LoRaWAN enables massive connectivity for Internet-of-Things applications. Many published works employ stochastic geometry to derive outage models of LoRaWAN over fading channels assuming fixed transmit power and distance-based spreading factor (SF) allocation. However, in practice, LoRaWAN employs the Adaptive Data Rate (ADR) mechanism, which dynamically adjusts SF and transmit power of nodes based on channel state. The community addressed the performance of ADR using simulations, but analytical models have not been introduced. In this letter, we seek to close this gap. We build over an analytical LoRaWAN model to consider the performance of steady-state ADR-enabled LoRaWAN. We derive outage expressions and an optimization procedure to maximize the number of users under reliability constraints. Results show that power allocation reduces interference and improves network capacity while reducing average power.