Monday, March 21, 2011

This is the end III

[Regular visitors to our humble blog might recall that at the end of January I wrote a post about the state of chess journalism.  It concluded:-

"Contemporary print media publications are currently under huge pressure from the internet and this is true in the chess world just as much as it is everywhere else ...

The old BCM way of going about things simply won't do any more. 'Facts' you can get on the internet for free, and much quicker than a printed magazine or newspaper could ever manage. To survive, magazines will have to provide something that the internet doesn't ...

Quality magazines will survive, but I'm quite sure that whether it turns out that we dig its grave in the short-, medium- or long-term, old-school chess journalism is inevitably doomed ...."


Well, if this thread about the imminent "extinction" of the BCM is to be believed, it seems the crisis point was closer than I had imagined. We'll get back to this later in the week, possibly with a bonus post tomorrow. In the meantime, here's something on B&N v K]







"I didn't understand why you were playing on.  I should have been more careful."

When I heard this I looked over to see a bishop and knight facing a lone king, a scoresheet with an awful lot of moves ticked off and, very obviously, no mate on the board.  I took a guess at what must have happened over the last fifty moves: our man had let his opponent sacrifice a piece, or perhaps even a rook, for the last pawn and then found he wasn't able to win the ending.  Ouch!

It could easily have been me.  At the time, just a couple of months ago (shortly before I started writing this series in fact), I wouldn't have known how to mate with bishop and knight against king either.  I think I'd always assumed it would be beyond me.  It does, after all, have the reputation of something that's incredibly tricky to do.

Just have a look at Epishin trying to finish off from the diagram at the head of today's blog:-






Oh deary me.  If an 2550+ GM can't do it, what chance does a club chesser have? None you might think, but here's the thing: while it's virtually impossible to work out over the board, if you do know what you're doing it's really rather straightforward.

OK, you could say that the chances of this ending arising coming up in actual play are so remote that it's not worth learning anyway.  This, in any event, is Silman's justification for not covering KBN v K in his book (Complete Endgame Course, Siles Press, 2008).  Having spent a fair bit of time on it over the past few weeks, however, I'd have to argue against this point of view.  It's worth studying, I think, not so much because of all the extra points you will get, but because in doing so you learn how bishops and knights work together.

So feel free to ignore the rest of this post if you'd rather - you're more than welcome just to pop back to Kempinski - Epishin and have another chuckle at a Grandmaster proving himself to be just as incompetent as the rest of us - but, for those who want it, here's a beginner's guide for how to mate with bishop and knight.  I'm sure it's not the only way, I doubt it's even the quickest way, but it will work and it's practical in the sense that it comes down to remembering just three simple positions.







Still with me? Good. Here we go with how to mate with bishop and knight against king in five easy stages. The technical information in what follows is based on what I learned from Pandolfini's Endgame Course (Simon & Schuster 1988). What I've tried to add is a little something to help make the process a touch easier.  Anyhoo, like Pandy, I’m going to go backwards and begin with the end.


PHASE ONE
The first stage of learning to win bishop and knight against king is to know the positions in which checkmate is actually delivered. There are quite a few, all of them with the enemy king trapped in a corner of the same colour as your bishop [If you have read this far then congratulations - you already know more about KBN v K than Epishin did in 2001: JMGB]. In fact it is possible to deliver mate in the ‘wrong’ corner, but these positions can’t be forced so ignore them.

A tip I picked up from Geoff Chandler (in his section of Master Chess: A course in 21 Lessons) is not just to look at the diagrams below and say ‘ah yes’. Instead, get a chess set out and actually set them up on a board. Seeing them in 3D, or perhaps just going to the effort of doing this, really seems to help burn the mating patterns into the brain.

The other thing that you should do is not just focus on mating on or around h8. Repeat the exercise for each corner on the board. When you have to actually do this for real your opponent might not be cooperative and put his king where you’d like him to.






Once you've done this, the next step is to solve mates in two, three and then four. Pandolfini’s Endgame Course gives five such exercises - pretty good - but, like far too much of the book, one of them is marred by a misprint.

This position is supposed to be a mate in three,




but after 15 minutes of trying I could only find a mate in four. When I looked at the solution the first move is Ng6 check! The King’s supposed to be on h8 – very annoying. Anyhoo, you might want to amuse yourself by finding the mates (start with the king on h8) before moving on.


PHASE TWO
After you've got the basic mating patterns firmly in your head, it's time to move on to what Pandolfini calls “The Lock”. Have a look at this position.




White seems to have Black’s king surrounded, but when you take a closer look you realise it doesn’t just appear that Black might be in danger of becoming trapped: he actually already is. In fact, the bishop and knight are doing all the work on their own. White doesn’t even need the king to keep Black under control.




So the king is free to drive back into the corner. It’s not all that hard to do either, just a simple matter of gaining the opposition. We’ll give Black the first move and play will go something like,

1 … Kf8, 2 Kd7 Kg7, 3 Ke7 Kg8 (otherwise the king gets even closer) and now 4 Bh6!



For some reason this bishop manoeuvre really tickles me. Now after 4 … Kh7, 5 Bf8! Kg8

This is how you get from Pandolfini’s lock to a position where mate is imminent. How you do you get to Pandolfini’s position in the first place? That’s the next stage. Before you move on, though, it’s worth practicing the lock in each corner and going sideways as well as up and down.


