Future wireless communications will face the dual challenge of supporting large traffic volume while providing reliable service for various kinds of delay-sensitive traffic. In light of this challenge, this paper investigates the throughput performance of a wireless communication system under channel coding over finite “packets.” When we apply coding over multiple “packets,” the underlying wireless channel can be effectively modeled as a finite-state Markov process. By linking the characteristics of the underlying physical channel to the parameters of the Markov process, we are able to derive the channel dispersion of the corresponding system. The channel dispersion is then used to assess the coding performance of various communication strategies. It is interesting to find that when there is a constraint on the total blocklength, coding over large “packets” will give better performance than that of coding over small “packets”.