In this paper, a novel concept of time connectivity is devised to evaluate the quality of network-level synchronization. The time connectivity is defined analogously to communication connectivity. To facilitate time connectivity analysis for wireless networks, a probabilistic synchronization model is developed for pre-synchronization and post-synchronization scenarios to evaluate synchronization performance. By utilizing recent results on stochastic geometry and percolation theory, time connectivity is transformed into the two-dimensional space for network-level analysis. Notably, the network time connectivity is investigated in two network schemes: partial-coverage and out-of-coverage. Extensive simulations are conducted to evaluate the time connectivity framework. To the best of our knowledge, this is the first paper that incorporates stochastic geometry and percolation theory to study the performance of network-level synchronization.