Another way to think about it:
Let's say my Triac gets turned on at the peak of the AC cycle. We are then switching the load circuit from 0V to 170V in the turn-on time of the Triac. Although it's a resistive load, there is some inductance and capacitance in the line. The "step" function of the Triac turning on looks like a square wave edge, which will look like a blast of energy at a wide frequency range, particularly across the AM band.
By starting at the zero-cross, there is no meaningful sudden voltage switching - it just happens to have a higher load on that next cycle.
Turning off at the zero-cross is also important.