BACKGROUND AND PURPOSE: Strategies to address gait and balance deficits early poststroke are minimal. The postural and motor control requirements of Backward Walking Training (BWT) may provide benefits to improve balance and walking speed in this population. This pilot study (1) determined the feasibility of administering BWT during inpatient rehabilitation and (2) compared the effectiveness of BWT to Standing Balance Training (SBT) on walking speed, balance, and balance-related efficacy in acute stroke. METHODS: Eighteen individuals 1-week poststroke were randomized to eight, 30-minute sessions of BWT or SBT in addition to scheduled therapy. Five-Meter Walk Test, 3-Meter Backward Walk Test, Activities-Specific Balance Confidence Scale, Berg Balance Scale, Sensory Organization Test, and Function Independence Measure-Mobility were assessed pre- and postintervention and at 3 months poststroke. RESULTS: Forward gait speed change (BWT: 0.75 m/s; SBT: 0.41 m/s), assessed by the 5-Meter Walk Test, and backward gait speed change (BWT: 0.53 m/s; SBT: 0.23 m/s), assessed by the 3-Meter Backward Walk Test, preintervention to 1-month retention were greater for BWT than for SBT (P < 0.05). Group difference effect size from preintervention to 1-month retention was large for Activities-Specific Balance Confidence Scale, moderate for Berg Balance Scale and Function Independence Measure-Mobility, and small for Sensory Organization Test. DISCUSSION AND CONCLUSIONS: Individuals 1-week poststroke tolerated 30 min/d of additional therapy. At 1-month postintervention, BWT resulted in greater improvements in both forward and backward walking speed than SBT. Backward walking training is a feasible important addition to acute stroke rehabilitation. Future areas of inquiry should examine BWT as a preventative modality for future fall incidence.Video Abstract available for more insights from the authors (see Video, Supplemental Digital Content 1, http://links.lww.com/JNPT/A193).