• The past few days of consolidation had a bullish tilt 
• Meaning: when yields broke out of a narrowing pattern, it was slightly more likely to be a positive break
• So far, so good, but room to run will eventually run out

In today's chart, you'll easily be able to see the rally that began the year for bond markets.  That rally lasted every bit of 6 weeks and covered quite a bit of ground.  The current rally trend is already almost 3 weeks old and has been covering ground at a similar pace.  Given that the January rally arguably had more going for it in terms of motivation, we may well wonder what kind of staying power the current trend will have.  In other words, the trend is good, until it turns bad.

In judging when good trends go bad, we can look at simple trendlines over varying amounts of time.  For instance, we can hold the current trend to its recently more narrow range (phase 2) or we can give it a bit wider berth with the phase 1 trendlines.  In either case, we'd be considering the trend broken, or at risk of being broken when one of the upper trendlines is crossed.

The simple linear trendlines are complimented by technical overlays (other forms of charting trends with some math applied).  The chart above has Bollinger Bands in yellow.  These are based on a 21-day exponential moving average in the middle, flanked by a certain value of standard deviations above and below the moving average, weighted by current trading levels.  Who knows what any of that means?!  But it is a very useful, mainstream technical study that does a good job of capturing almost all of the trading activity and serving as a clear rally or sell-off signal when the outer lines are broken.  In this chart, yields have some room to run before hitting the lower band, which can set the stage for a bounce higher unless rates break forcefully through the lower band.

Then there are the ever-popular stochastics in the lower section of the chart.  The parameters of this study are more open to user interpretation than Bollinger Bands.  Technicians can set up the chart so that the 2 lines rarely cross each other, thus giving later, but more reliable signals.  In this chart, my parameters are pretty conservative.  They didn't show the "cross" until March 21st, when the trend itself arguably began in the previous week.  On a positive note, there is also 'room to run' based on these stochastics.  If we charted them with narrower parameters, stochastics would already be well into overbought territory, meaning they were ripe for a correction/bounce.

Bottom line, it's all a question of how paranoid/sensitive you want to be to changes in the charts.  The most sensitive technicals already suggest caution, but the broader technicals suggest more room to run.  It's when they both say the same thing that we know things are definitely changing.  Unfortunately, rates will already have risen for several days by the time such a thing can be observed.  Therefor, the easiest conclusion is, and always will be: the trend is our friend, until it's not.