Yes, there is a difference between “unable to replicate because we couldn’t even attempt to replicate” (code and/or data are missing) and “unable to replicate because we tried and the results did not match”. Either both or only the second case could be called “failure to replicate”, depends on your preferred definition.
Still, while the second case is clearly “bad science”—it’s either mistakes or fraud—the first case is “not science” because science doesn’t work by trusting the word of the researcher. A well-known example of the first case is cold fusion.
Yes, there is a difference between “unable to replicate because we couldn’t even attempt to replicate” (code and/or data are missing) and “unable to replicate because we tried and the results did not match”. Either both or only the second case could be called “failure to replicate”, depends on your preferred definition.
Still, while the second case is clearly “bad science”—it’s either mistakes or fraud—the first case is “not science” because science doesn’t work by trusting the word of the researcher. A well-known example of the first case is cold fusion.