The most popular moves

[sje]

Symbolic's interactive command processor has the dobm command (display opening book moves). The moves, if any, are presented one per line in order of descending popularity. Each move is followed by its win/lose/draw/total counts and then its scoring expectation.

The book in use was generated from a collection of over two million games coming from several different sources. For the initial array. the dobm command produces:

Code: Select all

e4 [446,009/355,049/313,166/1,114,224] 0.540818
d4 [291,726/211,519/231,027/734,272] 0.554617
Nf3 [66,111/47,277/62,325/175,713] 0.553593
c4 [65,668/48,149/55,078/168,895] 0.551864
g3 [6,399/4,881/5,670/16,950] 0.544779
f4 [5,592/6,036/3,541/15,169] 0.485365
b4 [5,807/5,215/2,683/13,705] 0.521598
b3 [3,333/3,134/2,334/8,801] 0.511306
Nc3 [2,759/2,296/1,665/6,720] 0.534449
e3 [810/1,036/378/2,224] 0.449191
g4 [889/797/258/1,944] 0.523663
d3 [437/510/266/1,213] 0.469909
a3 [394/382/198/974] 0.50616
c3 [231/241/142/614] 0.491857
h3 [86/100/43/229] 0.469432
h4 [91/102/28/221] 0.475113
a4 [87/77/23/187] 0.526738
f3 [83/87/14/184] 0.48913
Nh3 [53/82/21/156] 0.407051
Na3 [49/39/10/98] 0.55102

What happens if the most popular move is played until the book is exited? See:

Code: Select all

1 e4 c5 2 Nf3 d6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 a6 6 Bg5 e6 7 f4 Be7 8 Qf3 Qc7 9
O-O-O Nbd7 10 g4 b5 11 Bxf6 Nxf6 12 g5 Nd7 13 f5 Nc5 14 f6 gxf6 15 gxf6 Bf8 16
Rg1 Bd7 17 Rg7 Bxg7 18 fxg7 Rg8 19 e5 O-O-O 20 exd6 Qb6 21 Ne4 Bc6 22 Nxc6 Qxc6
23 Qc3 Qxe4 24 Qxc5+ *

[d]
What is the most popular sequence from your book?

[sje]

Symbolic can also display the book moves in order of descending scoring expectation. The scoring expectation is given by the number of wins plus one half the number of draws with this sum divided by the total number of occurrences. Data is presented from the viewpoint of the side-to-move.

For the initial array:

Code: Select all

d4 [291,726/211,519/231,027/734,272] 0.554617
Nf3 [66,111/47,277/62,325/175,713] 0.553593
c4 [65,668/48,149/55,078/168,895] 0.551864
Na3 [49/39/10/98] 0.55102
g3 [6,399/4,881/5,670/16,950] 0.544779
e4 [446,009/355,049/313,166/1,114,224] 0.540818
Nc3 [2,759/2,296/1,665/6,720] 0.534449
a4 [87/77/23/187] 0.526738
g4 [889/797/258/1,944] 0.523663
b4 [5,807/5,215/2,683/13,705] 0.521598
b3 [3,333/3,134/2,334/8,801] 0.511306
a3 [394/382/198/974] 0.50616
c3 [231/241/142/614] 0.491857
f3 [83/87/14/184] 0.48913
f4 [5,592/6,036/3,541/15,169] 0.485365
h4 [91/102/28/221] 0.475113
d3 [437/510/266/1,213] 0.469909
h3 [86/100/43/229] 0.469432
e3 [810/1,036/378/2,224] 0.449191
Nh3 [53/82/21/156] 0.407051

Note that 1 e4 is sixth on the list. Not a big support for Fischer's famous comment "P-K4, best by test".

[Kempelen]

I suppose all popular positions of our books would be a result of a Sicilian defense. This is the result in my book:

[d]r2qnrk1/3nbppp/3pb3/5PP1/p2NP3/4B3/PPpQ3P/1K1R1B1R w - -

[Ajedrecista]

Hello Steven:

Symbolic can also display the book moves in order of descending scoring expectation. The scoring expectation is given by the number of wins plus one half the number of draws with this sum divided by the total number of occurrences. Data is presented from the viewpoint of the side-to-move.

For the initial array:

Code: Select all

d4 [291,726/211,519/231,027/734,272] 0.554617
Nf3 [66,111/47,277/62,325/175,713] 0.553593
c4 [65,668/48,149/55,078/168,895] 0.551864
Na3 [49/39/10/98] 0.55102
g3 [6,399/4,881/5,670/16,950] 0.544779
e4 [446,009/355,049/313,166/1,114,224] 0.540818
Nc3 [2,759/2,296/1,665/6,720] 0.534449
a4 [87/77/23/187] 0.526738
g4 [889/797/258/1,944] 0.523663
b4 [5,807/5,215/2,683/13,705] 0.521598
b3 [3,333/3,134/2,334/8,801] 0.511306
a3 [394/382/198/974] 0.50616
c3 [231/241/142/614] 0.491857
f3 [83/87/14/184] 0.48913
f4 [5,592/6,036/3,541/15,169] 0.485365
h4 [91/102/28/221] 0.475113
d3 [437/510/266/1,213] 0.469909
h3 [86/100/43/229] 0.469432
e3 [810/1,036/378/2,224] 0.449191
Nh3 [53/82/21/156] 0.407051

Note that 1 e4 is sixth on the list. Not a big support for Fischer's famous comment "P-K4, best by test".

Please note that 1.- Na3 does not have a significant number of games (the error bars for 95% confidence are larger than the reported white advantage; I also have doubts about the ratings of those players). The same for other moves like 1.- f3. So, in view of your data, I would not claim that 1.- e4 is sixth in winning chances [(wins + draws/2)/games] by white. Anyway, I find your data very interesting. Please keep up your good work!

Regards from Spain.

Ajedrecista.