Toward an empirical approach to evidentiary ruling.

This paper responds to criticisms/misconstruals of our measure of the maximum probative value of evidence (D. Davis & W. C. Follette, 2002), and our conclusions regarding the potentially prejudicial role of "intuitive profiling" evidence, including motive. We argue that R. D. Friedman and R. C. Park's (2003) criticisms and example cases are largely based on inappropriate violation of the presumption of innocence. Further, we address the merits of our absolute difference measure of probative value versus those of the Bayesian likelihood ratio championed by D. H. Kaye and J. J. Koehler (2003). We recommend methods for presentation of measures of evidence utility that convey complexities of interdependence between new and existing evidence. Finally, we propose a "probable cause" standard for admission of potentially prejudicial evidence, dictating that admissibility of such evidence should be contingent upon other substantial evidence of guilt.