White 280 vs Black 21
White starts from setup 280 (NBRKBNRQ), Black from setup 21 (NQBBNRKR). In the encyclopedia since 2026-10-15: neither side starts better.
|cp| ≤ 5 at 400M · 800M · 1.6B · 4B nodes, two independent engines agree
How to play with different setups
White sets up the pieces by position No. 280, Black by position No. 21, mirrored on their own side; the pawns stand as usual.
Moves as usual: White starts; everything else follows the Chess960 rules.
The two setups
White: setup #280 N B R K B N R Q
King on the d-file, rooks on c, g, queen on the h-file; no bishop starts in a corner; the knights start away from the centre. Castling follows the Chess960 rule: king and rook end on the classical squares. In ordinary symmetric play setup 280 gives White +0.41 (Morozov Balance Index, 400M nodes).
Black: setup #21 N Q B B N R K R
King on the g-file, rooks on f, h, queen on the b-file; no bishop starts in a corner; the knights start away from the centre. The king already stands on a castled square, so one of the castlings moves only the rook. In ordinary symmetric play setup 21 gives White +0.29 (Morozov Balance Index, 400M nodes).
Why the pair is balanced
At the deepest rungs (400M, 800M, 1.6B nodes) the evaluation is exactly 0.00 each time, reaching search depth 57.
Admission rule: |cp| ≤ 5 at four budgets up to 4 billion nodes, no one-sided drift along the ladder, and confirmation by two independent engines — Berserk and Lc0 — each below its own median on symmetric positions. The same corridor is required at every deeper budget (8 G, 16 G) at which the pair has been evaluated.
The shallow rungs wandered between -27 and +28 cp before settling — a reminder that quick engine checks overstate imbalances, and why the ladder goes deep.
For comparison, in symmetric play setup 280 gives White +0.41 and setup 21 gives White +0.29: with both sides on the same rank the first move is worth something. In this pair Black's own setup absorbs that edge completely.
The engine's first choice for White is g4; the alternatives are slightly weaker but keep the game level.
Along the engine's main line (30 moves) the evaluation stayed within ±16 cp the whole way: the balance holds as the pieces come out, not only on move one.
Engine self-play agrees: 10 games, White won 0, drawn 10, Black won 0 (White scored 50.0%).
Verification ladder
Nine independent searches of the starting position, each with a larger node budget. Evaluation in centipawns from White's point of view.
| Stage | Nodes | Depth | Eval, cp | Main line |
|---|---|---|---|---|
| S0 | 100 000 | 17 | -27 | 1. c4 h5 2. Ne3 c6 3. d4 Nf6 4. g3 Nb6 |
| S1 | 500 000 | 21 | -2 | 1. g4 c5 2. Nb3 c4 3. Nd4 b5 4. O-O-O Nb6 |
| S2 | 2 000 000 | 23 | +28 | 1. g4 Nb6 2. c3 d5 3. O-O-O c5 4. Nb3 Qc7 |
| S3 | 8 000 000 | 26 | +10 | 1. g4 Nb6 2. c4 c5 3. h4 h5 4. g5 g6 |
| S4 | 32 000 000 | 31 | +14 | 1. g4 c6 2. c4 h5 3. gxh5 Nb6 4. Ng3 d5 |
| S5 | 128 000 000 | 38 | 0 | 1. g4 Nb6 2. f4 c6 3. Bg3 h5 4. gxh5 Nf6 |
| S6 | 400 000 000 | 52 | 0 | 1. g4 Nb6 2. f4 c6 3. Bg3 h5 4. gxh5 Nf6 |
| S7 | 800 000 000 | 51 | 0 | 1. g4 Nb6 2. f4 c6 3. Bg3 h5 4. gxh5 Nf6 |
| S8 | 1 600 000 000 | 57 | 0 | 1. g4 Nb6 2. f4 c6 3. Bg3 h5 4. gxh5 Nf6 |
Deep verification
The search ladder reaches 1.6 billion nodes; admission needs a 4-billion rung and two further engines on top of that. Beyond even that, every pair was re-checked at budgets up to ten times larger and in thousands of self-play games. This section shows what each check returned for this pair.
| Nodes | Eval, cp |
|---|---|
| 4 000 000 000 | 0 |
| 8 000 000 000 | not checked |
| 16 000 000 000 | not checked |
This pair stays within the ±5 cp corridor up to 4 000 000 000 nodes.
