In the current system, numbers can be decomposed into the sum of coefficients times powers of ten - a power series. Understanding this (at some level) is crucial, and using a zero in, say, 20 is consistent with this. Using some other symbol in that position is not.

The same is true with other bases in number systems. We use hexadecimal a lot for representing computer bytes, and it works the exact same way, except the digits represent powers of sixteen. Ditto for octal, and older system used ind computers, where the digits are powers of 8. And binary, the heart of all digital systems works the same way, where digits are powers of 2.

Using some arbitrary symbol in place of the zero may result in a pleasing symmetry, but it obscures the fundamental nature of all common number systems - powers based sums.