What is the purpose of APawn

“The Pawn class is the base class of all Actors that can be controlled by players or AI. A Pawn is the physical representation of a player or AI entity within the world”

Consider a multiplayer RTS game with one client controlling many units and potentially multiple cameras. I.e. thinking of diablo, civilization, even chess, monopoly, etc…

From reading the forums, it sounds like the best option is to have a custom PlayerController that handles all input binding then calls appropriate action on actors in the game.
What is the point of “possessing” a pawn from a controller when you can just spawn actors from the controller, manage member pointers to them,
and move them around as needed? Why even create a restriction on being able to only possess one pawn at a time from a controller?
Why would there need to be a single unit to physically represent a player (any game where you control multiple units, like chess) and why wouldn’t i just use an actor to do so? Also true for any game with a menu/lobby/etc.
Lastly, I don’t see any APawn member functions that useful in regards to physical representation from the api docs https://docs.unrealengine.com/en-US/…awn/index.html

In FPS games the Pawn class makes perfect sense being able to easily get the Pawn that you are controlling and know its PlayerState and Controller. The Pawn class is not very good in a game with lots of “Pawns” then I would probably just make Actors with one or many entities (components) representing units.