First, it only applies to three-phase systems. Secondly, it has to do with the timing of the rising and falling of currents (or voltages) in each of the three wires.
Consider three conductors that are run across a room, from a power source to a motor. They are run in open air, supported every few feet or so, and you can see all the conductors. Label one of them ?Phase A,? one of them ?Phase B,? and one of them ?Phase C.? Put a clamp-on ammeter on each conductor, and run the signal cables to a computer display. Suppose you see that, at one moment in time, the current in Phase A is about to reach its peak, and the currents in the other two phases are much lower. A short time later, you observe that the current in Phase B reaches its peak, and a short time after that the current in Phase C reaches its peak. Then the pattern repeats itself. The pattern is A, B, C, A, B, C, A, B, C. . . .
It is just as possible that, after Phase A reached its peak, the next one to reach a peak might have been Phase C, with Phase B coming next. The pattern would then be A, C, B, A, C, B, A, C, B . . . .
Those are the only two possible phase rotation patterns. The most significant difference between the two is that one will spin a motor clockwise, and the other will spin the same motor counter-clockwise.