The frustrating thing about today is that we can't really know if NFP came in at just the right levels to avoid stirring the pot too much, or if markets weren't that keen on having a big reaction, regardless of the result.  Personally, I think a big reaction was very much in the cards, and as we discussed throughout the week, the burden of proof would be on weak data to dissuade the Fed.

In other words, an 'as-expected' report or better would likely have done damage.  At first glance, this report was pretty close to 'as-expected.'  Not only that, but some of the positive attributes offset the negative headline.  Labor-Force Participation held steady, which makes the drop in the unemployment rate more meaningful than it otherwise would be.  Hourly earnings rose by more than expected, and positive revisions over the past two months meant that today's headline was only 3k jobs away from forecasts (47k miss on the headline vs a positive revision of 44k).  Top all that off with a comment from the Labor Department that this report tends to be revised higher in coming months, historically. 

The preceding list of "yeah, buts" goes a long way toward helping us reconcile the initial move into weaker territory.  Late day weakness in European markets and oil prices helped bonds find their footing.  Or if you want to look at things from a pure Fed-based perspective, you could say that markets had been leaning a bit more assumptively toward a September hike and today's data made it a closer call.  In that sense, being in modestly stronger territory is understandable.

A better explanation for the initial weakness and subsequent strength would be that the initial weakness wasn't weak enough to create momentum.  So traders squared positions ahead of a 3-day weekend.  In today's case, that meant exiting bets on rates moving higher (aka "short covering," which is simply "buying bonds").