PHASE THREE



You can’t force a checkmate with the opposition king in the ‘wrong’ corner so that's where the other guy is going to run with his king. What we have to do is force him across (or down or up) the board to the right place.

In the basic starting position for this phase the bishop can be on any square on the b1-h7 diagonal. It could also be somewhere between a2 and g8 too, although in that case we would be pushing Black’s king down to h1 and not a8 which is where he’s heading here.

So, from the diagram,
1 Nf7+ Kg8, 2 Bb1 (losing a move; it doesn’t really matter where the bishop goes) Kf8, 3 Bh7 (cutting the king off) … Ke8, 4 Ne5 Kd8, 5 Ke6 Kc7

 


This is a critical position. Pandolfini doesn't pay much attention to it, but in my experience (and playing out the ending with T.C. he seemed to have a similarly difficulty) what follows is very difficult to grasp.

It looks like Black’s going to escape doesn’t it? He’s going to b6 next move, but if we cover that square with our knight he’ll get away through c6. What to do? Have a think about it then scroll down …

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.




Got it? If not, this should help you work out where the pieces should be.




It’s easy enough when you know how, I think, but a real bugger to visualise in advance if you don’t know what you’re looking for.

Anyhoo, 6 Ne5 Kc6, 7 Bd3! and Black’s king has nothing better than 7 … Kb7.




From here, you need to know how to get to Pandolfini’s Lock. It shouldn’t be too difficult to work out how (quick review, you want the bishop on b5, the knight on d5 and the king somewhere around e6).

When you’ve got Phase Three down it’s time to put the whole thing together. Set up the position at the beginning of this stage and force checkmate via the bishop/knight/king formation you learned in Phase Two.

This is pretty much where Pandolfini leaves things. We have a couple of more stages to go though.


PHASE FOUR
So, you’ve practice mating that lone king with your bishop and knight every which way. You can start in any corner and go up down or across according to your will. You’re so familiar with The Lock that you can visualise it several moves in advance.

What’s next? Well, something like
1 Nf7+ Kg8, 2 Bb1 Kf8, 3 Bh7 Ke8, 4 Ne5 Kd8, 5 Ke6 Kc7, 6 Ne5 Kc6, 7 Bd3 Kb7 8 Bb5
is all very well but what if Black doesn’t follow our plans? In this line, Black can vary with

4 … Kf8 (this is effectively what Kempinski did against Epishin - i.e. went Ka5-a6 instead of Ka5-a4) or 5 … Ke8 or 6 … Kd8

but ultimately it won’t do him any good. In fact sometimes it saves White time and we can mate a move or two quicker if Black plays like this. Not that for practical play we’re particularly bothered if we give checkmate after 19 moves or 20, but it’s good practice to work out what to do if Black tries these other moves.

After you’ve done all that there’s just one thing left to do, and that’s solve a problem that may already have occurred to you.


PHASE FIVE
Mating from the wrong corner is fine, but how do you get that lonesome king into the wrong corner in the first place? I’m not sure there’s much of an answer to that except just to practice. Set up the pieces in random positions and just do it playing against a computer or a friend or just moving the other king yourself.

Here’s a couple of test examples that you might want to try yourself against.




Readers who’ve been with us for a while might remember EJH winning this ending a year or two back (you can find out how he did it here).  It’s also a good idea to make things as difficult for yourself as possible by starting with the enemy king in the middle and your pieces scattered around the edge of the board. Here’s a position I set up and played out against myself (my computer tends to head straight for the wrong corner anyway which isn't much help).




I’m not saying that how I finished Black off is the best way, but it’s effective enough and that’s what we’re aiming for – the ability to mate with bishop and knight in a practical game when we need to and not theoretical perfection.


THIS IS THE END
Congratulations. If you made it this far you can now mate with bishop and knight against king. The last thing you need to do is review things every now and again so the method stays in your brain.  These are the three critical positions that you probably must simply know:-


1. King in the wrong corner



2. Prevent an escape



3. Pandolfini’s Lock



Keep these in mind and the hardest thing you’ll have to do is maintain your patience while you wait for the opportunity to demonstrate your knowledge! Good luck. Oh, and if you do manage to mate somebody in this ending, whether a result of this post or not, do let us know.  We'd love to hear from you.





PREVIOUSLY:-
This is the end
This is the end II







4 comments:

ejh said...

The satisfaction involved in actually delivering this mate is worth not only the time it takes to learn it (not much, to be honest) but the twenty-five years it takes to get it in a game. Honestly.

Michael said...

Thanks for this article!

I also found this video very instructive:

http://www.youtube.com/watch?v=PWZ7h2yrJME

I think it took me only 2 viewings - ie 20 minutes - to learn this 'method of 3 triangles' approach off by heart. Still waiting to try it out in practice though...

Lopsy said...

Awesome, this is really helpful! Thanks a bunch; now I'm hoping this ending shows up in an actual game instead of fearing it.

(I'm hoping the next one will be about that tricky Queen v. Rook ending- but I don't want to be presumptuous of course.)

Jonathan B said...

@Michael:
That video is pretty interesting
(clickable here for anybody unwilling to cut and paste). Kind of like a full-board Pandolfini's Lock system.

Have you tried it when Black doesn't keep heading for the wrong corner?


@Lopsy:
We'll see, but I fear you over-estimate my endgame prowess somewhat.

Thanks for the comments.

J