In scientific work, those who refuse to go beyond fact rarely get as far as fact.

Facts do not arrive unaccompanied; you have to guess first in order to know what to look for. Refusing to speculate sounds rigorous and mostly produces collections of data with no question attached. Guessing boldly and testing hard are two halves of one discipline, and dropping either half stalls the work.

On a problem you are stuck on, write down what you suspect is happening before you have proof. Then design the cheapest test that would show you were wrong. Being wrong quickly is the fastest way in.