AQ coined a handy little phrase in 2009.

"Play the range until the range plays you."

It's implications are far reaching but as long as one's mind is wrapped around the various things it can mean, it has been strikingly profitable.

Ranges exist in dynamic inter-relationship to each other.  Sometimes they are horizontal.  Sometimes they're sloped.  Often there are numerous tiers within the range, but there's always an absolute limit that ultimately defines the range if you move away from the more concentrated movements.

It's like anything really, where there is a distribution around a general trend.  If the trend is really a trend, then the CONTEXT for the distribution of potential data points will be moving in the same direction, as a whole, with more data points near the mean/median, and with fewer and fewer data points the more you increase the standard deviation.

But just because there are fewer data points at the extremes, it doesn't mean there is less volume.  In fact, at these less frequently visited data points, volume tends to spike, as it is today.  This is a beautiful thing because it simply reiterates the validity of the range extreme, AND suggests to us just how epic it would be if that range extreme breaks.  In the following chart, which is your best big picture view of bonds at the moment, we do not currently expect the range limit to meaningfully break, but we don't rule out it COULD happen. 

The bottom line is that it's more likely to hold than not.  And if you're "playing the range until the range plays you," this is the last stop on the train for both MBS and treasuries.  Put it this way... Until/unless these outer limits break, this is the most significant argument for a lock bias so far this year.  Feel free to sprinkle in some of your own biases or understanding though!  For instance, if you're convinced stocks go lower and we experience some sort of double dip, you might not be so convinced in the firmness of the range.

At a certain point, however, we must contend with, and ultimately accept the element of uncertainty inherent in predicting mortgage rate future.  There are valid arguments for continued selling in stocks/strength in bonds.  Use your estimation of those arguments to make nominal course corrections to how you approach the higher probability outcomes.

We've mentioned in the past that technical trends are at their MOST significant AFTER they've broken. In other words, this particular trendline CANNOT, by the very nature of bond pricing, last forever. So when it does break, that will actually be a new instance of the trends PROVIDING INFORMATION. If you're inclined more towards floating, whatever movement we get around this trend will certainly be more informative tomorrow or next week when it's had a chance to either go with the expectation, or break out and confirm the break.

Stocks are most likely set up to rally...  Bonds are sharing a ride with stocks...  Plan accordingly...