Random data is not compressible.
Adding redundant data to random data will make it more compressible - less random.