Independent engines, run on the same position:
- Berserk 14: 0
- Lc0 0.32.1: -15
Independent confirmations: 3 of 3. An engine counts as confirming when its reading of the pair is closer to equality than its own reading of a typical symmetric Chess960 position (+10 for Berserk, +18 for Lc0). Where only one engine confirms, the pair is balanced according to that engine alone, and this page says so.
Own weight of each setup, measured against the classical array: White +43, Black +33, difference +10. A balanced pair needs White's setup to be weaker by about the price of the first move.
Extended self-play: 10 games, 0 White wins, 10 draws, 0 Black wins, White scored 50.0%.
Best first moves
The 3 best first moves for White, searched together at 400 000 000 nodes.
| # | Eval, cp | Depth | Main line |
|---|---|---|---|
| 1 | +15 | 40 | 1. g4 c6 2. c4 Nb6 3. Ne3 h5 4. g5 g6 5. h4 Bc7 |
| 2 | 0 | 40 | 1. c4 h5 2. d4 c5 3. dxc5 b6 4. Nb3 Nf6 5. g3 bxc5 |
| 3 | -17 | 40 | 1. O-O-O h5 2. c3 c6 3. e4 d5 4. exd5 cxd5 5. d4 Nb6 |
Main-line play-out
30 moves, 5 000 000 nodes per half-move. Largest |evaluation| 16 cp, mean 1.1 cp, final 0 cp.
| # | White | cp | Black | cp |
|---|---|---|---|---|
| 1 | g4 | +16 | c6 | -8 |
| 2 | c4 | -1 | Nb6 | 0 |
| 3 | Ne3 | 0 | h5 | +4 |
| 4 | g5 | -4 | g6 | -2 |
| 5 | h4 | -1 | Bc7 | -3 |
| 6 | d4 | 0 | Ng7 | 0 |
| 7 | Bb4 | -4 | Re8 | 0 |
| 8 | c5 | -11 | Na8 | 0 |
| 9 | O-O | 0 | Bf4 | 0 |
| 10 | Nb3 | 0 | O-O | 0 |
| 11 | Qf3 | 0 | Bh2+ | -10 |
| 12 | Kg2 | 0 | d5 | 0 |
| 13 | cxd6 | 0 | Bxd6 | 0 |
| 14 | Bxd6 | 0 | exd6 | 0 |
| 15 | Qg3 | 0 | Nb6 | 0 |
| 16 | Nd2 | 0 | Nd5 | 0 |
| 17 | Ndc4 | 0 | Nxe3+ | 0 |
| 18 | fxe3 | 0 | Bf5 | 0 |
| 19 | Nxd6 | 0 | Bxb1 | 0 |
| 20 | Rxb1 | 0 | Re7 | 0 |
| 21 | Qf4 | 0 | Ne6 | 0 |
| 22 | Qg3 | 0 | Ng7 | 0 |
| 23 | e4 | 0 | Rd8 | 0 |
| 24 | Rf6 | 0 | Ne8 | 0 |
| 25 | Nxe8 | 0 | Qxg3+ | 0 |
| 26 | Kxg3 | 0 | Rexe8 | 0 |
| 27 | Rbf1 | 0 | Re7 | 0 |
| 28 | e3 | 0 | Rxe4 | 0 |
| 29 | R1f3 | 0 | Re7 | 0 |
| 30 | Kg2 | 0 | a5 | 0 |
Engine self-play
10 games at 500 000 nodes per move: White won 0, drawn 10, Black won 0 — White scored 50.0%. Average length 50.2 moves.
