The purpose of this study is to determine the portfolio determination of the Single Index Model, the Random Method, and to find out how the differences between the optimal portfolio returns of the two are. The sample in this study uses purposive sampling, namely the selection of samples with certain characteristics, so as to get 29 stocks in the KLCI index. The technique of data analysis and hypothesis testing uses the Wilcoxon Rank Sum-Test, to find out whether there is a difference in the average of two paired samples, the samples are the same sample. To test the hypothesis using the IBM SPSS Statistics 25 program tool. The results of the research from SPSS output show that the Asym Sig. (2-tailed) is 0.021 where the basis for making the decision is if the significance value (Sig < 0.005 means Ha is accepted). It can be concluded that there is a difference in portfolio returns between using the Single Index Model and the Random Method.