Background Elderly patients with early-stage triple-negative breast cancer are underrepresented in clinical trials, and recurrence prediction models tailored to this population are lacking. We aimed to develop a prediction tool using routinely available clinicopathological indicators. Methods We retrospectively enrolled 215 women aged ≥60 years with stage IA–IIB triple-negative invasive ductal carcinoma who underwent radical or modified radical mastectomy. Candidate predictors were screened by univariate Cox regression and entered into a multivariable Cox model with backward elimination. A nomogram was constructed and internally validated using bootstrap resampling. Discrimination was assessed by Harrell’s C-index and time-dependent area under the curve (AUC). Risk groups were defined using tertiles of the model-derived risk score. Decision curve analysis compared the net benefit with TNM stage alone. Results Four variables were retained: TNM stage, Ki-67 percentage, neutrophil-to-lymphocyte ratio (NLR), and postoperative chemotherapy. The bootstrap-corrected C-index was 0.793. Time-dependent AUCs were 0.934, 0.866, and 0.796 at 1, 3, and 5 years, respectively. The three risk groups showed clearly separated disease-free survival curves. The nomogram yielded a higher net benefit than TNM stage alone across threshold probabilities of 5% to 40%. Conclusions All model inputs were derived from standard preoperative evaluations and postoperative treatment records. For an older population with uneven access to molecular biomarkers, this nomogram can inform adjuvant treatment decisions using data already available. External validation in an independent cohort is needed before clinical adoption.