The term 'pivot point' can mean slightly different things depending on your frame of reference and market-watching goals.  For our purposes, the litmus test for a pivot (or inflection) point is the following:

• Have bonds been more likely to bounce vs break through when approaching the level in question?  
• Have bonds sometimes broken through and then returned to bounce at the same level coming from the other direction?

If we can answer "yes" to both of these questions, that's a pivot point.  If we're only seeing bounces on one side (i.e. continually hitting a certain ceiling in yields), we would refer to that merely as a "technical level," "ceiling," or "support."  If yields were treating the level as a floor, the technical term is "resistance."  

Whether we're talking about ceilings, floors, or pivots, the more frequent the bounces, the more relevant the level in question.  "Relevance," in this case, doesn't offer very much in terms of a crystal ball for predicting the future (though, to be sure, if you're going to make a wild guess as to where a particular rally or sell-off will end, your chances are better than random if you choose relevant pivot points).  Rather, the highest and best use of any of these technical levels is to serve as milestones for the legitimacy and significance of market movement.

In other words, if bond yields consistently bounce at a certain and then break above that ceiling, that break is more significant (or relevant, legitimate, meaningful, ominous, etc...) than any other random move.  If bond yields then return to the same level and now treat it as a floor (by bouncing there consistently), that level gains further technical significance.  If yields happen to break through the floor, we could infer more about a positive shift in momentum than another random move of the same size.

Pivot points also help us define ranges.  If we have several adjacent pivot points with one of the central getting the lion's share of the floor/ceiling bounces, that level becomes the de facto center of the current range.  Higher and lower pivots let us know if bonds are more apt to consider breaking higher or lower out of the range.  

With all of the above in mind, there are several important pivot points in play at the moment.  Even as I write, 10yr yields just bounced at 2.35% again.  

With yields at 2.34% currently, the biggest risk is obviously a break above the 2.35% level.  From there, a break of 2.42% would be grounds for ducking and covering for further bond market carnage.  

If we manage to hold below 2.35%, it allows us some small measure of reassurance, and an opportunity to attack the 2.29% pivot (which has been the most active pivot point of the past few weeks--the de facto center of the range).  If we can break below 2.29%, it would be the best case scenario for near-term momentum.  We'd need to see a break below 2.18% to get our hopes up and ultimately below 2.07% in order to get downright excited.  

Sorting out the relative impact of news, econ data, and incidental trading is tricky business right now.  Data matters to some small extent.  News can matter if it's the right news, but unfortunately, we tend to learn which news matters based on how markets react to it.  Obviously, news that affects the potential policy path of the Trump administration is high on the list of potential relevance.  

European geopolitical news is also a consideration, with the Italian election this weekend being one of the hotter topics.  "Incidental trading" (a term I'm using to attempt to capture the trading that will occur regardless of news or data) is always a factor at month-end, and is even more of a factor after bonds have made a big run and stopped at levels with long-term technical significance (indeed, the 2.2-2.5 range in 10yr yields contains a hall of fame when it comes to long-term pivot points).

Considering the current trickiness of sorting out what's moving markets and why, the pivot points become our best friends.  Anyone looking at market movement could make a case for one factor over another right now, but breaks and bounces at the pivot points don't lie.