Physics also uses statistics. Statistical mechanics is, in fact, one of the important branches of physics and is the underlying science of thermodynamics. However, the crucial difference between the statistics used in physics and the one used in other disciplines is that physics derives the probability laws governing its statistics, and is very general: it is used in the behavior of a gas, a liquid, a superconductor, a diode, a transistor, a glowing object, … .
If the removal of statistics takes away the prediction capability of a discipline, then that discipline is not a science.
9. Relation to Technology
Science improves the old technology and paves the way for the new. As in the prediction property of science, the time frame in which it becomes useful in technology varies. Newtonian theory of gravity was used in space technology almost three hundred years after its discovery. It took almost eighty years before general relativity became relevant in technology through the invention of GPS. On the other hand, less than thirty years after the inception of quantum mechanics, transistors were invented. Genetic engineering became possible about thirty years after the discovery of DNA. I don’t want its relation to technology to be a solid identifier of science. However, if a discipline claiming to be scientific has absolutely no application in technology throughout its history, then it may not be a science.
10. Science is not “Too Complicated”
The absence of decisive observation in non-science renders any theory questionable. A popular phrase that it used by both the proponents and opponents of the theory is that the problem under investigation is “too complicated,” with the intended meaning that no theory can explain the problem because “too many variables” are involved in that problem. Thus the proposed theory is neither right nor wrong, it is just incomplete. In this sense, science is not “too complicated” because scientific theories are verifiable.