The Elastic Net Algorithm (ENA) for solving the Traveling Salesman Problem is analyzed applying statistical mechanics. Using some general properties of the free energy function of stochastic Hopfield Neural Networks, we argue why Simic's derivation of the ENA from a Hopfield network is incorrect. However, like the Hopfield-Lagrange method, the ENA may be considered a specific dynamic penalty method, where, in this case, the weights of the various penalty terms decrease during execution of the algorithm. This view on the ENA corresponds to the view resulting from the theory on `deformable templates', where the term stochastic penalty method seems to be most appropriate. Next, the ENA is analyzed both on the level of the energy function as well as on the level of the motion equations. It will be proven and shown experimentally, why a non-feasible solution is sometimes found. It can be caused either by a too rapid lowering of the temperature parameter (which is avoidable), or by a peculiar property of the algorithm,namely, that of adhering to equidistance of the elastic net points. Thereupon, an alternative, Non-equidistant Elastic Net Algorithm (NENA) is presented and analyzed. It has a correct distance measure and it is hoped to guarantee feasibility in a more natural way. For small problem instances, this conjecture is confirmed experimentally. However, trying larger problem instances, the pictures changes: our experimental results show that the elastic net points appear to become `lumpy' which may cause non-feasibility again. Moreover, in cases both algorithms yield a feasible solution, the quality of the solution found by the NENA is often slightly worse than the one found by the original ENA. This motivated us to try an Hybrid Elastic Net Algorithm (HENA), which starts using the ENA and, after having found a good approximate solution, switches to the NENA in order to guarantee feasibility too. In practice, the ENA and HENA perform more or less the same. Up till now, we did not find parameters of the HENA, which invariably guarantee the desired feasibility of solutions.