Amphiphilic peptides hold great potential as drug delivery systems. A popular peptide design approach has been to place amino acids in the peptide sequence based on their known properties. On the other hand, the directed discovery approach aims to screen a sequence space for a desired property. However, screening amphiphilic peptides for desirable drug delivery properties is not possible without a quantity that is predictive of these properties. We studied the predictive power of critical aggregation concentration (CAC) values on the drug delivery performance of a series of amphiphilic peptides with different hydrophobic tails and close CAC values. The CAC values were predicted by our previously developed model and doxorubicin was used as a model hydrophobic drug. All peptides showed close drug loading, entrapment efficiency, and release profile. They also formed similar spherical particles by assembling in reverse β-sheet arrangements regardless of drug presence. Moreover, the assembled particles were able to accumulate doxorubicin inside ordinary as well as drug-resistant breast cancer cells and enhance its toxicity up to 39 and 17 folds, respectively. It can be concluded that similar drug delivery properties displayed by the peptides can be attributed to their similar hydrophilic-lipophilic balance as reflected in their close CAC values.