Genome mapping by anchoring random clones has recently been the subject of intensive theoretical study. In this paper, differences between published predictions of properties of anchored groups of clones ("contigs") are analyzed and simplifications of the mathematical formulae describing these properties are presented. The theoretical predictions are compared with the experimental results from the physical mapping of the genome of Schizosaccharomyces pombe. Information about the number of genome sections with no anchored clone on them ("oceans") and the number of undetected overlaps between the contigs at a given stage of the experiment is required for the decision to change from the random strategy to that of a directed closure of gaps. We demonstrate that the expected number of oceans can be approximated by the number of groups of clones anchored by a single probe ("singletons"), as can the expected number of undetected overlaps between contigs by the number of contigs containing more than one anchor.