Yeah the issue is probably partially weird results can be lazily explained as a number of sources; genuinely new science, experimental design problems, or equip

Yeah the issue is probably partially weird results can be lazily explained as a number of sources; genuinely new science, experimental design problems, or equipment errors to list a few, so if you're quickly reviewing a paper it's easy to just gloss over checking all the math because it's easy to trust on that. With computers we know exactly what we expect a piece of code to do and if it doesn't we're not discovering some new truth about the universe it's just that something was done incorrectly or our assumptions were wrong if it's someone else's code/program/library.