Chess test

This arose out of a thread in General Questions (Luck and skill in card and board games) discussing luck in games. I’d like to establish just how lucky you have to be to randomly find good moves at chess.

So I play White and anyone can play a Black move. I’ll give my assessment each time of every Black move, so we get an approximation of the above chance.

Hope that makes sense!

  1. e2-e4

Possible replies total 20.

Very good replies (4):
1… e7-e5
1… c7-c5
1… e7-e6
1… c7-c6

Good replies (4):
1… d7-d6
1… g7-g6
1… Ng8-f6
1… d7-d5

Reasonable replies (2):
1… b7-b6
1… Nb8-c6

Poor replies (8):
1… h7-h6
1… h7-h5
1… a7-a6*
1… a7-a5
1… f7-f6
1… Nb8-a6
1… Ng8-h6
1… g7-g5

Blunders (2):
1… b7-b5
1… f7-f5

Random chance of good or very good choice: 40%

So please pick a move (preferably a good or very good, and I’ll see how long I can keep this up!

*yes, I know Miles beat Karpov with this, but it was a one-off!

  1. … c7-c5
    Of course to compile some data you could pit a random-move player against a decent computer program for many hundreds of games. Even playing, say, 3-minute games the random player would have almost no chance in my opinion.

Although what’s going on here is more interesting – what is the probability of making a good move at any particular point in a game? I’d like to see the game at least make it past the opening … my expectation is that as a game progresses, the margin of error gets tighter and tighter. The tolerance in some endgames is very tight, plus there may be more possible bad moves on a more open board.

Thanks, panamajack.

  1. e2-e4 c7-c5
  2. c2-c3

Possible replies total 22.

Very good replies (3):
2… d7-d5
2… Ng8-f6
2… e7-e6

Good replies (2):
2… d7-d6
2… Nb8-c6

Reasonable replies (4):
2… b7-b6
2… g7-g6
2… Qd8-a5
2… e7-e5

Poor replies (10):
2… h7-h6
2… h7-h5
2… g7-g5
2… f7-f6
2… a7-a6
2… a7-a5
2… Nb8-a6
2… Ng8-h6
2… Qd8-b6
2… Qd8-c7

Blunders (3):
2… b7-b5
2… f7-f5
2… c5-c4

Current random chance of good or very good choice: 5/22, say 23%.

Cumulative random chance of good or very good choice:
40% (move 1) * 23% - say 9% by move 2.

Black to move.

Please point out errors.
(I’m going to round off to the nearest integer - there’s not enough accuracy to do otherwise in my opinion).

Let’s go Nf6…

Thanks, Bill H.

  1. e2-e4 c7-c5
  2. c2-c3 Ng8-f6
  3. e4-e5

Possible replies total 22.

Very good replies (1):
3… Nf6-d5

Good replies (0):

Reasonable replies (0):

Poor replies (1):
3… Nf6-g8

Blunders (20):
12 pawn moves
5 knight moves
3 queen moves

Current random chance of good or very good choice: 1/22, say 5%.

Cumulative random chance of good or very good choice:
40% (move 1) * 23% * 5% - say 0.5% by move 3.

Black to move.

I’ll add the average number of moves Black has each time.
So far it’s (20 + 22 + 22) / 3, say 21

Good thread glee - I’ll play Nf6-d5!

Thanks, glee :confused:

Oh, and there was a rook move (Rh8-g8) on move 3 I missed. (I’m a chessplayer, not a random move generator!)
It joins the other blunders, but the maths is still OK.

  1. e2-e4 c7-c5
  2. c2-c3 Ng8-f6
  3. e4-e5 Nf6-d5
  4. d2-d4

Possible replies total 27.

Very good replies (2):
4… c5xd4
4… e7-e6

Good replies (2):
4… d7-d6
4… Nb8-c6

Reasonable replies (0):

Poor replies (1):
4… Qd8-a5

Blunders (21):
11 pawn moves
8 knight moves
1 rook move
2 queen moves

Current random chance of good or very good choice: 4/27, say 15%.

Cumulative random chance of good or very good choice:
40% (move 1) * 23% * 5% * 15% - say 0.07% by move 4.

Black to move.

Average number of moves for Black so far:
(20 + 22 + 23 + 27) / 4, say 23.

Interesting idea, glee. Let’s try

  1. e2-e4 c7-c5
  2. c2-c3 Ng8-f6
  3. e4-e5 Nf6-d5
  4. d2-d4 e7-e6

Thanks, Cabbage.

  1. e2-e4 c7-c5
  2. c2-c3 Ng8-f6
  3. e4-e5 Nf6-d5
  4. d2-d4 e7-e6
  5. Ng1-f3

Possible replies total 33.

Very good replies (2):
5… c5xd4
5… d7-d6

Good replies (2):
5… b7-b6
5… Nb8-c6 (sharp play!)

Reasonable replies (3):
5… Bf8-e7
5… f7-f5
5… Qd8-a5

Poor replies (14):
6 pawn moves
4 knight moves
1 rook move
3 queen moves

Blunders (12):
5… b7-b5
5… c5-c4
5… Nd5-b4
5… Nd5xc3
5… Nd5-e3
5… Nd5-f4
5… Nd5-f6
5… g7-g5
5… Qd8-f6
5… Qd8-g5
5… Qd8-h4
5… Bf8-d6
Current random chance of good or very good choice: 4/33, say 12%.

