The objective of the study was to compare conventional biomechanical outcomes, movement regularity, and local dynamic stability among habitual rearfoot strikers during acute non-rearfoot-strike adoption (HRFS-AN) and after 12 weeks of gait retraining (HRFS-GR), and habitual non-rearfoot strikers (HNRFS). Thirty-three male recreational runners enrolled. Analyses included 13 habitual rearfoot strikers assessed during acute NRFS adoption and after 12 weeks of gait retraining, together with 14 HNRFS. Participants ran on a treadmill at 9 km·h⁻¹ while ground reaction forces and lower-limb kinematics were recorded. Coefficient of variation (CV), sample entropy (SampEn), and largest Lyapunov exponent (LyE) were used to characterize discrete kinematic variability, temporal regularity, and local dynamic stability, respectively. Runners in all three states adopted an NRFS pattern. HRFS-AN showed a more anterior foot strike than HRFS-GR and HNRFS. Peak vertical ground reaction force and most spatiotemporal and discrete kinematic outcomes did not differ among states; selected joint angle and CV outcomes differed, while VILR and VALR were lower in HRFS-AN than in HRFS-GR. Compared with HRFS-AN, HRFS-GR and HNRFS exhibited lower SampEn for sagittal-plane knee and ankle motion and vertical COM motion, and lower LyE for sagittal-plane knee and vertical COM motion. Compared with HNRFS, HRFS-GR exhibited higher SampEn for sagittal-plane knee and transverse-plane hip motion and higher LyE for sagittal-plane knee and coronal-plane hip motion. Most conventional biomechanical outcomes were similar among states; however, HRFS-AN exhibited a more anterior foot strike than HRFS-GR and HNRFS, as well as lower VILR and VALR than HRFS-GR. Compared with acute adoption, the retrained state showed greater regularity and local dynamic stability in selected knee, ankle, and COM trajectories, although selected differences from HNRFS remained. These findings suggest that a gradual 12-week gait-retraining programme may be an appropriate approach for transitioning habitual rearfoot strikers to NRFS.

• HRFS-AN exhibited a more anterior foot strike and lower VILR and VALR than HRFS-GR, while most other spatiotemporal and discrete kinematic outcomes were similar among states.

• Compared with acute NRFS adoption, 12-week gait retraining resulted in greater regularity and local dynamic stability in selected knee, ankle, and COM trajectories.

• Although selected nonlinear differences from HNRFS remained, the findings suggest that gradual 12-week gait retraining may be an appropriate approach for transitioning habitual rearfoot strikers to NRFS.