Citing Occam's Razor as "If two theories explain the facts equally well then the simpler theory is to be preferred," is both misleading and gratuitously incorrect. That explanation is, reflexively, the simplest interpretation, but neither necessary nor true. Let me explain, beginning from the beginning.
Depending on your source, the original Latin is one of:
- "Pluralitas non est ponenda sine neccesitate."
- "Frustra fit per plura quod potest fieri per pauciora."
- "Entia non sunt multiplicanda praeter necessitatem."
All meaning, roughly "Entities should not be multiplied without cause." Unfortunately, this is a little opaque for the English-speaking crowd, who tend to prefer:
"We are to admit no more causes of natural things than such as are both true and necessary."
This is a very important distinction. First of all, the word "simple," as used in the common misinterpretation, is highly ambiguous. Many people interpret "simple" to mean "obvious" or "straightforward" while others take it to mean "easy to understand" or "unburdened by rigorous proof." In all of these cases, wildly different conclusions can be drawn, and the use of the this, faulty, Occam's Razor is almost totally useless. This leads to many circular debates, especially about religion, and cosmology in particular, with both sides gesticulating madly and invoking Occam's Razor to no effect.
The correct interpretation, however, does not share these faults. It is also not nearly as powerful as its faithful abusers would like it to be. An attempt to explain it is: Given an understanding of natural phenomenon, it is incorrect to conclude that which is not required by the nature of that phenomenon.
A correct use of Occam's Razor would be a counter to the following argument: "There is thunder and lightening. Therefore, we know that Thor is upset." This is an invalid argument because a thorough understanding of thunder and lightening does not require that Thor is the cause. Note that the use of Occam's Razor in this case does not refute that Thor is the cause, it just denies that assumption based on the evidence. The distinction between denying the assumption and denying the fact is important, because, really, Thor could be causing thunder and lightening indirectly, and modern science just hasn't discovered this. Occam's Razor is useful because it admits multiple, indirect causes of events, not all of which are clear at any given time.
The flaw of Occam's Razor is that it presupposes a correct understanding of the natural phenomenon. Which reduces its usefulness greatly in discussions of cosmology, although it may still be used to counter leaps of logic.
In summary: It's a razor. Don't play with it unless you know how to use it.