Derek Nalls wrote on Thu, Apr 17, 2008 07:54 PM UTC:
CRC
practical attack values
http://www.symmetryperfect.com/shots/values-capa.pdf
Although Aberg's method of estimating the relative piece values for CRC pieces upon the 10 x 8 board was just an expedient extrapolation from established relative pieces values for FRC pieces upon the 8 x 8 board, his values are actually more accurate than yours.
One major flaw in your system is approximately equating the values of the archbishop and the chancellor. This is a radical contention which implies that the values of the rook and bishop are equal (since the archbishop equals a knight plus a bishop and the chancellor equals a knight plus a rook). This is inconsistent with your own system internally whereby the rook is (correctly) ascribed a higher value than the bishop.