Uniqueness in Chess Studies

Article type: Research Article

Authors: Haworth, G.McC.¹ | Bleicher, E. | van der Heijden, H.M.J.F.

Affiliations: [a] Reading, U.K. | [b] Berlin, Germany | [c] Deventer, The Netherlands

Note: [1] The University of Reading, Berkshire, UK, RG6 6AH. email: guy.haworth@bnc.oxon.org.

Abstract: Van der Heijden’s ENDGAME STUDY DATABASE IV, HhdbIV, is the definitive collection of 76,132 chess studies. In each one, White is to achieve the stipulated goal, win or draw: study solutions should be essentially unique with minor alternatives at most. In this second note on the mining of the database, we use the definitive Nalimov endgame tables to benchmark White’s moves in sub-7-man chess against this standard of uniqueness. Amongst goal-compatible mainline positions and goal-achieving moves, we identify the occurrence of absolutely unique moves and analyse the frequency and lengths of absolutely-unique-move sequences, AUMSs. We identify the occurrence of equi-optimal moves and suboptimal moves and refer to a defined method for classifying their significance.

DOI: 10.3233/ICG-2011-34104

Journal: ICGA Journal, vol. 34, no. 1, pp. 22-24, 2011