Many credit scoring systems depend on scorecards which order applicants by
credit risk. However the scorecards may also have other properties with cer
tain scores reflecting certain good:bad odds or differences in scores havin
g the same property throughout the score range. Other properties like posit
ivity of attribute points may be required for palatability or internal mark
eting reasons. The paper outlines the results of a small survey of what pro
perties scorecard builders require of their scorecards. It then discusses h
ow these properties can be obtained and describes a linear programming form
ulation which recalibrates scorecards so as to produce the best approximate
scorecard with the properties required.