-
-
Save jake9696/ddc7149a84761a94cb10f01721e9164e to your computer and use it in GitHub Desktop.
| Chess notes |
notes on the gifs - created a gif of the full game with chess.com's gif creator, then used ezgif.com to Resize from 720x720 to 360x360, then Cut snippets from the full game
Here's a game I played recently as white, that I should have won but didn't.
- d4 d5
- c4 e6
- Nc3 Nf6
up to this point, it's a Queen's Gambit Declined opening
- Nf3 Be7
here white is better to play 4. cxd5, probably black will recapture ..Nxd5 then 5. Nf3
- Ne5 O-O
stockfish says white has made a mistake and given away the advantage; 5. Bf4 or 5. cxd5 are better - I (white) saw an outpost for my knight and thought it would be a good resource later in the game
- e3 Nbd7
white plays e3 to open up the light square bishop; stockfish prefers 6. g3
after ..Ndb7, as white I was thinking if I found another useful move, letting black play ..Nxe5 8. dxe5 would leave me with an advanced (good), doubled (bad) pawn that could eventually be defended with f4, and was threatening black's other knight
- b3 Nxe5
- dxe5 Nd7
- f4 Bb4
my brilliant plan has come to pass, but stockfish reckons white is only deeper in a hole, close to -1 now
black is threatening white's hanging knight on c3, which will come with a fork on the king and rook
- Bb2 b6
there are three obvious ways for white to defend (Qc2, Bd2, Bb2); I chose Bb2 because I was looking at the b1-h7 diagonal for the queen which Qc2 would have pulled the queen away from due to having to recapture on c3, and I also prefer to finachetto my bishop ruling out Bd2
- a3 Bxc3+
white provokes Bxc3+ when that should have been avoided
- Bxc3 Nc5
- Bb4 Re8
- Bd3 a5
white has given away a slight advantage here with 14. Bd3, but black gives it back by not checking white's king and using that to open the d file
- Bxc5 bxc5
white cannot play Bd2 as it blocks the queen's defense of the light-square bishop, and avoiding the trade with Bc3 allows black time to play ..dxc4 and open the d file under favorable conditions
- Qg4 dxc4
stockfish prefers 16. Qc2 to create a battery with the queen and bishop, but as white I was thinking to triangulate rather than create a battery
now is the wrong time for black to take on c4, since it gives white's bishop an outpost there without any pawns to threaten it along the b or d files
- Bxc4 Bb7
as white, I am thinking that I would rather castle into black's fianchetto'd bishop than have to anchor the queen to defending the g2 pawn, and this also improves my ability to threaten black's queen on d8 and take control of the d file
- O-O Be4
- Rad1 Qe7
stockfish thinks black should have put pressure on the b3 pawn with Qb8, but it's hard to turn down a chance to connect your rooks
- Bd3 Bxd3
as white, I am thinking that a trade on d3 will leave me with a rook there, ready to double up on the d file with Rfd1
- Rxd3 Rab8
- Rfd1 Qf8
- Qh4 Kh8
black should probably not put their king in line with white's queen; it was probably better to push the a pawn
- Rd7 h6
as white, I was thinking a rook on the 7th rank would be an excellent resource, with the pawn on c7 looking vulnerable and giving a chance to get the other rook on the 7th rank as well, but stockfish much prefers white to move their queen to f2 or e1
black could have capitalized on this with ..Rxb3 or ..c4, with the plan of splitting white's rook pair
- Rxc7 Rxb3
white could have played Qh5 instead, to increase pressure on f7 and, after white does play Rxc7, the threat of both rooks on the 7th rank would be deadly
- Rdd7 Rb1+
black could have played ..c4 to open an attack by the queen on white's a3 pawn, and gotten some counterplay with the queen and rook on the 3rd rank, but ..Rb1+ will run out of checks too soon
- Kf2 Rb2+
- Kf3 Qg8
white's king is safe for now, and black can't hold off white's rooks any longer
