# Material value of the Berolina pawn

The Berolina pawn is a fairy piece that moves like a normal pawn, but can move two (and one in all other cases) steps diagonally and capture one forward. How much better (or worse) is this piece compared to a normal pawn?

• Put a pawn on an empty board. Count the number of moves it can make. Then put the Berychucky pawn on the board, do the same thing. That'll give you a fair idea of their relative values. Sep 20, 2020 at 5:11
• If you're describing it correctly, it's probably worth a slight bit more than a pawn, because it has two moves and one capture, while the normal pawn has one move and two captures (which doesn't happen as often). Sep 20, 2020 at 5:14
• I wonder if anybody has tried a variant where one side has normal pawns and the other side has Berolina pawns? That might be fun. Who would have the advantage>?
– bof
Sep 20, 2020 at 10:44
• "but can move two (and one in all other cases) steps diagonally" Do you mean "two in its first move and one in all other cases"? Sep 20, 2020 at 19:22
• @Acccumulation Yes. Just as you would expect. Same as a normal pawn, but with "straight ahead" and "diagonally forward" swapped. A standard fairy chessman. chessvariants.com/piececlopedia.dir/berolina.html
– bof
Sep 21, 2020 at 1:02

According to Gnu, which has a page dedicated to Berolina chess, the value of the Berolina pawn is the same as in standard chess, = 1. This is unsatisfactory because they give no explanation of how this evaluation was made.

The evaluations for standard chess existed before computer chess programs and were based on hundreds of years of experience playing standard chess. That doesn't exist for Berolina chess which was only invented in 1926 and is little played.

In the absence of similar amounts of over-the-board games of Berolina I think the best approach to getting an accurate evaluation would be an AlphaZero-like simulation of millions of self-learning games which could provide the empirical data upon which a good evaluation could be made.