We study networked control of nonlinear systems where system states and tentative plant input sequences are transmitted over unreliable communication channels. The sequences are calculated recursively by using a pre-designed nominally stabilizing state-feedback control mapping to plant state predictions. The controller does not require receipt acknowledgments or knowledge of delay or dropout distributions. For the i.i.d. case, in which case the numbers of consecutive dropouts are geometrically distributed, we show how the resulting closed loop system can be modeled as a Markov nonlinear jump system and establish sufficient conditions for stochastic stability.