Exteroceptive sensors can be used to estimate the robot pose in order to suppress inaccuracies coming from the accumulation of modelling errors when using the classical control approach. In some cases, it is impossible to directly observe the end-effector. Thus we can replace it efficiently by the observation of the legs directions. However, with such an approach, unusual results were recorded, namely: (i) the possibility of controlling the robot by observing a limited number of legs, and (ii) in some cases, the robot does not converge to the desired end-effector pose, even if the observed leg did. These results can be explained through the use of the hidden robot concept, which is a tangible visualisation of the mapping between the observed leg direction space and Cartesian space. In the present paper, it is explained (1) why the tools used in mechanical design can be efficently applied in control analysis via the use of the hidden robot concept which is, in our opinion, a way to unify the analysis of the mechanical and control performances and (2) why we believe that the hidden robot concept must be used to modify the robot design methodologies in order to include control-based performance indices.