OTOH this is a pretty conservative bunch.
It may "just" be something in the details (much like the Pioneer anomoly: http://www.planetary.org/blogs/bruce-betts/3459.html?referre... which was solved after 20 years of digging and LOTS of speculative ideas that did not work out).
At one point, the motion of Mercury didn't fit into the rules of mechanics, which we thought we understood. It turns out we didn't, and general relativity filled that gap.
It's far more likely the explanation is a mundane force not considered before, than new physics. But we have to look.
Note, this is an important anomaly not because of its magnitude, but rather, its relation to the measurement error.
The Pioneer anomaly was solved by precise calculation of the heat radiance of spacecraft.
But of course, it might end up being a trivial explanation (something on the measurement equipment, for example)
Another one of those questions: doesn't the acceleration of such a fly-by depend in large amount on the masses of the various objects, and if that's correct could this indicate an error in the mass of one of the objects (say, Cassini having a mass (not a weight) 1/1000000 less than we think it does)? And also, could it be indicative of a number of factors all being off by a fraction of the resulting discrepancy, or even two factors where one is a negative and the other is a slightly different positive?