Error-correcting codes are widely used in modern coding theory, and their applications in networks and communication cannot be omitted. Together with the error detecting codes, they are the core of every possible transmission and communication. In coding theory, information theory and telecommunications, error-correcting codes are used to control errors in data which are transmitted over different communication channels. Convolutional block codes are one of the most popular error-correcting codes which are applied in many networks. On the other side, Random Codes Based on Quasigroups (RCBQ) are cryptcodes defined elsewhere. These codes provide a correction of a certain number of errors in the transmitted data and an information security in one algorithm. There are a few modifications of RCBQ, but here we will consider performances of Cut-Decoding algorithm. In this paper, we investigate and compare the bit error probability (BER) of these two codes for rate 1/4 and different values of bit-error probability in the binary symmetric channel. From the obtained experimental results, we conclude that for lower bit-error probability in the binary symmetric channel, the RCBQs are slightly better than convolutional codes. The advantage of RCBQs is that they have some cryptographic properties, but convolutional codes are faster than RCBQs.