Members · Prefs · Collections · Openings · Endgames · Sacrifices · History · Search Kibitzing · Kibitzer's Café · Chessforums · Tournament Index · Players · Kibitzing

register now - it's free!
Richard Reti vs Jose Raul Capablanca
"A Knight in Capablanca" (game of the day Jul-11-07)
New York (1924)  ·  English Opening: Anglo-Indian Defense. King's Indian Formation (A15)  ·  1-0
To move:
Last move:

Click Here to play Guess-the-Move
Given 72 times; par: 52 [what's this?]

explore this opening
find similar games 7 more Reti/Capablanca games
PGN: download | view | print Help: general | java-troubleshooting

TIP: You can display posts in reverse order, by registering a free account then visiting your preferences page and checking the option "Display newest kibitzes on top."

PGN Viewer:  What is this?
For help with the default chess viewer, please see the Pgn4web Quickstart Guide.

Kibitzer's Corner
< Earlier Kibitzing  · PAGE 7 OF 7 ·  Later Kibitzing>
Jun-30-11  DrMAL: <psmith:> Like many others I am aware of Alekhine's notes from this tournament. He also mentions a nice plan for the Gruenfeld in the same book. Good stuff to learn on the way to mastery. If you are interested in learning, perhaps you should not quote what was written in 1924 as gospel and check for yourself. Engines are good for this, it is why they are used at all levels.

This game is very famous there are many websites with it and even blogs on it. I do not know offhand of some "reference" that purposely states Alekhine was wrong, nor do I care to check as such things are unimportant. If you are "interested in learning" I hope you do so. A more basic reference for learning is Kingscrusher's video

Jun-30-11  Olavi: The real and only point between 26.Rd5 and 26.Nd4 is that every stong player, after compering the resulting positions, immediately chooses the latter, it is a clear win, and easy to play.
Jul-07-11  qqdos: <Olavi> Tarrasch does not agree with you. 26.Nd4? (+0.64/depth 15) is not even third choice. 26.Nh4 (+0.99/16) is second; 26.Ne1 (+0.82/15) is next; with 26.R1d5! (+2.04/16) way out in front. Perhaps our squabbling twins <maxi> and <DrMAL> may care to chew over the above!
Jul-07-11  psmith: <qqdos>
Is Tarrasch the human being, Dr. Siegbert Tarrasch, or is there a computer program I don't know about named "Tarrasch"? We already know that programs set on this position don't choose Nd4 as their first choice or even their second. The question is what more extensive analysis shows. I think <DrMAL> is actually more right on this than I had first thought, though he never provided analysis to back it up -- but examining the position for a while seems to bear out R1d5 as a quicker and more efficient way to win -- though I also think both Nd4 and R1d5 are winning moves.
Jul-07-11  qqdos: <psmith> Tarrasch is a chess engine using an earlier version of Rybka. Rybka4 confirms the supremacy of 26.R1d5! (+2.47/17) and gives 26.Ne1 (1.01/17) as second choice. Reti, who deserves tremendous credit for piling on the pressure and inducing Capa into this historic loss, won so quickly with 26.Nd4? because Capa played a couple of indifferent moves - specifically: 27...Qe5?! (better 27...Rex3); and 30...Ne5?? (better 30...Rxe3) which was instantly punished by 31.R1d5! (3.66/19).
Jul-07-11  DrMAL: <qqdos> Houdini_15a_x64 (lines truncated):

25/76 30:06 9,113,397,661
+4.06 26.R1d5 Bxd5 27.g4
+1.13 26.R6d5 Bxd5 27.Rxd5
+1.07 26.Nh4 Rc7 27.R1d4
+1.04 26.Ne1 Bxg2 27.N1xg2 Qf3
+0.65 26.Nd4 Bxg2 27.Kxg2
+0.60 26.Qe2 Re6 27.Rxe6

Just sayin' LOL

Jul-08-11  qqdos: <DrMAL> Many thanks for the comparative listings. Spot on! I had looked at 26.R6d5 but this allows the BQ to escape with 27...g5. Also 27.g4 doesn't work. My concern for <psmith> is that all this doesn't prove that 26.Nd4(?) would allow Black with best play to prolong the game for many more moves and might even lead to a draw. With 26.R1d5!! the BQ is lost in 2 or 3 moves and caput.
Jul-08-11  psmith: <qqdos>
I think the judgment "the BQ is lost in 2 or 3 moves and caput" is a bit quick -- I agree that on analysis (and the computers support this) White is winning more quickly in the R1d5 line but it is not just because the BQ is lost -- Alekhine saw that variation but misjudged the position after 26. R1d5 Bxd5 27. g4 Bxf3 28. gxh5 Bxh5 29. Bc6. Looking at that position more closely Black's pieces are poorly co-ordinated and that leads to further catastrophic loss fairly quickly. There are of course situations in which a Rook, minor piece and pawn is not a bad trade for a Queen, but this is not one of them, and the "resulting tussle" is not nearly as prolonged or difficult as Alekhine thought. But although the programs may see this, citing their evals doesn't really illuminate things for the human chess player, as much as actually playing through some lines does.

