Fish stock assessment procedure is initially based on the assumption that the frequencies in length/weight-frequency samples used for analysis of the stock status follow approximately the normal distribution. Many of the statistical procedures are based on specific distributional assumptions. The assumption of normality is very common in most classical statistical tests. In case that analysis of data implies techniques that make normality or some other distributional assumptions it is essential that this assumption is confirmed. If distributional assumption is proved, more powerful parametric techniques can be applied and if it is not justified an application of non-parametric or robust techniques may be required. The present article aims to present MATLAB-based algorithm for commercial fisheries length-frequency samples distribution analysis of samples of Sprat and Anchovy caught in the Bulgarian waters in the Black sea. The statistical analysis uses engineering approaches in statistical data processing and the method used for analysis of sample frequencies distribution is chi-square normality or goodness of fit test. For the provision of this analysis, specific program is developed in MATLAB programming environment to support and confirm the assumption that length-frequency samples follow the normal distribution.