Since 2016, the Basel Committee on Banking Supervision has regulated banks to switch from a Value-at-Risk (VaR) to an Expected Shortfall (ES) approach to measuring the market risk and calculating the capital requirement. In the transition from VaR to ES, the major challenge faced by financial institutions is the lack of simple but powerful tools for evaluating ES forecasts (i.e., backtesting ES). The best available backtest for ES builds on the assumption of the correct specification of the conditional tail distribution, which itself is stronger than the assumption of the correct specification of ES. In this paper we propose some new backtests for ES without assuming the correct specification of the tail distribution. We establish the asymptotic properties of the tests, and investigate their finite sample performance through some Monte Carlo simulations. An empirical application to stock indices data highlights the merits of our tests.