You're right: paracetamol doesn't have a chiral centre, whereas carvone does, so synthesis of paracetamol will produce 100% of just paracetamol
However, 50% of the carvone synthesis mixture will be (-)-carvone only if carvone is produced as a racemate, i.e. equimolar amounts of (-)-carvone and (+)-carvone. Even if it isn't produced as a racemate, you will still have a mixture of enantiomers, which will require separation to obtain one specific enantiomer and this is a difficult process
If you're performing the synthesis with one specific enantiomer in mind, then extra pathways/reagents/conditions may be required
In essence, synthesis of one specific enantiomer of an optical compound requires a lot more effort than the synthesis of a non-optical compound