It's not the voltage that is the problem it's the amp draw from the servo. I was getting brownouts from the servo movements. The original circuit board design I did included a capacitor to try and reduce the issue. In the end powering the servo separately was the most stable solution.
While the nano can take in 6-12v unregulated (I wouldn't push past 12 with the clones) the voltage regulator on the board is not as efficient as a dedicated buck converter. Setting the output voltage to 5v and powering the arduino on the 5v pin bypasses the onboard regulator for better power to the board and to the servo.