A random player is similar to legal
player.
That it maintains a state and selects an
action for each state
based solely on its legality in that state
without consideration of the consequences.
A random player differs from a legal
player
is not simply take the first legal move
it finds but rather selects randomly from
among
all of the legal actions available in that
state.
Usually choosing a different move on
different occasions.
The implementation of a random player's
almost
identical to the implementation of a legal
player.
In fact it's hard to see the difference
here.
The only difference is in the play
handler.
In the legal player the play handler
simply returns the first legal move.
In a random player the play handler first
computes all legal moves
using find legals and then selects one at
random from this list.
Random players are no smarter than legal
players,
however they often appear more interesting
because they're unpredictable.
Also sometimes they avoid tracks, traps
that befall consistent players like legal,
which
sometimes can ma, maneuver themselves into
corners
from which they are unable to escape.
Random players are also used as standards
to show that
general game players as specific methods
perform better than chance.
now one downside on a random player is it
consumes slightly more compute time than
a legal player, since it must compute
all legal moves rather than computing just
one.
Now, for most games, this is not a problem
but for games
with a large number of possible actions
the difference can be noticeable.