The objective was to evaluate the reproductive performance and productivity of Pelibuey ewes in two flocks in Southeastern Mexico. Animals grazed on irrigated star grass (Cynodon plectostachyus). In flock 1, lambs grazed with their mothers from 7 to 70 days of age (weaning age), whereas in flock 2, they were confined and offered chopped grass, a commercial concentrated and managed under restricted suckling until weaning at 60 days of age.. The statistical model included the fixed effects of flock, year and season of lambing, parity number, litter size at weaning, first order interactions and the random effects of ewe and the error term. The means ± SD for age at first lambing, prolificacy, lambing interval (LI), weaning weight, and productivity were 477 ± 85.5 days, 1.41 ± 0.51 lambs, 255.9 ± 51.8 days, 16.5 ± 4 kg, and 18.1 ± 4.42 kg of lamb weaned, respectively. Repeatability estimates for the latter four traits were 0.09, 0.07, 0.11, and 0.13, respectively. All main effects in the models were significant (P < 0.001), except for year on LI. Prolificacy and productivity increased with parity number. The shortest LI was in the rainy season. Ewes with two or three lambs produced 8 and 14 kg of lamb more at weaning than single lamb ewes. In conclusion, reproductive performance and productivity could be improved, and the low repeatability estimates suggest that management should be used in a short term rather than genetic approaches.