Increasing wind power integration and coal-fired unit retirements increases the strain on the power system’s spinning reserve and increases the pressure on peak regulation. With the ability to stock extra power generation and supply the peak load, the energy storage system (ESS) can alleviate the rising demand on the spinning reserve and play an increasingly important role in the power system. In this paper, a tri-level robust ESS planning model is proposed to accommodate uncertain wind power investment as well as coal-fired unit retirement. The upper-level of this model is to determine the planning scheme of ESSs, which iteratively takes the worst-case scenario of wind power investment and coal-fired unit retirement into consideration. The middle-level and lower-level of this model are to make the optimal daily economic dispatch under the worst-case realizations of uncertainties. We derive an equivalent reformulation of the proposed robust ESS planning model and solve it with a dual column-and-constraint generation algorithm. Case studies are conducted using the IEEE RTS-79 system. The results demonstrate the superiority of the proposed planning method in comparison with other methods. Furthermore, the effects of the capital cost of ESS, the expected proportion of wind power, and the uncertainty budget on the development of ESS are studied. Taking the uncertainties of unit retirement and wind power investment into consideration achieves a better trade-off between the ESS investment cost and the operational cost.