- Rxf7 Rb3
- g4 c4
white could have kept putting pressure on black's position with Qg4, as the endgame with all the pieces traded off should be an easy win for white as long as the black pawn on c5 can be dealt with
as white, I was thinking that black didn't have enough time to get their c or a pawns down the board before I could use the g pawn to break up black's pawn fence in front of their king
- g5 Qh7
- Rxg7 Qxg7
white could have played 32. f5 to provoke an exchange on f5 and leave a black pawn there to prevent the black queen's escape after Rxg7, and won a rook in the exchange
- Rxg7 Kxg7
- Qxh6+ Kf7
- g6+ Ke7
as white, my plan here was to keep checking black's king and using tempo where possible to get the g pawn to the back rank and force black to sacrifice a rook
- Qg5+ Kd7
since black couldn't play Kd6 due to white's pawn on e5, white should have played 36. g7 instead, to force the rook sacrifice where white could play Qh7+ and pick up the rook on g8
black's only other legal move, ..Kf8, is a mate in 3 (36. ..Kf8 37. Qf6+ Kg8 38. Qf7+ Kh8 39. Qh7#)
- g7 Rbb8
now that black has both rooks on their back rank, white has lost the chance to force a rook sacrifice
- f5 exf5
- Qxf5+ Kc7
- Qf7+ Kb6
- Qxc4 Rbc8
white has finally dealt with black's only passed pawn
- Qd4+ Kb7
- h4 Rg8
as white, since I missed forcing a rook sacrifice, now I'm thinking I should connect the g and h pawns; surprisingly, stockfish agrees with this rather than throwing out further checks on the black king
- e6 Rce8
white's e6 opens up the queen to defend the pawn on g7 and creates another pawn threat, trying to overwork black's rooks
- Qd7+ Kb6
- h5 Rd8
white's h5 will work, but 46. e7 would have really put black's rooks in a pickle
- Qe7 Rde8
for white, 47. Qf7 would have been more accurate, but black walks into the rook pickle anyway
- Qd6+ Kb5
- h6 Rd8
- Qe5+ Kc6
much better for white was 50. h7, because it forces black to either take white's queen or the g pawn, and lose one of the rooks while white will keep at least one queen
- h7 Rd5
stockfish sees a mate in 11 if white had played 51. e7 instead
- Qc3+ Kd6
white has mate in 5 here with (53. hxg8=Q Rf5+ 54. Ke2 Ke7 55. Qgc8 Rf2+ 56. Kxf2 a4 57. Qd7#)
- hxg8=Q Rf5+
- Ke2 Rc5
black has diverged but white still has mate in 3 (e.g., 55. Qd8+ Kc6 56. Qc8+ Kb5 57. Qb3#)
- Qxc5+ Kxc5
not as quick for white but should still be mate in 5 with (56. Qc8+ Kb5 57. g8=Q Ka4 58. Qc2+ Kb5 59. Qb8+ Ka6 60. Qd3#)
- Qf8+ Kb5
white's inaccuracy wastes a move but mate in 5 is still on the board
- g8=Q a4
- Qb4+ Kc6
- e7 Kc7
white could have played any check by the queen on the 8th rank and had mate in 2 (e.g., 59. Qa8+ Kc7 60. Qc5#), but e7 should still only delay it as long as white plays accurately (60. Qd8+ Kc6 61. e8=Q#)
- e8=Q 1/2-1/2
white nooooooooooo!
In this game, white barely escapes checkmate along the a file. White was JimWest and black was Neil1454 on chess.com.
- e4 c5
- Nf3 Nc6
- d4 cxd4
- Nxd4 g6
- Nc3 Bg7
- Be3 d6
- Bc4 Nf6
- f3 Nd7
- Bb3 O-O
- Qd2 Nc5
- O-O-O Nxb3+
- axb3 Qa5
- Kb1 Nb4
- Bh6 {Critical move.}
14... Bd7 - Bxg7 Kxg7
- h4 Rac8
- Nde2 Bb5? ± {MISTAKE (+2.30)} ({(+0.10) The best move was} 17... Rc6 18. h5 Ra6 19. Na4 g5 20. h6+ Kh8 21. Nec3 f6 22. Nd5 Nxd5 23. Qxa5 Rxa5 24. Rxd5 Rxd5 25. exd5 Bxa4 26. bxa4 Kg8 27. a5)
- Nc1 {Critical move.}
18... b6 {INACCURACY (+3.54)} ({(+2.50) The best move was} 18... f6 19. g4 b6 20. f4 Bc6 21. Rhe1 Kg8 22. f5 Rf7 23. Rg1 Qe5 24. Rde1 a5 25. g5 Bd7 26. gxf6 Qxf6) - h5 Rh8
- hxg6?? {BLUNDER (+1.48)} ({(+5.57) The best move was} 20. e5 d5 21. Rh4 Be8 22. hxg6 fxg6 23. Qh6+ Kg8 24. Qf4 h5 25. Qxb4 Qxb4 26. Rxb4 e6 27. Rbd4)
20... hxg6 - Rxh8 Rxh8
- Nxb5 = {INACCURACY (+0.22) Critical mistake.} ({(+1.58) The best move was} 22. Qd4+ f6 23. Nd5 Nxd5 24. exd5 Qa6 25. g4 Qb7 26. Re1 Rh4 27. Qe3 Kf8 28. Qe4 Be8 29.
