Ovinomancer
No flips for you!
You're confusing the scientific method for observation. Also, in a world of magic, maggot may sometimes come from eggs and other times spontaneously appear. You're making the assumption that there is a consistent underlying mechanic in a fantasy world that isn't mutable. Once you acknowledge that causes can change from one example to the next or even that causality isn't consistent, the scientific method fails.Or in a world where maggots do appear spontaneously, the scientific method will show that.
Today, maybe. Tomorrow the god changes it's mind, relents on Fred, and instead gets angry at people with brown hair. Next week it's just Tuesday's that draw it's ire, and you need double the number of eyes on Tuesday. Until a bit later when it's back to people named Fred.Then the scientific method will show that after the other variables are eliminated, there is still an additional factor that is affecting the number of newt eyes required.
Then given enough data to be able to apply the scientific method, we will see that experimenters called Fred need to use more eyes than those called Bob.
In time, other variables can be calculated and eliminated, and eventually the number of newt eyes needed will actually be able to be used as a measure of the god's happiness.
Causality does not need to be consistent in a fantasy world. In fact, in any game run by a DM and not natural principles, causation in game isn't consistent. Sometimes it doesn't even exist, or have you never had something happen in game and then gone back and detailed why that thing happened like that?
At least some of that concept is the basis for the Eberron setting.
Note that there is a difference between the scientific method flat out not working, and someone being unable to resolve something using the scientific method due to lack of data.
Actually creating a world in which the scientific method doesn't work is going to be really difficult: - it would require interference by outside beings in all processes, right down to individual thought processes and basic physical interactions.
Sure, if you want to say that you can always attempt to use a screwdriver, even if it will never tighten a nut, we can agree -- you can. But the argument wasn't that you can always use the screwdriver, it was that the screwdriver always works. And science is just a tool people use to determine aspects of their existence. It isn't a magical force that always finds truth. Science can't even say that science is the best tool to use, but it's ridiculous for a tool to be able to prove it's the best tool. We, as a society, tend to elevate science to some pinnacle of awesome and as the ultimate finder of truth and facts. And it's damn good -- best tool we have, in my opinion -- but it's not that, it's just a tool used by people, people that can be flawed or mistaken or incapable of properly applying the tool. I love science. It's great. But it isn't everything, can't be everything, and isn't a magic wand that can be waved at any problem in any context and find truth. We don't really even know if what we've discovered using it is actually true or just a good abstraction of what we can perceive of our universe.
Putting science on a pedestal is a mistake akin to worshipping your toolbox.
As for you last statement, that's trivially untrue. YOU decide how your world works, and you generally are deciding that at the macro level where it directly affects your players and not at the causation/underlying physical principles of your game. Don't confuse a limited consistency for actual causal mechanisms.