We illustrate the possible connection that exists between the extremal prop
erties of entropy expressions and the renormalization group (RG) approach w
hen applied to systems with scaling symmetry. We consider three examples: (
1) Gaussian fixed-point criticality in a fluid or in the capillary-wave mod
el or an interface; (2) Levy-like random walks with self-similar cluster fo
rmation; and (3) long-ranged bond percolation. In all cases we find a decre
asing entropy function that becomes minimum under an appropriate constraint
at the fixed point. We use an equivalence between random-walk distribution
s and order-parameter pair correlations in a simple fluid or magnet to stud
y how the dimensional anomaly at criticality relates to walks with long-tai
led distributions.