Members · Prefs · Laboratory · Collections · Openings · Endgames · Sacrifices · History · Search Kibitzing · Kibitzer's Café · Chessforums · Tournament Index · Players · Kibitzing
Bent Larsen vs Lajos Portisch
Portisch - Larsen Candidates Quarterfinal (1977), Rotterdam NED, rd 3, Mar-02
Sicilian Defense: Paulsen. Bastrikov Variation (B47)  ·  1-0



Annotations by Stockfish (Computer).      [22200 more games annotated by Stockfish]

explore this opening
find similar games 54 more Larsen/Portisch games
PGN: download | view | print Help: general | java-troubleshooting

TIP: To see the raw PGN for this game, click on the PGN: view link above.

PGN Viewer:  What is this?
For help with this chess viewer, please see the Olga Chess Viewer Quickstart Guide.


Kibitzer's Corner
Mar-30-06  prinsallan: Long ardeous endgame in which Larsen eventually manages to win against a splendid Portisch.
Oct-18-15  Howard: For those of you with computers, where was the point of no return for Portisch---in other words, where did he throw away the draw for good ?
Oct-23-15  NeverAgain: Howard: move 82, although the engines aren't unanimous on this.

This long game has a couple of other interesting points.

At move 6 we have a Taimanov/Paulsen tabiya. Here Portisch opted for the unusual 6...b6. Unusual, as in the Sicilian the queenside fianchetto is typically achieved via ...a6 and ...b5, gaining space and threatening the c3 knight at the same time.

The tame 6...b6 appears to be playable if unambitious. Maki (2355) - Pyhala (2295) (1985 Finnish ch, Espoo), the only other game in Megabase 2012 featuring this move, varied here with 7.Nxc6; and although White went on to grind out another long KRPP-vs-KRP endgame win (1-0, 75), it was thanks to Black's blunder before time control not the opening (the position was even until 12...Be7?!).

However, with his next move (7...Bb7?!) Portisch apparently mixed up the move order and allowed White a chance for a solid opening advantage. What Larsen missed was 8.Nbd5:

click for larger view

White gets the two bishops and Black chronic weaknesses on d5 and d6:

<8.Nbd5 Qb8 9.Qd2 Be7 <<9...a6 is now just a waste of time, White still plays 10.0-0-0 as 10...axb5?? Qxd7 is mate>> 10.0-0-0 Nf6 11.Bf4 e5 12.Bg3 0-0 13.Nd6 Bxd6 14.Qxd6 > [1.08/33 Komodo 9.2]

click for larger view

Having missed that, White had only a small plus until 17.Qg3. Here Black baited him again with the dubious 17...f6 (the position called for ...b5 and ...f5 at this point), and once again While let his chance pass by:


click for larger view

It is hard to believe an attacking player like Larsen missing such an obvious move. <18...Rf7> forced, as <<18...g6? 19.Rxg6+! Kh8 <<<19...hxg6 20.Qxg6+ Kh8 21.Qh6 Kg8 22.exf6 is hardly better>>> 19.Rh6>> gives White a winning attack; <19.exf6 Qxg3 20.fxg3 g6 21.h4 > [0.93/35 K9.2]

click for larger view

After that things were close to even and following the bishops exchange a double-rook ending arose that Black shouldn't have had too much trouble drawing, despite his weak and scattered pawns, had he played the active <30...Rb4> instead of the lame 30...Rg5:

click for larger view

<30...Rb4 31.Rbb7 <<or 31.Rxb4 axb4>> Rxb7 32.Rxb7 a4=> [0.00/45 K9.2] White can create an outside passed pawn on the queenside but his Rook will be in front of it (= draw).

[to be continued]

Oct-23-15  NeverAgain: This chance Larsen didn't let slip and for the next 47 moves he relentlessly nurtured his advantage. On move 77 Portisch finally cracked with 77...Rc8?. The black king had to head for the enemy passed pawns immediately, simultaneously clearing the way for his own g-pawn - <77...Kf4>

click for larger view

<77...Kf4 78.Ra8 g4 79.Rf8+ Ke3 80.b4 Rc3+ 81.Kb2 Rc8 > [1.37/54 Stockfish 211015] with a long KQP-vs-KQ endgame to follow, where - to quote a classic (Tal, IIRC?) - "you can give as many checks as you want". Perhaps there is a win there, but it would have required another adjournment to find. As it is Portisch went down without much of a fight.

Here's a small mystery for ya. Komodo concurs with SF; however, Houdini 4 Pro x64 (using the same 6-men Syzygy tablebases as the other two engines) claims that the position after <77...Kf4> is a dead draw! (0.00/37) It gives White a at best with TB turned off. Curiously enough, the much lower-rated Deep Fritz 14 x64 arrives at after only a couple of minutes (d=25, and that's without any tablebases). I was wondering whether Houdini saw something all the other engines missed, but it turned out that he prefers the illogical <78.Ra6> and allows the black king to get in front of the white passed pawns. Hmmm, apparently, Houdini is not exactly an endgame wizard, TBs or not TBs...

