Let's try and explain cancellation and wave interference and why an o-scope can't measure them at the source of the signal (amplifier):
I can try to explain cancellation this way, and yes this does correspond to an o-scope so bear with me. Let’s take a pool, the typical backyard variety. When a person dives in the pool, it makes a relatively small splash/wave. When they do a belly flop, the wave/splash is much larger. It is the same person, from the same height, but different waves. You can measure the force, velocity, etc of the person hitting the water and you can change them (much like you can add power with an amplifier to a subwoofer with the same specs). Problem is, the area the person presents to the surface changes the wave/splash characteristics, ala the more area you present to the water the bigger the waves, given everything else is kept the same.
Now if 2 people jump in the pool at different ends and everything about them (all the variables) is/are the same, weight, area hitting the water, speed, etc, the waves will interfere by getting larger but a small bit of cancellation will still exist or else they would be double the size of the original wave. You can't really predict where they will cancel without some serious math, theory, guesstimation, but the waves do interfere and cause cancellation. You can measure every aspect of the “person” and how they hit the water (the subwoofer and the amplifier power -- much like you would on a scope and by measuring T/S params) but the best way to reduce the cancellation that occurs is by not using things with different variables or characteristics. There are so many factors that influence cancellation that reducing the major variables such as amplifier power differences and different woofers (sizes, brands, etc) will yield the cleanest response. When you start messing with the big variables, you start experiencing major problems that can't be easily measured unless you have really expensive equipment. And yes, they will exhibit cancellation, even if you can’t hear it.
Last, remember you aim the box in a car to maximize the additive properties of sound waves, much like water waves. You are trying to achieve maximum addition and minimal cancellation. So if you can measure cancellation on an oscilloscope, rotating the box 90º won’t affect the sound at all because at the amp, it still ‘looks’ the same.
*If this is totally confusing, that is because I am not the best explainer*