One of the tricky things about science is that it takes considerable skill to reconcile experiment with even very solidly established theory, even with a good scientific background. For example, some poor soul at Physics Forums has set up a muon detector seemingly according to Doyle, and done calculations to show him how many hits he should get on his detector.
The problem: "I have observed a count rate of about 50 muons/day, . . . but 50 muons/day is no where near the 331,000/day I should be getting."
The noise has overwhelmed the signal and the person doing the experiment has only vague conjectures to explain why this is happening. Given the context of the post, the quality of past experimental work, and the accuracy of this part of particle theory, the problem is almost certainly one of experimental design or overlooked terms in a prediction equation. But, much of science is solving these basic kinds of problems, rather than contemplating new physics or squeezing the last part per million out of an observation.
One of the things that science education omits, until one reaches the quite high level at which one begins to do experiments like this one (junior or senior undergraduate science major or graduate student level), is the extent to which this kind of problem is the norm, rather than the exception.