Play vs computer

Play chess against a 2000-rated computer

Expert Nadia plays like a real 2000 player. Free, in your browser, no sign-up.

Last updated

touch_appWarm up with a 2000-level tactic, then play the full game.

How the 2000 computer plays

A 2000 player rarely blunders, punishes loose pieces immediately and plays the endgame well. Expect a long game where small advantages matter.

54
average loss per move (centipawns)
2.5
big blunders per 100 moves
4.1s
average think, 10-minute game
8
moves it looks ahead

For comparison, real Chess.com rapid players around 2000 lose about 49 centipawns per move (measured on their own games, the same way). The bot is Expert Nadia (2000): see how the bots are built.

Who this level is for

You are a strong club or tournament player looking for a serious opponent.

A sample game at 2000

  1. 1.d4
  2. d6
  3. 2.Nf3
  4. Nd7
  5. 3.c4
  6. g6
  7. 4.a4
  8. e5
  9. 5.Nc3
  10. Ngf6
  11. 6.e4
  12. Bg7
  13. 7.Be2
  14. O-O
  15. 8.Qc2
  16. c6
  17. 9.h3
  18. Nh5
  19. 10.Be3
  20. exd4
  21. 11.Nxd4
  22. Qh4
  23. 12.Bxh5?
  24. Qxh5
  25. 13.g4
  26. Qa5
  27. 14.O-O-O
  28. Ne5
  29. 15.Nf5?
  30. gxf5
  31. 16.exf5
  32. Nxc4
  33. 17.Bf4
  34. Qb4
  35. 18.Rd3
  36. d5
  37. 19.Re3
  38. a5
  39. 20.Rf3
  40. Ra6
  41. 21.Re1
  42. b6?
  43. 22.b3
  44. Na3
  45. 23.Qe2
  46. Bxc3
  47. 24.Qe7
  48. Qxe7
  49. 25.Rxe7
  50. Bb4
  51. 26.Bd2?
  52. Bxe7
  53. 27.Bf4
  54. Ra8
  55. 28.Rg3
  56. Bb4
  57. 29.Bh6
  58. Bb7
  59. 30.Rd3
  60. f6
  61. 31.Re3
  62. Rfe8
  63. 32.Rxe8+
  64. Rxe8
  65. 33.Bd2
  66. d4
  67. 34.Bg5??
  68. Re1+
  69. 35.Kb2
  70. Bc3+
  71. 36.Ka2
  72. Ra1#
  73. 0-1
The 2000 bot against itself, every move scored by Stockfish: 1 blunder (??). Hover a marked move to see what was better. A sample game, chosen because it contains a mistake.

How to beat a 2000 computer

  1. 1Play openings you know well and aim for positions you understand.
  2. 2Avoid giving it targets: every loose pawn or piece will be attacked.
  3. 3When the position is level, keep pieces on and create imbalances.

How we measured this

Measured on 41 Chess.com players' public games, November 2026.

We picked random public Chess.com players rated 400 to 2000 and took their 5 most recent rapid games. Stockfish scored every move they played from move 5 on, comparing it with the best move, and we averaged the loss per move. We fitted a smooth trend through all the players so no single game sways a rating, then ran our bots through exactly the same measurement.

Correction: our first version used textbook reference values that overstated human mistakes; the bots are now tuned to these measured numbers.