What I find more interesting is the preference for several other moves over Reti's choice of Nd4. I would say that Reti's move is overwhelmingly natural for a human player, but the tactics seem to favor other moves (and that's where the programs beat us). There seem to be a couple of things going on here.

First, 26. R6d5: After 26.... g5 Black's Queen is still very close to being trapped. But I'd like to know how Houdini (or any other program) proves White to have an advantage (more than a pawn) in this position. I don't see the plan, myself.

Second, why are other Knight moves except 26. Nd4 better? -- against 26. Nd4 Bxg2 27. Kxg2 Rxe3! is a surprising resource pointed out by <qqdos>. Black gives up the exchange for a weakened White Pawn structure and activity for his Queen. I am not sure that White doesn't have a win from this position. But the other Knight moves -- to e1 or h4 -- have the result that after 26...Bxg2 27. N(h4 or e1)xg2 the Knight is covering e3 and the communication between the Rooks on the d-file is not broken, so if 27... Rxe3 is pointless, and White is better.

Premium Chessgames Member
  Sastre: After 26.R1d5 g5, 27.g4 Qg6 28.Rxg6+ wins the Black queen.
Jul-08-11  psmith: <Sastre> The question is what happens after 26. R6d5 (that's Rook on Six to d 5) g5. (26. R6d5 is the second choice of the programs after 26. R1d5).
Premium Chessgames Member
  Sastre: <psmith> Houdini's main line after 26.R6d5 g5 is 27.Ne5 Bxd5 28.g4 Nxe5 29.gxh5 Bxg2 30.Kxg2 Rd7 31.Rxd7 Nfxd7 32.Qd4 (d=22/+2.73).
Jul-08-11  psmith: <sastre>

Thanks. But --

Sorry, I've been stupid.

Of course the question is what about 26. R6d5 Bxd5 27. Rxd5 g5.

Premium Chessgames Member
  Sastre: <psmith> 26.R6d5 Bxd5 27.Rxd5 g5 28.Nh4 f6 29.Bf3 Qf7 30.Nhf5 Qg6 31.Nd6 Re6 32.Rd1 Kh8 33.Bg4 (d=22/+0.90).
Jul-09-11  qqdos: <psmith> Thanks for you thoughtful piece. I was trying to convince you that 26.R1d5!! is a quicker win than 26.Nd4, when Black could struggle on for quite a number of moves. Whether human or machine, after 26.R1d5 I believe Black would resign by move 28. or 29. As a human and with hindsight (!), I think Reti might well concede that he did not play the best move. <Sastre> Thanks also for your contributions.
Jul-09-11  DrMAL: <qqdos> Agreed on all accounts. Nice to see some friendly chess banter rather than some sort of personal trashing. When one examines historical evidence, particularly Alekhine's comment that 26.Nd4 was the "most compelling move" it helps to consider that this game was a huge shocker, everyone at that tournament was completely stunned by Reti's win.

In any event, if you and <psmith> and others want to look at another game I have analyzed, I am most interested on comments regarding Judit Polgar vs Karpov, 2003 cheers.

Jun-09-12  Wyatt Gwyon: What a dirty little pun.
Premium Chessgames Member
  whiteshark: Reti's score sheet :
Dec-02-13  erniecohen: I didn't see any followup on the suggestion (from 6 years ago) of 27...♖xe3 28. fxe3 ♘e5}, which seems like it might save the game, e.g. 29. ♕c1 ♘g4 30. ♘f3 ♕xb5.
Premium Chessgames Member
  Eduardo Bermudez: Ninety years ago !
Apr-08-14  JimNorCal: <whiteshark>: "Reti's score sheet"

It appears that Reti started to write Alekhine ("Al") and then crossed it out to write "Capablanca".

I guess that even for strong GMs, players on the level of Alekhine, Capa and Lasker all look alike :)

Dec-28-14  thegoodanarchist: Reti takes the first serious win from Capa in about a decade.

Richard could have quoted George Costanza from "Seinfeld", when George was lying about designing the Guggenheim addition:

"It really didn't take me that long."

Mar-30-16  DrGridlock: Some discussion above on the merits of 26 R1d5 vs the game continuation of 26 Nd4. Modern engines (as noted in Kingscrusher's video of the game), confirm a preference for R1d5 against Alekhine's notation from 1924:

"White would have won the queen for a rook, knight and a pawn, but the final tussle in that case would have been much more difficult and tedious than after the best defense possible against the move in the text."

