The growing tendency of wind energy technology usage has induced researchers to seek more studies on the improvement of wind energy harvesting systems with highly reliable power injected into the electric grid. The amount of extracted wind energy by a wind turbine depends on the accuracy of the maximum power point tracking (MPPT) algorithm associated with the control scheme of the wind energy conversion system (WECS). This paper aims to present a novel development of MPPT algorithm based on parabolic prediction techniques (PPT) for variable speed wind turbines (VSWT) so as to achieve maximum wind power extraction and efficiency enhancement. The developed PPT-MPPT algorithm uses parabolic convex functions to estimate the parabola wind turbine output curve and locate the maximum power point on that curve. Simultaneously the algorithm establishes a systematic procedure for adjusting the concavity and the optimal part of the estimated parabola curve to make sure that the proposed technique is in continuous convergence. The feasibility of the developed PPT-MPPT algorithm has been validated under rapid and random changing wind speed profiles via the MATLAB/Simulink environment. The performance of the proposed PPT-MMPT algorithm is evaluated by comparing the simulation finding to those obtained using the benchmark perturb and observe(P&O) algorithms. The simulation outcomes affirm that the proposed PPT-MPPT algorithm outperformed the other algorithms in terms of tracking efficiency and extracted power which can be regarded as an efficacious manner to improve the VSWT system. © 2013 IEEE.