Nd3 Rh2 30. Nb4 g5 31. Nc6 Bxc6 32. dxc6)
22... Qxb5 - Qc3+ Kg8
- Qc8+ Kg7
- Qc7 Qa6?? ± {BLUNDER (+6.25)} ({(-0.59) The best move was} 25... Qc6 26. Qxc6 Nxc6 27. c3 Rh2 28. Rd2 f5 29. exf5 gxf5 30. Nd3 e5 31. b4 Kf7 32. Ka2 f4 33. b5 Ne7 34. Nb4 Nf5 35. Nc6 Ke6 36. Nxa7 d5 37. Nc6)
- Qc3+ f6
- Qxb4 Rc8
- Qa4? {MISTAKE (+4.81)} ({(+7.36) The best move was} 28. e5 Qb7 29. exf6+ exf6 30. Rxd6 Rf8 31. Nd3 Qe7 32. Rxb6 Qxb4 33. Rxb4 f5 34. Rb7+ Rf7 35. Rxf7+ Kxf7 36. Ne5+ Ke6 37. Nxg6 a5 38. Kc1 Kd6)
28... Qb7 {INACCURACY (+6.62)} ({(+4.81) The best move was} 28... Qxa4 29. bxa4 Rh8 30. a5 Kf7 31. e5 dxe5 32. c4 Ke8 33. b4 g5 34. c5 bxc5 35. b5 g4 36. Nd3 gxf3 37. gxf3 Kd7 38. Nb2+ Ke6) - Qd4 Qc6
- Qd2 a5
- Rh1 g5
- Ne2 a4
- Nd4 Qd7
- Nf5+ Kg8
- g4 Qc7?? {BLUNDER (Mate in 5)} ({(+11.86) The best move was} 35... Qe8 36. bxa4 Ra8 37. Rh6 Ra7 38. b3 Ra5 39. Qh2 Rxf5 40. Rh8+ Kf7 41. Qh5+ Ke6 42. exf5+ Ke5 43. Qxe8)
- Qd5+ (36. Qd5+ e6 37. Qxe6+ Qf7 38. Nh6+ Kh8 39. Qxf7 ... 40. Nf5#)
1-0
Chess variant idea
Human v. Computer
The computer plays on a timed interval, smaller intervals will be more difficult for the human player.
The human plays their next move at any point during the interval between the computer player’s moves.
The computer plays its next move based on the board position after its previous move, not yet accounting for the intervening move of the human player.
If the human player does not move before the interval, they forfeit. As related variations, the human player can choose to skip the turn or bank the turn and move multiple times during one interval by spending banked turns.
There is no draw or stalemate. The human player may choose to resign at any point.
Repetition, either of a position on the board or of a sequence of moves, does not lead to stalemate.
A stalemate is not declared after any number of moves without a pawn move occurring.
A stalemate is not declared if both players have insufficient material to checkmate the opposing player’s king.
The computer player may not resign or skip a turn. The computer player has one second to move on each interval. If it does not move within that second, it forfeits.
Here’s a chess game I recently lost. I think it illustrates the importance of tempo, pressure, and finding checks. I played as black.
(moves 1-5; white builds a strong center, black defends with the Pirc)
(moves 6-13)
(move 14; white blunders a bishop)
(moves 15-34; black has been trying to press the advantage, white should have grabbed the open h file on move 34)
(Rh3 is better for white on either move, 36... Nd7 is how black should have saved the knight)
(black wanted an in-between move to create a passed pawn before capturing white's bishop, but black's blunder gives white tempo with a string of checks)
(moves 38-40; white blunders away the tempo and now black has the string of checks, with mate in 9 if played accurately: 40… Qxf4+ 41. Kg1 Qd4+ 42. Kf1 Qxd1+ 43. Kf2 Qd4+ 44. Kf3 Qe4+ 45. Kf2 Ra2+ 46. Kg3 Rxg6+ 47. Kh3 Qxg2+ 48. Kh4 Qg4#/Qh1#/Qh2#)
(moves 41-46; with e6, white is threatening mate in 1 with Qf7#, but black misses the threat; black has mate in 2 now with 47. Kxg2 Ra2+ 48. Kf1 Qd1# but misses that as well)
(move 47; black traded rooks and no longer has enough material to checkmate the white king without Ra2+, but at least this keeps the checks going)
(move 48; back to mate in 3 for black, but black doesn’t find it: 49. Ke3 Qe4+ 50. Kf2 Ra2+ [recurring theme since move 44 — black should get that back rank rook into the action] 51. Kf1 Qb1#)
(move 49; so far, so good…)