Quite a few talking heads and analysts referred to last Wednesday's post-Fed rally as 'overdone.'  They pointed to Thursday's moderate weakness as evidence that bonds would have a hard time making further progress at the new long-term lows.  

When the rally extended sharply on Friday in the overnight session, pundits doubled down on the 'overdone' calls, saying surely a bounce was imminent.  A few hours later, weak Markit PMIs in the US launched yields even lower prompting--you guessed it--even more calls for a jarring return to reality in the following week.

If you were around yesterday, you know that such a return (to higher yields) never happened.  Bonds continued to defy calls for a bounce by surging as low as 2.337 yesterday afternoon.  At that point, there were far fewer experts willing to declare the likelihood of a bounce--just what the market needed in order to finally bounce!

To be clear, bounces ALWAYS happen (eventually) after big rallies like this, but they relish the opportunity to throw ankle-breaking head fakes in the process.  It's fair to say that the past 3 business days broke more than a few ankles.  My position has been and continues to be that we'd need to see a certain amount of weakness before battening down the hatches and bracing for a negative shift in momentum (even if we'd only be concerned about short-term momentum at first).

How to identify such a shift?  This is one of the highest and best uses of technical analysis.  That esoteric wizardry that some analysts use in an attempt to predict the future is actually best suited to helping us identify meaningful shifts.  The most basic technical patterns are in good supply right now with several instances of yields bouncing at clearly-defined pivot points.  Of these, 2.47% in 10yr Treasuries is probably the most significant.  That's the ceiling I'd be watching for a break before getting too concerned about anything more than a small, brief consolidation.  If it's broken convincingly, 2.55% becomes the next level to watch.

As the chart suggests, 2.42% is a technical level on the other side of this morning's trading.  Breaking below there would reopen short-term rally potential.  But the emphasis would be on short-term.  Past precedent suggests bonds will most likely be drawn back at least to 2.47% (if not 2.55-2.62%) before being able to make significant progress.  There are historical exceptions to that pattern, but they've all involved massively obvious motivations (like the European crisis and US fiscal drama in mid-2011).