82...Ke5 - hard to give this move a question mark but it makes no sense at all. <82...Rg8 83.b6 g1Q > [5.68/41 SF] was the only reasonable alternative to an immediate resignation.

After 82...Ke5 the engines report a tablebase win. The final position is mate in 16:

<86.Qe8+ Kd3 87.Qg6+ Ke3 88.Qg3+ Ke2 89.Qxg2+ Ke3 90.Kb2 Rd3 91.a6 Rd2+ 92.Qxd2+ Kxd2 93.a7 Ke3 94.a8=Q Kf4 95.Qa5 Kg4 96.Kc3 Kh4 97.Qf5 Kg3 98.Kd4 Kg2 99.Qg4+ Kh2 100.Ke3 Kh1 101.Kf2 Kh2 102.Qg2#>

One has to give Larsen his due for his persistence and endgame skill. For 47 moves (30-77) the engines report a steadily increasing (from about +0.80 to +1.30, in gradual increments) White advantage, with no significant reversals. By all signs he was quite ready for a third playing session, lol.

Oct-23-15  Howard: Thanks very, very much !

Larsen missed a win on the 18th move ?! Remind me to bring that to Pal Benko's attention---he covered the match for Chess Life & Review back then.

Again, I very much appreciate this !

Oct-23-15  Howard: But 82...Rg8 would also have lost, correct? The tablebase would probably support that contention.
Oct-24-15  Howard: Yes, the tablebase says it's a win...but you probably don't need to use it to see for yourself.
Nov-14-15  NeverAgain: Own quote from a couple of posts ago:

<Here's a small mystery for ya. Komodo concurs with SF; however, Houdini 4 Pro x64 (using the same 6-men Syzygy tablebases as the other two engines) claims that the position after <77...Kf4> is a dead draw! (0.00/37) It gives White a at best with TB turned off. Curiously enough, the much lower-rated Deep Fritz 14 x64 arrives at after only a couple of minutes (d=25, and that's without any tablebases). I was wondering whether Houdini saw something all the other engines missed, but it turned out that he prefers the illogical <78.Ra6> and allows the black king to get in front of the white passed pawns. Hmmm, apparently, Houdini is not exactly an endgame wizard, TBs or not TBs...>

The mystery turned out to be a PEBCAK issue: I didn't have the complete 6-men Syzygy installed (didn't realize that the 6-men downloads don't include the files from the smaller, separate 3+4+5 package at the time). With that corrected, Houdini w/TB performs on par with the other TB-supporting engines.

Nov-30-15  Howard: So, in English, what does that mean? Was there a draw on the 77th move ?

Still hard to believe that Larsen might have missed a win on the 18th move !

Feb-26-16  NeverAgain: In plain English it means that I screwed up the analysis between moves 30 and 77. A steady +1 eval in a rooks-and-pawns endgame that lasts 30+ moves is nonsense, it points to the engine not being able to make any meaningful headway, not that the superior side played clever moves. Although SF7 confirms K9.2's eval of <30...Rb4> as the best move, it is clear that <30...Rg5> didn't throw away the draw either.

<77...Rc8??> was the losing move, giving White a tablebase win that SF can find as early as d=36 with the 6-men Syzygy. I'm sure if you leave it running with TB=off it will eventually resolve a mate in 30 or something, but that could take hours. The point is that <77...Kf4> was a clear draw:

click for larger view

Stockfish 7 64 POPCNT:

77...Kf4 78.Ra6 Rc8 79.Rf6+ Ke5 80.Rf1 Rg8 81.Re1+ Kd5 82.Rd1+ Kc6 83.Rc1+ Kd5 84.Rg1 g4 85.Rg3 Ke5 86.a5 Kf4 87.Rg1 g3 88.b4 Ke4 89.Kb3 Kd5 90.Rg2 Kc6 91.Kc4 Rg4+ 92.Kc3 Kb5 93.a6 Kxa6 = (0.00) Depth: 60/34 00:05:45 6763MN, tb=156716733

I don't know how I managed to squeeze out that 1.37/54 eval from SF in October; running it now I get a dead zero starting with d=28. At d=55 it gives the same PV as SF7 for the first six plies, then diverges with <80...g4>, but the result is the same. Another PEBCAK on my part, I'm afraid; more likely, a whole little series of them. ;)

As for Larsen's chance at move 18, I think my Oct analysis should stand up, although the lines that emerge after <18.Rg4!?> merit a deeper look. Like K9.2, SF7 evaluates the position after <18...Rf7 19.exf6 Qxg3 20.fxg3 g6 21.h4> as , although the PV it gives at the current depth doesn't seem like a real winning attempt:

click for larger view

21...e5 22.Rg5 Raf8 23.Rxe5 Rxf6 24.Re3 Rxf1+ 25.Bxf1 Kf7 26.g4 h6 27.Rb3 Rb8 28.h5 g5 29.Be2 Kf6 30.Kf2 Ke5 31.Kg3 a5 32.Bf3 Kd6 33.Rd3+ Ke6 34.Re3+ Kd6 35.b3 Ra8 36.Rd3+ Ke7 37.Rd1 Rf8 38.Re1+ Kd6 39.Re2 Rc8 40.Rd2+ Ke6 41.Rf2 Ke7 42.Bxc6 dxc6 43.c5 bxc5 44.Rf5 a4 (0.88) Depth: 37/55 00:04:58 4549MN, tb=151614

I'll let it run for a while and then see what H4 Tactical has to say.