Looking at computer analysis after 26 R1d5:

Richard Reti - Jose Raul Capablanca

click for larger view

White's advantage lies on concrete lines, and not just on an abstract idea of how material values compare. After:

26 ... Bxd5
27 g4 Bxf3
28 gxh5 Bxh5
29 Bc6

Richard Reti - Jose Raul Capablanca

click for larger view

Black's pieces are rather tied up, and he has no defense to some further forced exchanges. Komodo likes an exchange sac for black as a way to unlock his knights and rook, but after:

29 ... Rxe3
30 fxe3 Bg4
31 Bxd7 Nxd7

Richard Reti - Jose Raul Capablanca

click for larger view

Black's rook is still stuck on a7, guarding (and blocked by) his knight on d7. Another forced exchange:

32 Qg2 h5
33 h3 Be6
34 Rxe6 fxe6

Richard Reti - Jose Raul Capablanca

click for larger view

Now White's queen is active, and black's pawns are going to start to fall.

35 Qxg6+ Kf8
36 Qxe6

Black has still not found a more active role for his rook than to sit on a7 guarding his knight, and his pawn on h5 is also going to fall after Qf5+ and Qxh5.

R1d5 is simply the better move for White at move 26 because:

(i) White can force exchanges in the following moves

(ii) Black's rook on a7 is inactive with no way to enter the game

(iii) White's queen is quite active and is soon going to start chomping black's pawns.

Mar-30-16  DrGridlock: <erniecohen: I didn't see any followup on the suggestion (from 6 years ago) of 27...♖xe3 28. fxe3 ♘e5}, which seems like it might save the game, e.g. 29. ♕c1 ♘g4 30. ♘f3 ♕xb5.>

Indeed, 27 ... Rxe3 is black's better continuation. Alekhine's notes here are rather silly: "Of course [black] could not play 27 ... Rxe3, on account of 28 fxe3 Qxd1 29 Nf5." Of course black is busted in this position, but Ne5 is a much better continuation for black at move 28 than Qxd1. After 27 ... Rxe3

Richard Reti - Jose Raul Capablanca

click for larger view

Komodo finds:

28 fxe3 Ne5 (not Alekhine's Qxd1)
29 Qe2 Qxe2+
30 Nxe2 Nc4
31 Kf3 Re7

Richard Reti - Jose Raul Capablanca

click for larger view

Here the game is not "saved" for black - White is clearly winning. But there's still "work to be done" for the victory.

May-26-16  edubueno: Having a better position, Capa made a mistake 14...Cf8? instead of 14...d5!
Sep-18-16  edubueno: 30...Txe3! would be done by Dr. Lasker without hesitation.
Jump to page #    (enter # from 1 to 7)
search thread:   
< Earlier Kibitzing  · PAGE 7 OF 7 ·  Later Kibitzing>
NOTE: You need to pick a username and password to post a reply. Getting your account takes less than a minute, totally anonymous, and 100% free--plus, it entitles you to features otherwise unavailable. Pick your username now and join the chessgames community!
If you already have an account, you should login now.
Please observe our posting guidelines:
  1. No obscene, racist, sexist, or profane language.
  2. No spamming, advertising, or duplicating posts.
  3. No personal attacks against other members.
  4. Nothing in violation of United States law.
  5. No posting personal information of members.
Blow the Whistle See something that violates our rules? Blow the whistle and inform an administrator.

NOTE: Keep all discussion on the topic of this page. This forum is for this specific game and nothing else. If you want to discuss chess in general, or this site, you might try the Kibitzer's Café.
Messages posted by Chessgames members do not necessarily represent the views of, its employees, or sponsors.
Spot an error? Please submit a correction slip and help us eliminate database mistakes!
This game is type: CLASSICAL (Disagree? Please submit a correction slip.)

Featured in the Following Game Collections [what is this?]
barb's favorite games
by barb
Reti spearheads the hypermodern revolution
from A Game for Each Decade by severus6
Reti, Richard (1889-1929)
from 1st Class Masters by PMKnight
Joe Stanley's favorite games
by Joe Stanley
Book of Five Rings' favorite games
by Book of Five Rings
Reti vs. Capablanca, New York 1924
from Calchexas' Famous Games by Calchexas
from chess-end-games by saveyougod
Réti King's Indian
from The Réti System by nkvd
18 May 1924, New York International
from Capablanca loses with the Black pieces by Calli
Game 97
from Garry Kasparov on My Great Predecessors Part 1 by MetalPlastic
A Reti knight in Capablanca
from Immortal games by KokeFischer
from Snorri's favorite games by Snorri
franskfranz's favorite games as white
by franskfranz
Tactics - 1
by obrit
1.Nf3: The opening of the future.
from These were the greatest... by nikolaas
New York 1924
by Benzol
assorted Good games
by rbaglini
The first win against Capablanca in quite a while...
from (1900-2000) Which is the game of the century? by ruylopez900
Capablanca's first loss in 8 years I think...
from simondt's favorite games by simondt
getting a feel for the opening (A)
by fourier
plus 97 more collections (not shown)

home | about | login | logout | F.A.Q. | your profile | preferences | Premium Membership | Kibitzer's Café | Biographer's Bistro | new kibitzing | chessforums | Tournament Index | Player Directory | World Chess Championships | Opening Explorer | Guess the Move | Game Collections | ChessBookie Game | Chessgames Challenge | Store | privacy notice | advertising | contact us
Copyright 2001-2017, Chessgames Services LLC