To investigate the effects of the contract-relax-antagonist-contract (CRAC) form of proprioceptive neuromuscular facilitation (PNF) stretching, with and without a warm-up, on postural stability. Thirty volunteers (15 men and 15 women, age: 25.17 +/- 5.4 years, height: 173.76 +/- 8.2 cm, and weight: 72.03 +/- 14.87 kg) were randomly assigned to 1 of 3 conditions: warm-up and stretch (WS), stretching only (SO), and a control condition (CON). Contract-relax-antagonist-contract PNF of the hamstrings, plantar flexors, and hip flexors was performed during WS and SO. A 6-minute treadmill warm-up was applied before CRAC in the WS condition. Measures of anterior/posterior and medial/lateral (M/L) postural stability were taken before and after treatment conditions. A 2 x 3 analysis of variance was used to assess for differences between conditions. Significance was set at p < 0.05. There was a time x condition interaction (F = 3.962,58; p = 0.024, Power = 0.69) for M/L stability. There was a difference between WS and CON (p = 0.037, Power = 0.57) and SO and CON (p = 0.041, Power = 0.51) posttesting. This study suggests that CRAC PNF stretching with or without warm-up improves M/L stability. Contract-relax-antagonist-contract form of stretching is a useful protocol for improving M/L stability.