Feb-27-16  NeverAgain: Three hours and eight iterations later:

21...a5 22.Rgf4 a4 23.R4f2 Ra5 24.Rd1 Rc5 25.Kh2 b5 26.cxb5 Bxb5 27.Bxb5 Rxb5 28.g4 h6 <snip> (0.90) Depth: 44/65 03:33:34 191496MN, tb=15555995

Looks like Black can hold this ending. The direct <22.h5> doesn't seem to promise more than a small plus either:

22.h5 Raf8 23.hxg6 Rxf6 24.gxh7+ Kxh7 25.Bd3+ Kh8 26.Rxf6 Rxf6 (0.61) Depth: 36/57 00:06:26 6160MN, tb=1610440

and as you skip towards the end of the PV White's advantage shrinks even more (0.46/37).

edit: Houdini 4 Pro x64 Tactical also prefers <18.Rg4 Rf7 19.exf6> to the game continuation (18.exf6) but doesn't see more than a 0.76 plus for White at d=32. Still, it's more than the practically equal position Larsen got with <18.exf6>.

Feb-27-16  Howard: Any possibility that Larsen may have missed a win in the endgame?

In other words, if Portisch threw away the draw by 77...Rc8??, that doesn't mean that Larsen might now have missed a better continuation earlier in the endgame.

Thanks very much, NeverAgain! I played over this game a couple times back in the late 70's.

Apr-17-16  Howard: Just got the excellent book Understanding Rook Endgames, and it reiterates what was said earlier here--77...Kf4! would have held.

But, to reiterate my earlier inquiry, is it possible Larsen had a forced win at some earlier point but overlooked it ?

Premium Chessgames Member
  Honza Cervenka: 77...Rc8 is a loss of tempo in the Pawn race, which is fatal here. Essential line to draw in comparison to text is 77...Kf4 78.Rd5 g4 79.Rd1 g3 80.a5 g2 81.b4 Rf2 82.Rg1 Kg3 83.a6 Rf1 84.Rxg2+ Kxg2 = (diagram)

click for larger view

Premium Chessgames Member
  g15713: E. 1.0
Black to move at 77

click for larger view

As others have pointed out only 77...Kf4 draws!

Syzygy also says this the only move to draw:


NOTE: Create an account today to post replies and access other powerful features which are available only to registered users. Becoming a member is free, anonymous, and takes less than 1 minute! If you already have a username, then simply login login under your username now to join the discussion.

Please observe our posting guidelines:

  1. No obscene, racist, sexist, or profane language.
  2. No spamming, advertising, duplicate, or gibberish posts.
  3. No vitriolic or systematic personal attacks against other members.
  4. Nothing in violation of United States law.
  5. No cyberstalking or malicious posting of negative or private information (doxing/doxxing) of members.
  6. No trolling.
  7. The use of "sock puppet" accounts to circumvent disciplinary action taken by moderators, create a false impression of consensus or support, or stage conversations, is prohibited.

Please try to maintain a semblance of civility at all times.

Blow the Whistle

See something that violates our rules? Blow the whistle and inform a moderator.

NOTE: Please keep all discussion on-topic. This forum is for this specific game only. To discuss chess or this site in general, visit the Kibitzer's Café.

Messages posted by Chessgames members do not necessarily represent the views of, its employees, or sponsors.
All moderator actions taken are ultimately at the sole discretion of the administration.

This game is type: CLASSICAL. Please report incorrect or missing information by submitting a correction slip to help us improve the quality of our content.

Featured in the Following Game Collections[what is this?]
Never_Again's analysis
by NeverAgain
from Chess Informant 23 by Chessdreamer
Game 94
from Veliki majstori saha 32 BENT LARSEN (Marovic) by Chessdreamer
#30 77...Kf4 draws! (a- & b-pawns vs. g-pawn)
from g15713's favorite Rook endgames Part I by g15713
IGM Bent Larsen
by 64rutor
Game 3 March 2
from Portisch - Larsen Candidates Quarterfinal 1977 by Tabanus
Game 3
from WCC Index [Portisch-Larsen 1977] by nescio

Home | About | Login | Logout | F.A.Q. | Profile | Preferences | Premium Membership | Kibitzer's Café | Biographer's Bistro | New Kibitzing | Chessforums | Tournament Index | Player Directory | Notable Games | World Chess Championships | Opening Explorer | Guess the Move | Game Collections | ChessBookie Game | Chessgames Challenge | Store | Privacy Notice | Contact Us

Copyright 2001-2021, Chessgames Services LLC