Christopher Columbus had a theory that there was a faster, more direct path to the East Indies which involved sailing west from Europe. He did find land, but not the East Indies.
During his entire lifetime Christopher Columbus believed, or at least he did not admit to anyone, that where he ended up was not the East Indies, but what would later be known as America. This failure to recognize the obvious resulted in North American aboriginal natives being referred to as Indians for 500 years, and America to be named after someone else, explorer Amerigo Vespucci.
The problem with theories is that they carry a lot of human baggage. Theories don't provide "the truth", but they are often used as such, even when subsequent observations are made that contradict such theories. In medicine, for example, frequent mammograms are promoted as a way to tackle breast cancer, even though statistics don't show an increase in life expectancy. Same goes for drugs that control blood cholesterol.
This is one reason why I find it odd that people insist that the "proper" way to develop a trading system is to first establish a theory, then design a system based on that theory. The end product, if useful at all, is typically rife with human biases, and justified with selective empirical data.
Many of the most brilliant theories in science have resulted from empirical data, not the other way around. Einstein's Theory of Relativity, for example, would never have come into existence without the Michelson-Morley and other experiments in the late 19th century. The Big Bang Theory came about after observations that galaxies were observed to be moving away from our own, determined by observing the Doppler Redshift. It is pretty much inconceivable that either of these two theories could ever have come into existence without prior empirical data.
Einstein once stated "It is very possible that without these philosophical studies I would not have arrived at the solution", referring to philosopher David Hume's 1738 'A Treatise of Human Nature'. "Hume was an empiricist and skeptic, believing that scientific concepts must be based on experience and evidence, not reason alone."