Not true. Before the theory of special relativity was proposed, there were experiments that seemed to indicate that the speed of light was independent of the observer, and there was reason to believe from the knowledge of electricity and magnetism that the speed of light might be independent of observer. Einstein just had the insight to explain these things in what now is the obvious way: the speed of light is independent of observer.
And once the special theory of relativity was proposed, it was obvious to everybody that there must be a more general theory that would deal with accelerating reference frames. Again, Einstein had the insight to claim that there was no difference between an accelerated reference frame and one under the influence of a gravitational field.
It took quite some time to work out the math properly for this general theory, but long before it was done, there was a known problem with the orbit of Mercury. Einstein showed that his theory of general relativity explained the observed orbit.
Now, it is amazing that these theories have proven to be so correct on such little evidence, but it seems like the primary reason why they have been so successful is due to the mathematical beauty of the theories. Mathematics, it seems, describes our universe to an unreasonable degree of accuracy.
So no, I still claim that we are usually wrong when we move into a new region of observation.