Monday, January 05, 2015

The Final Move of 2014

White to play his last move of 2014

My final game of 2014 concluded, rather fittingly, right here. Nothing left but kings and pawns. A year in chess that went as far as it could go. Well, almost.

My opponent and I could have finished with bare kings, I suppose. Since I have a move in this position that pretty much forces resignation, though, I wasn’t especially keen for that to happen.

Actually, this position could have been an even more appropriate finish to the year than it was. You see, before I opted for a Year of ISEs, I had intended that my 2014 blogging was going to be all about king and pawn endgames.

White to play
Anand - Chess Genius, London Intel GP (rapid) 1994

Why king and pawn endings?

Well, for a start I rather like them. Long standing readers may remember earlier posts such as The Penarth Pier Problem from July 2012 or This is the End VI (June 2011).

More than that, though, pawn endings are not just fundamental to chess they are an entirely logical follow-on from an interest in rook and pawn positions. The more I looked at rook endgames the more I realised that I needed to know when trading the pieces would lead to a win and when it would only be a draw (e.g. Random Rook Endings IV; SMA#24; 53).

So why didn’t I go with pawn endings last year then? Aside from wanting to change things up a little and make the switch away from endgame themes, it seemed to me that writing about pawn endings would be inherently problematic.

King and pawn positions do not crop up in the games of the very top players very often. You’re not likely to see many at a Candidates’ tournament, say, or in a World Championship match. Which would mean I’d be mining the same old sources that have already been well-covered by everybody else.

I was also worried that unlike rook endings where there are all sorts of concrete positions to study and learn - Lucena, Philidor, Vancura, rook plus f&h against rook and so on - king and pawn endgames are more about techniques and calculation. You simply have to work them out.

I wasn’t sure what I’d have to contribute. What would I have to say beyond, "here’s an interesting position that I found in a book" and "here’s another one"? Where were the stories going to come from?

Was there really a narrative that could be sustained for a whole year? I didn’t think so.

36 h3-h4!

Which was a pity because Muller and Lamprecht’s Secrets of Pawn Endings looked like it would be fun to work through and Joel Benjamin's forthcoming book sounded like it would be right up my street. The publication date of Liquidation on the Chessboard seems to have slipped so I still have no idea what’s going to be in it but if it’s material like Euwe - Alekhine, World Championship match (24) 1935Sokolov - Korchnoi, Tilburg 1987Shirov - Akopian, Oakham 1992 and Shirov - Timman, Wijk aan Zee 1996 then I’m fairly sure I'd want to read it.

And then there's Anand’s game against the computer at the Intel Rapid tournament. Vishy pushed his pawn to h4 and won easily. Why? Because whether Black captures or pushes on himself he (or "it" in this case) will leaves lines open for White’s king to penetrate. The machine can’t bring his own king over to stop White eating the kingside pawns because of White’s threat to create a passed pawn on the queenside.

Not that any of that happened in 2014. It all went down two decades earlier. Compare it to my game, my final game of the year, and the relevance is obvious.

The thing is, I knew. I knew h4 would win my game. Not just at the point I played it. Back when there were still bishops on the board half a dozen moves earlier. I knew I could bully my opponent into a passive position by putting my piece on an aggressive square because if he tried swapping the pieces off the king and pawn ending would be trivial.

And I knew this because I remembered Anand’s win against the computer from twenty years ago. If you’re good at chess I imagine this sort of thing happens to you all the time, but for me it - the illusion that I can actually play this game - is pretty rare.

So when the guy playing Black did go for an exchange of bishops I ended 2014 with a point and a feeling of being (unduly) pleased with myself. I finished the year with that and the realisation that maybe king and pawn stories exist after all.


Jonathan Rogers said...

Certainly you can last a year with king and pawn endings. Don't forget Karpov v Kasparov, Las Palmas 1996, where both players made mistakes.

One can probably work out the result of 95% or more of them with best play, but that doesn't make them trivial. I still don't know what best play would have been in Collins v Rogers, 4NCL 2011, for example. We were little wiser after analysing it for one and a half hours afterwards.

Oh, and Happy New Year!

Andrew Gelman said...

It's good to see some good news for a change! Also, I learned something from your post. I would've gone Kd3 without a thought, and then only a few moves later would I have realized that I couldn't do anything after pushing the pawn to a5.

Jonathan B said...

One thing I noticed after looking at a lot of Outside Passed Pawn positions recently is that the attacker usually seems to clarify the other wing first and only then push the passed pawn.

E.g the Lombardy - Fischer game from Friday.

Fischer gave back the exchange then traded rooks to get a pawn ending where he could engineer a passed pawn on the queenside. Then he makes sure the kingside is blocked (so no counter play from White) and only then pushed the a-pawn and forces the White king wide.

So I guess old Nimzo was right - the threat (of pushing a passed pawn) really is stronger than the execution

John Cox said...

Or indeed Topalov-Kasparov, Linares 2005. Seems to have been quite the weakness for the great man. Difficult to target, though.

Jonathan B said...

Thanks for the suggestions gentlemen.

I’d forgot about the Topalov - Kasparov. Not sure I was every aware of the Karpov - Kasparov game.

Skeptic said...

In the excellent book "Essential Chess Endings", Howell explains why strong players first deal with the other wing and only then push the passed pawn in such positions. It is not that the "threat is stronger than the execution" at all, but two very concrete reasons.

1). First of all, it's to kill the opponent's possible counterplay, which is the first priority (after creating the passed pawn itself). Otherwise a won game can easily turn into a drawn or lost one.

2). Second, in pawn endings with an outside passed pawn, the old adage "passed pawns must be pushed" is distinctly unhelpful: the further *back* the outside passed pawn is, the more moves the opponent's king has to make to capture it, giving the owner more time to wreck havoc on the other wing!

Jonathan B said...

Well Skeptic, you say Potato and I’ll say Potato but what you’re describing sounds exactly like threat/execution to me.

As you say, you push the pawn the enemy king can deal with it. Execution of the plan of pushing the pawn is therefore weak.

But if you don’t push the pawn and get busy on the opposite wing the enemy king can’t go across to help because of the threat to push the pawn.

So I’ll stay saying potato but do feel free to say potato if you wish.