Static translation (also called port forwarding), assigns a fixed address to a specific internal network resource (usually a server).
Static NAT is required to make internal hosts available for connection from external hosts.
It merely replaces port information on a one-to-one basis. This affords no protection to statistically translated hosts: hacking attacks will be just as efficiently translated as any other valid connection attempt.
NOTE FROM CLEMENT: Hiding Nat or Overloaded Nat is when you have a group of users behind a unique public IP address. This will provide you with some security through obscurity where an attacker scanning your network would see the unique IP address on the outside of the gateway but could not tell if there is one user, ten users, or hundreds of users behind that IP.
NAT was NEVER built as a security mechanism. In the case of Static NAT used for some of your servers for example, your web server private IP is map to a valid external public IP on a one on one basis, your SMTP server private IP is mapped to a static public IP, and so on.
If an attacker scan the IP address range on the external side of the gateway he would discover every single one of your servers or any other hosts using static natting. Ports that are open, services that are listening, and all of this info could be gathered just as if the server was in fact using a public IP. It does not provide this security through obscurity mentioned above.
All of the other answer are incorrect.
Reference used for this question: STREBE, Matthew and PERKINS, Charles, Firewalls 24seven, Sybex 2000, Chapter 7: Network Address Translation.