(Black to move)

click for larger view

The position is technically even.

Things seem to go wrong for Black in handling White's break on the a-file. It can't really be stopped, but if Black is careful removing handy targets for White, it doesn't do much (*).

But it seems Black makes subtle mistakes putting targets up for grabs...

<19...b6 20.g3 Kf7 21.Ra4 c5>

That's the problem with 21...c5, it gives the soon-to-be doubled b4-pawn a target. In fact, putting the pawn on c5 means White doesn't get doubled pawns when opening the file.

<22.Rfa1 Rac8 23.b4>

(After White's breakout 23.b4)

click for larger view

<23...axb4 24.cxb4 cxb4 25.Rxb4 Ne7>

The knight comes over a little late(*).

<26.Rc4 Ke6 27.b4 b5 28.Rc5>

Next, instead of playing 26...c5 and blocking off the bishop, Black plays 26...Ke3, where it becomes an X-ray target on the 6th. White immediately plays b4 and then occupies c5 with the rook, the very square Black could have owned.

<28...c6 29.Ra6>

(Black to move)

click for larger view

<29...Ra8 30.Rb6 Rfd8 31.Rcxb5>

The alarm bells are beginning to ring, yet Black invites White's rook onto the 6th with 29...Ra8 instead of playing the critical 29...Rb8, allowing the b5-pawn to fall to a pin. Once White wins the Q-side Black has little hope.

Mecking does a nice job handing the technical execution of the game here.

Side note- my <MillBase> version of the game continues with 37...Nc8 38.Rc7 (38.Bf8 or 38.Rh7 are both stronger though).

(*) E.g. instead of 21...c5 Black is better playing

A nice example of Black playing this line without pushing ...c5 is

Seems <RedHotPawn> and <CT> both give the extended version, as does Wojciech's <OlimpBase>.

Given the latter, I think the extended version is more likely.

Wojciech gives this blurb on the tournament:

http://www.olimpbase.org/1968/1968i...

listing the above game as a highlight, with this comment:

<Neat, subtle win of young Brazilian star.>