Agencies spend a lot of time and money screening and testing candidates who cannot possibly succeed in the overall selection process. To reduce these costs, agencies often divide their selection systems into phases, in which candidates must achieve a cut score in one phase to qualify for the next phase. Unfortunately, agencies often set these cut scores too low or too high, resulting in wasted time and money on candidates with little or no chance of being selected. Worse still, they eliminate hidden talent and risk litigation from candidates who did not get an opportunity to participate in the entire process. One solution to this problem is to conduct a computer simulation of the selection process and examine the results that would occur with various potential cut scores. This way, agencies can set cut scores that will help them reduce their costs and still maintain a completely effective and useful process that minimizes the risk of litigation.