The main advantage of the double random phase encryption technique is its physical implementation however to allow us to analyse its behaviour we perform the encryption/decryption numerically. A typically strong encryption scheme will have an extremely large key-space, which will make the probable success of any brute force attack on that algorithm miniscule. Traditionally, designers of optical image encryption systems only demonstrate how a small number of arbitrary keys cannot decrypt a chosen encrypted image in their system. We analyse this algorithm from a key-space perspective. The key-space of an encryption algorithm can be defined as the set of possible keys that can be used to encode data using that algorithm. For a range of problem instances we plot the distribution of decryption errors in the key-space indicating the lack of feasibility of a simple brute force attack.