I came across this philosophy paper (miter missive:  The Browser) which ponders whether the hypothesis of an omnipotent and omniscient God is any more likely to imply that God is good rather than God is evil.

Suppose, for example, that the universe shows clear evidence of having been designed. To conclude, solely on that basis, that the designer is supremely benevolent would be about as unjustified as it would be to conclude that it is, say, supremely malevolent, which clearly would not be justified at all.

The problem always appears at a much more basic level for me.  Suppose you are an omnipotent God.  What do you do?  Obviously to answer that question you should start by identifying all of the feasible alternatives (ok that one is easy, everything is feasible because you are omnipotent), rank them according to your preferences, and do the one that ranks at the top.  Wait a minute.  What are your preferences?

You are omnipotent remember.  Its not just that you get to choose your preferences.  Your preferences do not exist until you create them.  Ok.  So first you choose your preferences then solve the problem of what to do given those preferences.  How do you choose your preferences?  It is no help trying to choose the preferences that are easiest to satisfy blissfully because you are omnipotent.  All preferences are trivial to satisfy blissfully.  But why do you want to want that anyway?  How do you even know what you want to want?  You don’t have any preferences yet right?

So I think that an omnipotent God would be too neruotic to even get out of bed and decide whether to be good or evil.