Suggestion for a random position variant

One random position variant I've seen, on Roblox, is one in which the kings always start on the e file, each player starts with 7 non royal standard chess pieces on the back rank, and 8 pawns in front of the king, however which type of piece starts on each back-rank square is random, and the starting position is always mirror symmetric.

So one game the starting position might be this

and the next game the starting position might be this

I calculated that there are about 17 times more possible starting positions in this variant than there are in chess 960.

this would come with lots of unfair positions with one player posed to win, it's basically WoF but rigged.

But hold up, you got this from ROBLOX of all places? Dang.

@rohithbala, yes there's a room on Roblox in which it's possible to play chess and chess variants.  It also has pawn battle, in which pawn promotion is checkmate, and there's a sort of standard chess, however in all the variants and standard chess on Roblox pawns can only promote to queens.

how would this result in 17 times more possibilities than Chess 960? Unless of course you break the rules by placing both bishops on same colored squares and putting the rooks on the same side of the king

@TheCheesePhoenix When playing the variant called random on Roblox I have seen starting positions involving two bishops on the same colored squares, two rooks on the same side of the king, and even starting positions with a different number of rooks, bishops, knights, and queens than in standard chess, so the possibilities are even greater than you would get by just allowing bishops to be on the same colored colored squares, or allowing both rooks to be on the same side of the king.  I've always observed the kings to be on the e file.  There's 4 types of piece, meaning that on each square on the back rank that isn't on the e file there are 4 possibilities, and with 7 back rank squares not on the e file the number of possible starting arrangements of pieces is 4^7, which is 16,384 possible starting positions, which is slightly more than 17 times larger than 960.

Well if I had to do the math, I'd say that yours is flawed, because once, say, the queen, is placed on a square, you cannot place it again since there is only one queen for each player. Likewise after you place the king, and the same applies for the pairs of pieces. Either way, placing two bishops on the same colored square is not a good idea at all, especially because it's a mirrored game, where players control opposite color complexes, which would most likely lead to a drawn endgame. Putting the rooks on the same side of the king voids castling rules, but I guess you could play "no castling" if you wanted. Still seems weird to me.

@TheCheesePhoenix from what I've observed, for the Roblox random variant the number of each type of piece is not always the same in each game, and in some games some types of pieces may be absent.  For instance in the last game I played there was no queen, one bishop, two knights, and 4 rooks.  In the game before that there were 3 queens, two rooks, one knight, and one bishop.  From what I observed what piece is on any given square seems to be random, although this is just an educated guess from observing what positions I would get when playing the random variant on Roblox as opposed to from reading an official set of written rules.

well this adds a completely new side to the game idea. completely randomizing the number of each piece and their positions as long as there are 8 pawns and the king is on the e-file. i love the idea but for some reason i think some positions could result in dead draws and others could be completely winning for one side