That makes very little sense, as mathematics is completely independent from physics, and physics can exist without (the field of) mathematics.
Here's an introduction to various topics in physics
as working physicists understand them, pitched at the advanced layman's level (lay people who have some science and math background). They were famously delivered at Stanford University Continuing Education to an audience heavily laden with Silicon Valley techies by Leonard Susskind (the string theory guy).
http://theoreticalminimum.com/courses/classical-mechanics/2011/fall
So if "mathematics is completely independent from physics", then what's up with phase space, Lagrangians, least-action principles, Euler-Lagrange equations, symmetry and conservation laws, the Hamiltonian, Hamilton's equations, Liouville's theorem, poisson brackets, gauge fields and gauge invariance?
http://theoreticalminimum.com/courses/quantum-mechanics/2012/winter
If "mathematics is completely independent from physics', then what's up with vector states, vector spaces and vector operators, matrix algebra, linear operators, commutators, Pauli sigma matrices, wave functions, inner products of system state vectors, probability amplitudes, density matrices and Fourier analysis?
http://theoreticalminimum.com/courses/special-relativity-and-electrodynamics/2012/spring
If "mathematics is completely independent from physics", then what's up with the Lorentz transformation, reference frames, the relativistic velocity addition formula, invariant intervals, space and time-like intervals, space-time four-vectors, cross terms in the Lagrangian, the Lorentz force law, Einstein and Minkowski notation, tensors, Susskind's four fundamental principles that apply to all physical laws, Maxwell's equations, solutions of Maxwell's equations and the use of action, locality and Lorentz invariance to develop the Lagrangian for electrodynamics?
http://theoreticalminimum.com/courses/general-relativity/2012/fall
If "mathematics is completely independent from physics", then what's up with the equivalence principle and tensor analysis, the geometries of flat and curved spaces, covariant and contravariant vectors and derivatives, geodesics, the metric for a gravitational field, the Schwarzschild metric, Kruskal coordinates, Einstein's field equations, the four-current, the continuity equation and the stress-energy tensor, or with how Einstein's equations can be linearized and how the linearized equation is a wave equation?
My point is that
physics seems to be up to its eyeballs in mathematics.
So you agree with me that mathematics is not the basis of physics. Good.
I'm not convinced that it's possible to separate the two, without returning everyone to qualitative physics as it was in the late middle ages.
If we think of 'physics' as the system of ideas, concepts and models developed by human beings over the centuries so as to better understand the physical world,
I think that it's indisputable that mathematics is inextricably intertwined with physics. I'd go so far as to say that the marriage of mathematics with our understanding of physical reality is the definitive change that defines the early (17th century) 'scientific revolution'.
So the (more metaphysical) question then becomes:
What does the value of applied mathematics to physics tell us about the nature of physical reality? How must physical reality be such that mathematics is so fruitfully applicable to it?
My own (very tentative) answer would be that physical reality possesses a structure and behaves in ways that correspond to the mathematics. That's why I keep harping on the
isomorphism idea. If the word 'mathematics' doesn't just refer to the symbolism scrawled on so many chalkboards, but refers as well to the abstract relationships that the symbolism capture (that's my point about
mathematical Platonism), then
I don't see anything wrong with saying that those same abstract relationships exist in physical reality too.
That's why mathematical physics works.
I don't really accept the idea that mathematics is all that physical reality is, that physical reality can be reduced to mathematics without remainder. But I do believe that whatever those abstract relationships are that mathematics explores, that they are truly present in physical reality as well.
Suskind's first two classes have been repackaged in the form of inexpensive (but rather difficult) books:
https://www.amazon.com/Theoretical-Minimum-Start-Doing-Physics/dp/0465075681
https://www.amazon.com/Quantum-Mechanics-Theoretical-Leonard-Susskind/dp/0465062903