The state-of-the-art polymer dielectrics have been limited to nonpolar polymers with relatively low energy density but ultralow dielectric losses for the past decades. With the fast development of power electronics in pulsed power and power conditioning applications, there is a need for next-generation dielectric capacitors in areas of high energy density/low loss and/or high temperature/low loss polymer dielectrics. Given limitations in further enhancing atomic and electronic polarizations for polymers, this Perspective focuses on a fundamental question: Can orientational polarization in polar polymers be utilized for high energy density and low loss dielectrics? Existing experimental and theoretical results have suggested the following perspectives. For amorphous polar polymers, high energy density and low loss can be achieved below their glass transition temperatures. For liquid crystalline side-chain polymers, dipole mobility is so high that they saturate at relatively low electric fields, and only limited electrical energy can be further stored after dipole saturation. Crystalline polar polymers are promising and can be divided into three categories: normal ferroelectric, paraelectric, and novel ferroelectric. For normal ferroelectric crystalline polymers, switching of a high spontaneous polarization results in a large hysteresis. To reduce the hysteresis, ultrafine crystallites or ferroelectric domains are desired to reduce the spontaneous polarization. For paraelectric crystalline polymers, dipoles have the potential to align in an external electric field. However, a high degree of dipole reversibility is required for the high energy density and low loss application. Novel ferroelectric behaviors include relaxor ferroelectric and antiferroelectric-like behaviors are highly desired because of their high degree of dipole reversibility. To achieve the relaxor ferroelectric behavior, structural defects such as bulky comonomers need to be introduced into the crystalline lattice to expand the lateral unit cell dimensions and speed up the mobility and reversibility of crystalline dipoles. So far, true antiferroelectric crystalline polymers have not yet been discovered. Nevertheless, the antiferroelectric-like behavior has been realized by reducing the compensation polarization via nanoconfinement. In the future, more research is needed to develop new paraelectric and novel ferroelectric polymers for high energy density and low loss dielectrics.