This experiment was conducted using 48 female Markhoz goats, 3.5 years and 44± 1.2 kg body weight. Animals allocated in 2 groups including 24 female goats in each group. Group A and B received basal and flushing diets, respectively. Digestible energy, total digestible nutrients and crud protein in flushing diets almost was 1.5 times of basal diet. Estrus was synchronized using CIDR and insemination was done with fresh semen. Blood samples were collected 2, 48 and 92 hours after remove of CIDR and 20 days after artificial insemination. Blood samples were analyzed for concentrations of estrogen, progesterone, glucose, protein, blood urea nitrogen, phosphorus and beta-hydroxybutyrate. Applying backward modeling strategy in a logistic regression model indicated that among measured parameters which may affect the number of offspring in the follicular phase, only estrogen remained in the model. Therefore changes in estrogen concentration could be the most effective predictor for reproductive performance (P<0.01). Data analysis in the follicular phase revealed significant correlations between blood estrogen, glucose and protein concentrations and the number of offspring (P < 0.05). The highest correlation between number of offspring and blood parameters was found for glucose (r = +0.53) in the proestrus phase and for estradiol17-β and protein (r = + 0.6 and r = + 0.29 respectively) in the estrus phase. There was a significant correlation (r = + 0.58, P < 0.01) between blood progesterone level and the number of kids during pregnancy (20 days after artificial insemination).