Today's ISM services data (technically "non-manufacturing") was one of the week's two biggest reports apart from tomorrow's jobs data.  Like it's manufacturing-specific sibling, it came out much weaker than expected, and markets did what you'd expect.  Bonds quickly improved to the week's best levels.  MBS struggled to keep pace with Treasuries.  And stocks sold-off initially but ultimately reversed course as investors assumed the data implied better odds for more Fed rate cuts.

That's about as simple as today can be explained, and there's no need to make it any more complicated.  The bigger questions surround tomorrow's jobs report.  With labor markets as strong as they have been, does a strong NFP number do much damage to this week's rally?  With this week's other econ data coming in so bad, is a weak NFP number already assumed to some extent?  As always the truth and the reality usually end up falling somewhere between the logical answers to the questions above.

But for the sake of speculation, my view is that an 'as-expected' NFP reading (or better) would be a bit of a relief to traders at this point.  With the consensus at 145k, anything between 100-200k could go either way in terms of the eventual bond market response, and it could go the other way depending on the internals (unemployment rate, wages, etc).  A sub 100k number would likely be more meaningful than a number over 200k (unless it's WAY over 200k).