I just took a backup of a SQL Server database. The MDF and LDF files together total around 29 GB, but the .bak file was only 23 GB, about 20% smaller.
My first guess when one version of a set of data is smaller than another version containing the same data would be data compression, but compression usually yields a much better compression ratio than 20%, especially for highly-ordered data (such as database tables.) Also, compressed data can't easily be compressed further, but I know that .bak files can be compressed.
So if the data isn't being compressed, and nothing's being discarded, (because the whole point of making a backup is to be able to restore it to an identical state afterwards,) then what's that 20% that's unaccounted for?