statistical entropy
/ sta-TIS-ti-cal EN-truh-pee /
Suppose two boxes look identical from the outside, but you are told one can be arranged inside in a thousand secret ways and the other in only ten. The first box is, in a precise sense, richer in hidden possibilities. Statistical entropy is the number that captures exactly that richness: it measures how many microscopic arrangements a system could secretly be in while still showing the same outward face.
More precisely, statistical entropy connects the thermodynamic entropy S to a count of microstates. In Boltzmann's form, S equals the Boltzmann constant k times the natural logarithm of W, the number of microstates that share the same energy. In the more general Gibbs form, it is a weighted sum over the probabilities of all accessible microstates. Either way, more accessible arrangements mean higher entropy.
Why it matters: statistical entropy is the deep meaning behind the thermodynamic entropy you measure with heat and temperature — it explains why entropy tends to increase (systems drift toward macrostates with overwhelmingly more microstates) and gives entropy a true zero. The honest caveat is that the count W must be done correctly, including degeneracies and the right way of treating identical particles, or the bridge gives wrong numbers.
Take four distinguishable molecules and two equal-sized halves of a box. There is only one way to put all four on the left, but six ways to split them two-and-two. Boltzmann's S = k ln W then assigns the even split a higher entropy — which is why gases spread out to fill their container.
The even split owns more microstates, so it carries higher statistical entropy.
Statistical entropy and thermodynamic entropy are the same quantity reached by two roads: one counts microstates, the other tracks heat divided by temperature. Their agreement is one of the great confirmations of the atomic picture.