While the intentions are good, Fischer Random 960 is too random, no opening theory can be developed, and the castling rules are arcane. Some opening positions allow white to attack an undefended pawn on the first move.
Fix by placing the rooks and kings are their classical chess squares and placing knights, bishops, and queens on random squares.
Plenty of varied opening positions left, castling rules are same as classical position, and no first move by white attacks an undefended pawn.
Comments?
PS: Does anyone know how to calculate the number of legal positions in this variant?
I calculate as 5 possible unique squares for the queen, 3 for the white squared bishop, 2 for the black squared bishop, 4 squares for one knight, and 3 squares for the other knight.
5 x 3 x 2 x 4 x 3 = 360 possible random positions. Does this seem right?
Plenty of randomness but some logical consistency allowing some theory to develop. Rooks could still sacrifice the exchange on f3, f6, c6, and c3. The ‘fork trick’ would still occur in the opening from time to time. Bishops could be finachettoed from time to time. Kings could castle to opposite sides and all out blitzes could occur.
You could learn something which might be useful later.
I think the castling rules are cute, admittedly in an arcane sort of way.
At first I thought this was incorrect, but now I see that it’s right. White’s bishop on h1 could attack black’s (possibly undefended) pawn on b7 with 1.g2-g3 or 1.g2-g4.
Nah, that’s too much like regular chess.
Doesn’t this overstate the number of starting positions by a factor of 2? The two knights (unlike the two bishops) are indistinguishable. If, for example, “one” knight is on b1 and “the other” knight is on g1, it’s the same as “the other” knight being on b1 and “one” being on g1.
This is called “4 choose 2” in the permutations game. It’s the number of ways 2 indistinguishable pieces can be placed on 4 squares. In general, “n choose m” (where m<=n) is equal to n!/(m!(n-m)!). For example, “4 choose 2” is 4!/(2!(4-2)!), which is 24/(2*2), which is 6, not 12.
So the number of legal starting positions in your variant would be 180, not 360.
In “normal” Fischer Random, the number of legal starting positions is 4 for the light-squared bishop, times 4 for the dark-squared bishop, times 6 for the queen, times (5*4)/2 (“5 choose 2”) for the knights. This comes out to 960. The pieces (rooks and king) to be placed on the remaining three squares are uniquely determined, because of the requirement that the king must be placed somewhere between the rooks.
That would be kind of okay if multiple pieces could occupy a square and some squares were left unoccupied (and you are counting Knight1 on b1 and Knight2 on c1 as a separate position from Knight2 on b1 and Knight1 on c1.
3 possible squares for the light-squared bishop x 2 possible squares for the dark-squared bishop x 3 possible squares for the queen in each of those combinations. Looks like 18 possibilities.
Hmm, you’re right. My answer, too, overstated (vastly) the number of possibilities.
3 possible squares for the light bishop times 2 for the dark bishop gives 6 ways to place the two bishops. Each of these 6 allows 3 possible remaining squares for the queen. That’s 18. Once those are placed, there is only 1 way to place the knights.
been playing 960 for quite awhile now. seems that what most fail to realise is that the basic tenets of “regular” chess seem to work with 960, too. go figure. centre control, piece development. it ain’t really that difficult.
I saw an idea a few years ago that made a lot more sense to me: pick one alternate position, say 338 just completely at random, and have all 960 games use that position for, say, two or three years. With advance knowledge of the opening position, people could prepare and opening theory would develop because it wouldn’t be a waste of time. Of course there are likely some positions that are flat out unfair, but if one of those is picked before it is released a 10,000 game engine match could be played in a week and if either color scores more than, say, 60%, reject that initial position.