Equine pregnancy failure can occur throughout gestation with etiologies varying by region. In Europe, umbilical cord torsion (UCT) has been reported to account for up to 52% of abortions and stillbirths. Little is known about what causes excessive twisting of the umbilical cord (UC), and up till now, only the length of the UC has been identified as a risk factor for UCT. The aim of the present study was to determine the association of potential maternal and fetal risk factors with UCT and umbilical cord length (UCL). A retrospective cohort study was performed on aborted or prematurely delivered stillborn fetuses submitted for necropsy, and a follow-up questionnaire survey was sent to mare owners to obtain additional anamnestic data. In total 72 cases were included in the study. Multivariable regression analyses were performed to evaluate the risk factors (mare size, period of gestation, gestational length, parity, previous abortion, UCL, maternal age at abortion and fetal sex) association with UCT and UCL, and Wilcoxon rank sum test tested pairwise differences in UCL for the risk factors mare size, period of gestation and cause of abortion. Abortion caused by UCT (UCTA) was diagnosed in 37 cases, and these had shorter gestational length (8.1 ± 0.3 months) compared to fetuses lost to other causes (8.9 ± 0.2 months) (p = 0.002). UCTAs accounted for the highest prevalence of abortions in the mid gestational period (113–227 days) (14/21, 67%) compared to the late period (228–314 days) (22/50, 44%) (p = 0.05). The cause of abortion was associated with period of gestation (p = 0.009) and UCL (p = 0.002), and fetuses aborted as a consequence of UCT had longer UCs (98.84 ± 5.14 cm) compared to abortions due to other causes (67.2 ± 4.9 cm). Mares in the mid period of gestation had 7.9 times the odds of UCTA compared to the late period (CI: 1.93, 32.2), and UCL was associated with increased odds of UCTA (OR: 1.05, CI: 1.02, 1.07). UCL was associated with mare size (p = 0.012), period of gestation (p = 0.002) and cause of abortion (p = 0.001). In conclusion, UCTA was significantly associated with UCL and gestational length. The prevalence of UCTA was not influenced by mare size, but pony fetuses from UCTAs had significantly shorter UCs compared to horse fetuses. UCs longer than 54 cm may predispose ponies for UCTA. No other maternal or fetal risk factors were found to be associated with the cause of abortion.