Objectives To investigate the prognostic value of pre-procedure neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) and construct a nomogram to predict disease-free survival (DFS) in patients receiving radiofrequency ablation (RFA) for hepatocellular carcinoma (HCC) within Milan criteria. Methods The data of 515 patients of HCC within Milan criteria receiving RFA were retrospectively collected. The patients were divided into two groups: the training group (n = 382) and the validation group (n = 133). Several preprocedural variables were analyzed in the two groups to determine the prognostic factors. Results The median DFS time of the training and validation group was 28.4 months and 24.5 months, respectively. Multivariate analyses showed that number of lesions, alpha-feto protein levels, NLR and PLR were independent risk factors of DFS. According to the time-dependent receiver operating characteristic curve (t-ROC), the optimal cutoff value of the NLR and PLR was 1.55 and 75.30, respectively, with sensitivity of 0.737 and 0.648 and specificity of 0.541 and 0.508, respectively. The area under curve (AUC) of the t-ROC curves for the NLR was 0.662 and PLR was 0.597. The DFS was significantly higher in the NLR <= 1.55 group compared to NLR > 1.55 group and the PLR <= 75.30 group compared to PLR > 75.30 group in both training and validation datasets. Nomogram was developed based on the prognostic factors indicated by the Cox regression to predict 1-, 2-, 3- and 5-year DFS probabilities. Conclusions The cutoff value of the NLR and PLR was 1.55 and 75.30. This new nomogram based on NLR and PLR may provide good and individualized prediction of recurrence for HCC patients within Milan criteria after RFA.