgame it's a good way of finding out which pieces to exchange and which to keep; a piece that has no will, that finds no meaning in roaming the board, should immediately be exchanged.

"The concept of 'talking to your pieces' can be stretched infinitely, but I usually use it to 'decomputerify' my thoughts and make chess more animated."

Rowson concludes with this thought:

To add further weight to the idea of talking with your pieces, Emanuel Lasker based his theory of chess aesthetics on the idea of 'achievement', or more specifically 'the achievement of the pieces'. Lasker's ideal is for the pieces to achieve a task of vital importance when there is only one way for them to do so. In a sense this is what talking to your pieces is all about; you consult with your forces very deliberately and 'ask' them how they will make the most of themselves for the common good. 

So it is my opinion that 'talking with your pieces' is an effective way to cultivate your intuitive abilities.

In the spirit of Rowson’s suggestion, we don’t need to have many chats with our pieces to realize that our pieces have very different opinions about what pawn structures they each prefer.

Knights want closed positions, especially ones where they can post up in enemy territory with the support of a friendly pawn or two and the backing of a friendly rook and watch the game whirl around them.

Likewise bishops enjoy open games, especially ones with an open long diagonal of its own color where they can swoop down like a bird of prey.

For their part, rooks seek out open and half-open files where they can swing like a wrecking ball against the opposing forces. This is the whole idea behind Nimzowitsch’s mysterious rook moves. It’s also why engines give points to placing rooks on open or half-open files. As R.M Hyatt, A.E. Gower and H.L. Nelson write in Computers, Chess and Cognition:

Half-open files and king tropism are really long-range heuristics that try to anticipate attacks and file openings long before the search actually reaches the position. This attracts rooks to places where they will probably be useful later, although at times rooks end up on squares where they have to move again.

Thus, talking with our pieces will naturally lead us to paying closer attention to pawn structures, both our own and our opponents, and talking with our rooks will invite us to pay particular attention to open and half-open files.

Eavesdrop on the opposing pieces

To take Rowson's crazy idea even deeper into insanity, maybe we shouldn't just be talking with our own pieces. Maybe we should be eavesdropping on the opposing pieces, just like a baseball player who tries to steal signs or a NBA player who tries to sneak into the opposing team's sideline huddle during a timeout.

As @Silman writes, one thing that is almost ubiquitous among beginning players is that we are, more or less, dancing with ourselves, without paying attention to our opponent's plans. His solution at least sounds simple:

As one moves up the rating ladder, it’s extremely important to truly explore the ideas, plans, tactical nuances, and structures of both sides. Though grandmasters see most of those things at a glance, that doesn’t mean you shouldn’t start to train yourself to see as many possibilities for both sides as you can.

So the crazy idea occurs to me that if I were listening to my opponent's pieces in my blitz game, I might have heard the opposing knight and rook whispering between themselves about teaming up to beat up and knock out my queenside rook. 

Who knows if this listening to the pieces is gonna work but it seems intriguing enough to give it a try.