Background Dermatopathology is considered the gold standard for melanoma diagnosis, but a subset of cases is difficult to diagnose by histopathology. Objective The goals of this study were to measure the accuracy of histopathologic features in difficult-to-diagnose melanocytic tumors and the interobserver agreement of those features. Methods This is a case-control study of histopathologic features of melanoma in 100 difficult-to-diagnose melanocytic neoplasms (40 melanomas and 60 nevi). Slides were blindly evaluated by 5 dermatopathologists. Frequencies, predictive values, and interobserver agreement were calculated. Univariate and multivariate logistic regression analyses were performed to identify the most influential features in arriving at a diagnosis of melanoma. Results Asymmetry, single-cell melanocytosis, solar elastosis, pagetoid melanocytosis, and broad surface diameter were most influential in arriving at a diagnosis of melanoma. Asymmetry and single-cell melanocytosis were most predictive of melanoma. Fleiss kappa was <0.6 for interobserver agreement in 9/10 histopathologic features of melanoma. Limitations This study is limited by the small sample size, selection bias, and binary classification of melanocytic lesions. Conclusion Our results indicate histopathologic features of melanoma in difficult-to-diagnose lesions vary in accuracy and reproducibility.