Basically it boils down to the fact that rates of reactions (kinetics) depend on both thermodynamic factors (energetics of transition state energies, mostly) and non-thermodynamic factors (molecular geometric factors, e.g.). Especially, because reactions can proceed through any number of pathways, many of them with multiple stages and steps, kinetics in practice is largely phenomenological. While it technically may be possible to predict kinetics from first principles considerations, the escalating complexity of all but the most basic chemical systems renders the study of kinetics predominantly experimental in nature (with much of the modelling work taking the form of semiempirical approaches). On the other hand, reaction thermodynamics depend mostly on only the initial and final states of the system, making thermodynamics as a discipline more directly susceptible to theoretical treatments. Of course certain aspects of kinetics are amenable to first-principles approaches (often applied ad hoc, after semiempirical relationships have been formulated) but it's difficult to construct a comprehensive theoretical framework that leads to easy prediction of how all complex reaction systems will evolve in time.