PURPOSE: This study investigated the effects of a 6-week eccentrically-biased training intervention between Romanian deadlift (RDL) and Nordic hamstring exercises (NHE) on regional hamstring muscle architecture and morphology. METHODS: Participants were randomly allocated to an RDL or NHE intervention group and trained twice per week for 6 weeks followed by a 2-week detraining period. Biceps femoris long head (BFlh) architecture (fascicle length [FL] and pennation angle [PA]) and muscle thickness (MT) were measured using extended-field-of-view ultrasound. Anatomical cross-sectional areas (ACSA) of BFlh, semimembranosus, and semitendinosus were also assessed via ultrasound. Muscle architecture and morphology were measured proximally, in the middle, and distally along the thigh. Separate linear mixed effects models were run for each outcome. RESULTS: Thirty-two healthy participants (15 females, 17 males) aged 20.8 (0.9) years were enrolled and allocated to RDL (n = 16) or NHE (n = 16) groups. Following the intervention, BFlh FL increased (0.80 [0.6, 1.0] cm (9%), p <
0.001), while PA decreased uniformly across all regions (-1.2° [-1.5, -0.9] (10%), p <
0.001) and MT did not change (p = 0.29). Both BFlh FL and PA returned to baseline after detraining. No differences were observed between groups for changes in FL (p = 0.55), PA (p = 0.74), or MT (p = 0.48). Hamstrings ACSA increased after the intervention (0.78 cm2 [0.45, 1.11] (10%), p <
0.001) and remained elevated after detraining. There were no significant differences in ACSA between groups after the intervention (p = 0.60). CONCLUSIONS: Changes in FL, PA, and ACSA are consistent between an eccentrically-biased RDL and NHE intervention. These findings provide practitioners choice in exercise selection for eliciting adaptations relevant to hamstring injury prevention.