Production enhancement of tight reservoirs requires in-depth analysis of mechanisms governing hydraulic fracturing, especially in naturally fractured reservoirs. This study is dedicated to the optimization of hydraulic fracturing and drilling by integrating both rock fracture mechanics and petrophysics study. Fracture twist near the borehole adversely impair production rate or induce premature screenout, and is analyzed in terms of mixed-mode fracture propagation. The best fracture propagation criterion is selected by comparing experimental and theoretical fracture initiation angles, suggestions regarding the alleviation of fracture twist are summarized by sensitivity analysis. Accurate estimation of fracture gradient is critical for both drilling and hydraulic fracturing. Fracture gradient by traditional methods is greatly overestimated due to the ignorance of preexisting fractures, nonlinear near wellbore stress concentration acting on fracture surfaces, and nonlinear fluid pressure distribution inside fractures. A weight function method is firstly introduced to petroleum engineering for the calculation of stress intensity factor where there are preexisting fractures intersecting borehole. Weight function parameters of a pair of symmetrical fractures emanating from borehole are derived and verified against existing models. The weight function parameters are applied to predict breakdown pressure of preexisting fractures. The simulation results are compared against both measured breakdown pressure in both fields and laboratory and results of classical fracture models. The proven weight function method shows a great potential in improving the accuracy of breakdown pressure prediction. Screening fracture candidates plays a central role in hydraulic fracturing design. Identification of fracture barrier helps prevent freshwater aquifer from contamination and undesirable water breakthrough due to unintentional cross of fracture barriers. New definition of brittleness is developed and benchmarked for unconventional shale. Correlations of brittleness with neutron porosity and compressional slowness are built for predicting brittleness in the fields lacking specific logging data, and saving cost of logging service. Fracability index model is firstly developed for screening fracturing candidates and fracture barrier identification by integrating brittleness and fracture energy (or fracture toughness, Young’s modulus, tensile strength). Its interpretation result is proven by logging interpretation. It is found that it is not always right that formations with high brittleness are good fracturing candidates. The fracability index model refines the formation evaluation and narrows the thickness of target interval. This research establishes a comprehensive understanding regarding fracture twist, mixed-mode fracture propagation, fracture gradient prediction, fracturing candidate selection by integrating theoretical modeling, experimental efforts, and logging interpretation. The methodologies will not only help engineers improve hydraulic fracturing design, but also provide evidence and theoretical support for explaining previously failed cases.