Cumulative random chance of good or very good choice:
0.4 (move 1) * 0.23 * 0.05 * 0.15 * 0.12 - say 0.00008 by move 5.
I switched because I’m worried about losing a decimal place. Do you prefer the above, or 0.008%?

Black to move.
(You may like to try a good move, as opposed to a very good move.)

Average number of moves for Black so far:
(20 + 22 + 23 + 27 + 33) / 5, say 25.

  1. e2-e4 c7-c5
  2. c2-c3 Ng8-f6
  3. e4-e5 Nf6-d5
  4. d2-d4 e7-e6
  5. Ng1-f3 Nb8-c6

I love Knights.

Thanks, erislover.

  1. e2-e4 c7-c5
  2. c2-c3 Ng8-f6
  3. e4-e5 Nf6-d5
  4. d2-d4 e7-e6
  5. Ng1-f3 Nb8-c6

Possible replies total 37.

Very good replies (2):
6… Nd5-b4
6… Qd8-a5+

Good replies (0):

Reasonable replies (2):
6… Nd5-b6
6… Nd5-c7

Poor replies (0):

Blunders (33):
12 pawn moves
11 knight moves
2 bishop moves
2 rook moves
6 queen moves

Current random chance of good or very good choice: 2/37, say 5%.

Cumulative random chance of good or very good choice:
0.4 (move 1) * 0.23 * 0.05 * 0.15 * 0.12 * 0.05 - say 0.000004 by move 6 (or 0.0004%).

The chance of winning the UK lottery 1st prize is 14 million to 1. I make that about 0.0000007.
What is the chance of being hit by a meteorite?

Black to move.
(This is a sharp variation I analysed some years ago. I wouldn’t try it in a tournament game.)

Average number of moves for Black so far:
(20 + 22 + 23 + 27 + 33 + 37) / 6, say 27.

Wanna share white’s move with us? :stuck_out_tongue:

I held up the thread :eek:

I played the move on my board at home - how come I didn’t type it :confused:

I’ve kept you all in suspense :o

Actually, you could have worked it out from the other information :cool:

  1. e2-e4 c7-c5
  2. c2-c3 Ng8-f6
  3. e4-e5 Nf6-d5
  4. d2-d4 e7-e6
  5. Ng1-f3 Nb8-c6
  6. c3-c4
  1. e2-e4 c7-c5
  2. c2-c3 Ng8-f6
  3. e4-e5 Nf6-d5
  4. d2-d4 e7-e6
  5. Ng1-f3 Nb8-c6
  6. c3-c4 Qd8-a5+

Thanks, Sxyzzx.

  1. e2-e4 c7-c5
  2. c2-c3 Ng8-f6
  3. e4-e5 Nf6-d5
  4. d2-d4 e7-e6
  5. Ng1-f3 Nb8-c6
  6. c3-c4 Qd8-a5+
  7. Bc1-d2

Possible replies total 40.

Very good replies (1):
6… Nd5-b4

Good replies (0):

Reasonable replies (0):

Poor replies (0):

Blunders (39):
11 pawn moves
13 knight moves
2 bishop moves
2 rook moves
11 queen moves

Current random chance of good or very good choice: 1/40, say 3%.

Cumulative random chance of good or very good choice:
0.4 (move 1) * 0.23 * 0.05 * 0.15 * 0.12 * 0.05 * 0.03 - say 0.0000001 by move 7 (or 0.00001%).

The chance of winning the UK lottery 1st prize is 14 million to 1. I make that about 0.0000007.
What is the chance of being hit by a meteorite?

Black to move.
Since Black’s move is forced, I’ll reply myself.

Average number of moves for Black so far:
(20 + 22 + 23 + 27 + 33 + 37 + 40) / 7, say 29.

Thanks, me. :cool:

  1. e2-e4 c7-c5
  2. c2-c3 Ng8-f6
  3. e4-e5 Nf6-d5
  4. d2-d4 e7-e6
  5. Ng1-f3 Nb8-c6
  6. c3-c4 Qd8-a5+
  7. Bc1-d2 Nd5-b4
  8. d4-d5

Possible replies total 35.
Also I’ve been forgetting King moves! Never mind, using the rounding level I’ve chosen, I’m probably still accurate enough.

Very good replies (2):
8… Nc6-d4
8… e6xd5

Good replies (0):

Reasonable replies (0):

Poor replies (0):

Blunders (33):
10 pawn moves
9 knight moves
2 bishop moves
2 rook moves
8 queen moves
2 king moves

Current random chance of good or very good choice: 2/35, say 6%.

Cumulative random chance of good or very good choice:
0.4 (move 1) * 0.23 * 0.05 * 0.15 * 0.12 * 0.05 * 0.03 * 0.06 - say 0.000000007 by move 8 (or 0.00000017%).

The chance of winning the UK lottery 1st prize is 14 million to 1. I make that about 0.0000007.
So by move 8, you’re 100 times more likely to have won the UK lottery than to have reached a good position! (And the random player is facing an experienced player who has analysed the whole opening for several moves further on.)

Black to move.

Average number of moves for Black so far:
(20 + 22 + 23 + 27 + 33 + 37 + 40 + 35) / 8, say 30.

  1. Bump
  2. I’m satisfied I proved my point, but I’ll carry on if anyone wants to.