Bill Tindall, E.Tn.
Amazingly this myth persists in spite of the fact that refuting data has been provided on numerous occasions. I tried it by mashing a joint in a large machinist vice capable of applying more pressure and one would ever be able to achieve with clamps. The joint was not "starved". Franklin confirmed my observation, no doubt, with better controlled experiments.
We know from experience that adding 5% water to Titebond thins it substantially which is a nice trick when spreading a bunch on veneer. The opposite is also true as a consequence of the physics of these latex suspensions. As soon as the glue touches the wood water diffuses into the wood and the viscosity of the glue at the surface builds quickly. The challenge is to squeeze the excess glue out of the joint, with no risk of squeezing out the thin layer required to get an optimum bond. The high viscosity of the glue at the wood surface prevents this glue from moving under any pressure clamps can muster.