George Duke wrote on Wed, Aug 1, 2007 03:45 PM UTC:
Where are J K Lewis' piece values, Joe? It helps to familiarize with Ralph
Betza's 'Ideal & Practical Values I-VI'. How many years went into those 2001 articles, now a covenant? Later lists just build on Betza variables from 1990's and 1980's. However, Betza likes to compare pieces pairwise not whole CVs. In system to point-count entire game, available so far are elementary (1) Use unit Pawns 1.0, or 1.1 (2) Where practicable, Rook value & up, ignoring the other pieces, put piece X on empty board with 2 Kings and opposite-colour Pawns, and shuffle them around until finding Pawn-equivalence. These end-game positions are Joyce's reproducible 'many positions' versus '1 specific position'. Who should win to left? Trying 10, or 30, positions, we are interested in typical not
__ __ __ __ __ __ __ extreme cases with few pieces/pawns. Its experts
__ __ __ __K__ __ __ know who wins, but move one or two Pawns at a
__ __ __ __P__ __ __ time one or two spaces, and it goes the other
__ __P__P__ __P__P__ way. Pawn Equivalence, Rook = 5.0. There are
__ __ __ __ __ __ __ in mad-Queen (40+40 points)/ 64 squares
__ __ __ __ __ __ __ = 80/64 = 1.25. Should we always distribute
__r__ __ __k__ __ __ points so that there are approximately 1.25
__ __ __ __ __ __ __ the number of squares? So CVs may be like Bridge
hands: the total point-count for bidding in system of all four 13-card hands includes 40 high-card points plus short-suit counts of 5 or 10 or 12 more: 52 points/52 cards = 1.0. [Whole history of Bridge postdates Capablanca Chess]
