Decreased estradiol (E2) levels are associated with early miscarriage (EM), but the relationship between decreased times of E2 and EM has not been reported. We aimed to investigate the relationship between the average decreased times of E2 (ADTE) and EM. Women with a history of miscarriage were retrospectively recruited from the Reproductive Center of Lanzhou University Second Hospital (Lanzhou, China) between September 2019 and February 2022. Based on pregnancy outcome, they were divided into ongoing pregnancy group (n = 359) and EM group (n = 104). In addition, subgroup analyses were performed for the number of previous miscarriages and whether E2 levels decreased continuously. The exposure and outcome variables were ADTE and miscarriage before 12 weeks of gestation, respectively. Totally, 1171 patients were recruited and 463 patients were finally analyzed. ADTE was associated with EM (odds ratio [OR] = 1.346, 95% confidence interval [CI]1.154-1.571, P < 0.001). When ADTE ≥ 2.5, the EM risk increased 1.17-fold compared to patients with 0-1.249 times (OR = 2.170, 95% CI 1.144-4.117, P = 0.018). Moreover, a threshold effect existed in the ADTE and the risk of EM with a value of 4.9 times. When exceeding 4.9 times, the EM risk increased 4.713-fold for each increased unit (OR = 5.713, 95% CI 1.255-23.170, P = 0.024). Subgroup analysis showed that ADTE had a greater effect on the occurrence of EM in women with a history of 1-2 miscarriages than in women with 3 miscarriages. Decreased E2 was a risk factor for EM regardless of whether it dropped continuously or not. In conclusion, our study identifies a potential link between ADTE and early miscarriage risk in women with prior miscarriages, yet cautious interpretation is necessary due to inherent design limitations. Further research with prospective designs and large population samples is essential to validate ADTE's utility as a predictive indicator for early miscarriage in clinical settings.