A proper definition of the 'optimal' number of implants to support a full arch prosthesis should go beyond solely a listing of the number of implants used in a treatment plan; it should be based upon a biomechanical analysis that takes into account several factors: the locations of the implants in the jaw; the quality and quantity of bone into which they are placed; the loads (forces and moments) that develop on the implants; the magnitudes of stress and strain that develop in the interfacial bone as well as in the implants and prosthesis; and the relationship of the stresses and strains to limits for the materials involved. Overall, determining an 'optimal' number of implants to use in a patient is a biomechanical design problem. This paper discusses some of the approaches that are already available to aid biomechanically focused clinical treatment planning. A number of examples are presented to illustrate how relatively simple biomechanical analyses - e.g. the Skalak model - as well as more complex analyses (e.g. finite element modelling) can be used to assess the pros and cons of various arrangements of implants to support fullarch prostheses. Some of the examples considered include the use of 4 rather than 6 implants to span the same arc-length in a jaw, and the pros and cons of using tilted implants as in the 'all-on-4' approach. In evaluating the accuracy of the various biomechanical analyses, it is clear that our current prediction methods are not always perfectly accurate in vivo, although they can provide a reasonably approximate analysis of a treatment plan in many situations. In the current era of cone beam computerised tomography (CT) scans of patients in the dental office, there is significant promise for finite element analyses (FEA) based on anatomically-accurate input data. However, at the same time it has to be recognised that effective use of FEA software requires a reasonable engineering background, especially insofar as interpretations of the clinical significance of stresses and strains in bone and prosthetic materials.