Let me tell you why I love Richard Feynman, among the many reasons.  Feynman loved the process of science.  He was not wedded to some search for TRUTH.  He was a practical man.  He simply wanted a slightly more predictive model of the world than he had the day before.  And, he lived it.  The guy went to clubs and scientifically experimented with pick-up lines.

Due to my Feynmanesque pragmatic love of science (perfection is a fool's quest), I'm somewhat disturbed by Sean Carroll's logic
And, without fail, the scientific judgment comes down in favor of a
strictly non-miraculous, non-supernatural view of the universe.
This sounds all good and logical on its face, but what Carroll fails to address is the framework upon which the scientific method is built.  The scientific method makes two general assumptions, to which we have never observed an exception.

1. Any phenomena can be understood as an effect of the laws of nature.
2. The laws of nature are the same everywhere (i.e., not only are we in a special place, there are no special places)*

These are the base assumptions upon which the scientific method is built.  Logically, the method built on assumptions cannot be used to prove those assumptions.  It is like using a mathematics system based on the concept that 1+1=2 to prove that 1+1=2 .  Trivially, it is not surprising that the scientific method produces results that are consistent with natural cause and effect. 

It is our observations that additional assumptions are not necessary to explain phenomena and that those assumptions are not violated which demonstrate the continued validity. 

As a result, the scientific method cannot disprove the general existence of supernatural effects or the existence of a god.  The method requires that we assume they do not exist.  The method is capable of testing specific claims and effects, such as in Carroll's well phrased virgin birth example.

Similarly, atheism makes the basic assumption that there is no god.  In the realm of assumptions underlying faiths, atheism is by far the Occamiest of belief systems (i.e., makes the fewest assumptions).  That makes atheism, in the absence of the impossible direct test, the preferred hypothesis, but does not represent proof.

So, in summary, I agree with Carroll that our observations and experience do not require supernatural forces for accurate explanation.  I would, however, like to see the logic a tad less mangled.   

*Usually only astronomers and astrophysicists consciously worry about this one.

N.B. Predictably, PZ jumped on the bandwagon and wrote glowingly about Carroll's post (including making it mandatory reading for the PZealots), cause, like the rest of us, PZ loves him some confirmation bias.