Wednesday, September 28, 2011

Blue or Red Pill XII

White to play
JMGB v AN Other, Sunningdale Open September 2011

White, your humble scribe, to play. It's a real life BORP moment.

My pieces are looking menacing. How about I loosen up Black’s king a little with 26 e6? Something like 26 … Qb3, 27 exf7+ Qxf7, 28 Bxg6 will follow

and after Black moves his queen to cover the bishop so a discovered check won’t just win I’ll have an extra pawn as well as attacking chances. That looks pretty good.

What about 26 Bxg6, though? If Black takes on g6 he’ll drop all his kingside pawns and then his own bishop too so he only has one move. 26 … Qxg2+, 27 Qxg2 Bxg2, 28 Bxf7+

After the bishops come off, I’ll be a pawn up here too, this time with a pair of connected passers in a rook ending. That looks pretty good too.

It seems like a straight choice ...

Keep the queens on and go for an attack.

Swap them off and head for an ending.

The choice was mine. Now it’s yours.

Blue or Red Pill Index


Anonymous said...

It's a style issue. The club's board 1, one of the better known purveyors of lesser known opening variations, reckons that given a choice, I usually simplify, particularly in time limited games. So I would go for the Rook ending.

What frequently happens is that the judgement of simplification is good, but rather better than the ensuing technique, so in practical terms the complex approach might have worked better.

Anonymous said...

what about mixing up the colours here and going 1. e6 and then following ... Qb3 with Bxg6?!


Anonymous said...

Having cheated by looking at the actual game continuation, I'll have a dose of the blue pill, please.

The red pill was very tempting and I would have played it like a shot in quickplay. At longplay, unless one's surname is Grischuk, one should have the time on hand to evaluate the rook ending properly.

Seani's "white pill" looks very tempting as well, not that this patzer would ever have thought of it at all - either in analysis or in live play.

John Cox said...

Well, definitely not the rook ending, I'd say - at the end ...Kxf7; Kxg2 Ke6; Kf3 Kf5 is very annoying because you don't have time to give him a check, isn't it? Whereas one or perhaps both of the other two options must be winning.

Anonymous said...

My first thought on the rook ending agreed with John, but a second thought made me think the blockade was illusory. After ...Kf5, Re1 seems to leave the e-pawn unstoppable without acquiescing in a lost K+P ending.


Jonathan B said...

With apologies for the delayed replies ...

@Seani: I didn't consider mixing the plans during the game - I only noticed this possibility when going over the game with a computer running in the background. I think at the time I actually considered 1 e6 Qa2 (rather than ... Qb3). The reason I switched for the post was that my silicon pal doesn't like 2 Bxg6 after ... Qb3 (presumably because of some kind of ... Qe3+ resource) but thinks it's winning after ... Qa2.

Jonathan B said...

@anony2: "Having cheated by looking at the actual game continuation, I'll have a dose of the blue pill, please."

Indeed, the game ended in a draw after I swapped into a rook ending. I didn't swap off for the right reasons though.

@John Coxat the end ...Kxf7; Kxg2 Ke6; Kf3 Kf5 is very annoying

Indeed. That's exactly how the game went. For some reason I'd only considered ... Rxf7 after which I expected the black d-pawn would drop giving me a relatively straightforward win.

Jonathan B said...

@Richard: After ...Kf5, Re1 seems to leave the e-pawn unstoppable without acquiescing in a lost K+P ending.

That's what I thought my first, then I changed my mind - wrongly I think - and certainly for the wrong reason - and played something else.

@anony1: What frequently happens is that the judgement of simplification is good, but rather better than the ensuing technique

My technique was certainly not up to swapping off, so in practice perhaps trying to keep an attack going would have been a better idea, yes. That said, if I'd have done that and screwed it up I'd probably be here saying I should have swapped off.

Anyhoo, more on the rook ending on Monday.

John Cox said...

I don't follow, Richard - what's your move after Re1 d3?

Interesting point about ...Rxf7, JB. I bet I know why you only considered ..Rxf7; it's because in the position you were looking at that would be the natural move, since it's a middle game with the kings exposed. If you'd been shown the position after ...Bxg2 and asked to choose a move, you'd go ...Kxf7 like a shot, I suspect.

Maybe not - it does superficially block the rook in a way, I suppose. Still, I expect this (carrying on analysing middle-game moves instead of endgame ones when looking ahead) is a fruitful source of errors. It would be quite a good instructional position, I think - no doubt why you chose it.

Anonymous said...

Assuming i can't just round up the d pawn with Ke3, i think

1. Re1 d3 2.e6 d2 3.Re5+ Kf6 4.Ke2 Ke7 5.f5 Rd8 6. Kd1 Kf6 7. Ra5 is winning? Black R must leave 8th rank and then white plays Ra8.

Although doing it in my head so probably missed something.


Anonymous said...

On the other hand maybe it's a study like draw!

1. Re1 d3 2.e6 d2 3.Re5+ Kf6 4. Ke2 Ke7 5.f5 Rd8 6. Kd1 Kf6 7.Ra5 Kg5! 8. h4+ Kf6 9.Rb5 h5! 10.Ra5 Rd6 11. Ra8 Rd5! 12. Rf8+ Ke7 13. Rf7+ Ke8 14. f6 Rd6! 15. Re7+ Kf8 =

All rook endings are drawn. Avoid them like the plague.


JOhn Cox said...

Actually you can just go Re1 d3; Rd1 winning the d-pawn, can't you?

Maybe Re1 Ra8 would have been more sensible on my part. You'd think we'd know whether the pawn ending after 1 Re1 Ra8 2 e6 Re8 3 Re5+ Kf6 4 Ke4 Rxe6 5 Rxe6 Kex6 7 Kd4 was winning or not, wouldn't you, but I'm not sure. Baburin wrote an article about this on-line somewhere. White wins because he hasn't moved his h-pawn and has a choice of h3 or h4 at some moment; is that right?!

I have to do some work, tiresomely...

Anonymous said...

Tablebases say Kd4 Pf4, h(?) v Ke6 Ph6 is a win for White, wherever you put the White h-pawn: h2, h3, h4 or h5, and with or without the move. It's when the Black pawn is on h7 that the White h-pawn square is critical: only h2 wins, probably for the reason given by JC.

Does a barrage of checks help Black after ... Ra8 e6 Ra3+? It seems not: White can hide on either h4 or f1.

Anonymous said...

I think after 1.Re1 d3 2. Rd1 Black can play 2...Ra8 3.Rxd3 Ra4.

Winning the fpawn.

The K+P ending in the other line is winning - that was my whole justification for Re1!


Jonathan B said...

I was going to have another look at the rook ending in today's post but so much analysis got posted here that I thought I'd kick the CJ-ECF dog again.

I'll come back to it at some point, though, if only to explore the 'sources of error' question. Interesting point that JC makes above about the ... Rxf7/... Kxf7 confusion. At the time I couldn't understand why I hadn't considered the king capture but the middlegame/endgame mindset thesis sounds convicing to me.

Anyhoo, thanks for your input chaps.