Sample game 1 (1/2-1/2)
1. g4 Nb6 2. Ne3 h5 3. c4 c6 4. g5 g6 5. h4 d5 6. cxd5 cxd5 7. d4 Nd6 8. b3 Nb5 9. Nac2 e5 10. Nxd5 Be6 11. Nf6+ Bxf6 12. gxf6 exd4 13. Qe4 Bg4 14. O-O-O Re8 15. Qd3 Nc3 16. Bxc3 dxc3 17. Ne3 Qf4 18. Qd4 Qxf2 19. Nxg4 Qxd4 20. Rxd4 hxg4 21. Rgxg4 Rxe2 22. a4 Re1+ 23. Kc2 Re6 24. Rd8+ Kh7 25. Rxh8+ Kxh8 26. Rd4 a5 27. Rd8+ Kh7 28. Kxc3 Rxf6 29. Be4 Kg7 30. Bxb7 Re6 31. Bf3 Kf6 32. Kd4 Ke7 33. Rb8 Rd6+ 34. Kc3 Nd5+ 35. Kc4 Nb6+ 36. Kc3 Nd5+ 37. Kc4 Nb6+
Sample game 2 (1/2-1/2)
1. g4 c5 2. Nb3 b6 3. c3 Bb7 4. e4 Nac7 5. d4 Ne6 6. Ng3 a5 7. d5 Nf4 8. c4 h5 9. gxh5 e6 10. a4 Nxh5 11. Qf3 Nf4 12. O-O-O Nh3 13. Rg2 Nf4 14. Rg1 Nh3 15. Rg2 f5 16. exf5 exd5 17. cxd5 Nf4 18. Rg1 Bg5 19. Bd2 Rh4 20. Ne4 Bh6 21. Bc2 Bxd5 22. Kb1 Qe5 23. Qc3 Rxf5 24. Rde1 Qxc3 25. Nxc3 Be6 26. Bxf5 Bxf5+ 27. Ka1 Nf6 28. f3 Kf8 29. Nc1 d5 30. N1e2 Nd3 31. Bxh6 gxh6 32. Rd1 Nb4 33. Ng3 Nc2+ 34. Ka2 Nb4+ 35. Ka3 Bd3 36. Rd2 d4 37. Nce4 Nfd5 38. Nf2 Nc2+ 39. Ka2 Ndb4+ 40. Kb1 Bh7 41. Ng4 Ne1+ 42. Ka1 Nec2+ 43. Kb1 Ne3+ 44. Ka1 Nbc2+ 45. Kb1 Nb4+ 46. Ka1 h5 47. Nf6 Bd3 48. Ngxh5 Ke7 49. Rg7+ Ke6 50. Rb7 Nec2+ 51. Kb1 Ne3+ 52. Ka1 c4 53. Rxb6+ Ke7 54. Rb7+ Kd6 55. Ne8+ Kc6 56. Rc7+ Kd5 57. Nhf6+ Ke6 58. Nd7 Rf4 59. Nc5+ Kd5 60. Nxd3 cxd3 61. Rc1 Rxf3 62. Nc7+ Ke5 63. Kb1 Nf1 64. Re1+ Kf5 65. Rdd1 Ne3 66. Nb5 Ke4 67. Rd2 Nc6 68. Nc7 Ne5 69. Ne6 Kd5
Play it in the CH960 app
Open this page on a phone with the CH960 app installed, or find pair 280 / 21 in the app's encyclopedia, then choose how to play.
- Local game — two players on one device, with a clock.
- With a mentor — a clearly labelled training bot with hints, at your level.
- Challenge a friend — an online game by link; both players get the same pair, colours swap on rematch.
Play this pair against the computer, right here Play a friend
FEN (Shredder, castling rights by rook files): nqbbnrkr/pppppppp/8/8/8/8/PPPPPPPP/NBRKBNRQ w CGfh - 0 1
Mirror twin: White 589 vs Black 938 — a different position (castling squares are fixed), not evaluated by the search.