Esteve-Lanao, J, Del Rosso, S, Larumbe-Zabala, E, Cardona, C, Alcocer-Gamboa, A, and Boullosa, DA. Predicting marathon performance time throughout the training preparation in recreational runners. J Strength Cond Res XX(X): 000–000, 2019—The objective of this study was to predict marathon performance at different time points along the season using different speeds derived from ventilatory thresholds and running economy (RE). Sixteen recreational runners (8 women and 8 men) completed a 16-week marathon training macrocycle. Aerobic threshold (AeT), anaerobic threshold (AnT), and maximal oxygen uptake were assessed at the beginning of the season, whereas speeds eliciting training zones at AeT and AnT, and RE were evaluated at 5-time points during the season (M1–M5). Analyses of variance and hierarchical regression analyses were conducted. Training improved AeT and AnT speeds at M2 vs. M1 (p = 0.001) and remained significantly higher at M3, M4, and M5 (p = 0.001). There was a significant effect of time (p = 0.003) for RE, being higher at M4 and M5 compared with M1 and M3. Significant correlations were found between marathon performance and speeds at AeT and AnT at every time point (r = 0.81–0.94; p < 0.05). Speed at AnT represented the main influence (65.9 and 71.41%) in the final time prediction at M1 and M2, whereas speed at AeT took its place toward the end of the macrocycle (76.0, 80.4, and 85.0% for M3, M4, and M5, respectively). In conclusion, assessment of speeds at AeT and AnT permits for reasonable performance prediction during the training preparation, therefore avoiding maximal testing while monitoring 2 fundamental training speeds. Future research should verify if these findings are applicable to runners of different levels and other periodization models.