Statistically speaking, you're going to still encounter deviations from equilibrium, even though the expected value is equilibrium. But these rare deviations from equilibrium - which are inevitable - might have the power to do work. So does the universe inevitably descend towards a maximum-entropy state? Or is it only probabilistically destined towards a maximum-entropy state - that is - it will be in that state more than any other state.
After all, someone has even hypothesized a Poincare recurrence time, as described below:
Scale of an estimated Poincaré recurrence time for the quantum state of a hypothetical box containing an isolated black hole of stellar mass. This time assumes a statistical model subject to Poincaré recurrence. A much simplified way of thinking about this time is that in a model in which history repeats itself arbitrarily many times due to properties of statistical mechanics, this is the time scale when it will first be somewhat similar (for a reasonable choice of "similar") to its current state again.