The first two days of the week have essentially no actionable data on the schedule.  The effects were made painfully obvious yesterday as bond markets improved in the morning and sold-off in the afternoon on nothing more than tradeflows (i.e. trading decisions based on the nature of other incoming trades, as opposed to fundamental data or news).  There's a 3.5 day weekend ahead as well, and thus prime time for lighter participation among market participants.

The combination of lower participation and an absence of news makes it even easier for existing tradeflows to inspire movement.  Yesterday's probably originated with corporate debt hedging (large firms trading Treasuries to protect their impending bond deals from near-term rate volatility, much like "locking" in the mortgage world). 

The qualifier 'probably' is necessary because there's never any way to be 100% certain that corporate debt hedging activities are directly linked to how Treasuries and MBS are moving.  All we can know is that it was an active day for that section of the market and trading Treasuries is one of the main ways such deals are hedged.  In short, there's strong evidence.

Even then, it's important to keep in mind that this corporate credit market 'stuff' isn't an inherently positive or negative thing for bond markets.  It's only really useful to discuss when piecing together intraday market movers and only rarely has enough of an effect to act as a longer-term theme.

While we can't know the exact extent to which that is affecting markets, we can always assume that traders are cognizant of the technical picture when things are slow.  'Technical picture' refers to technical analysis, or simply trading based on various mathematical assessments of previous trading.

One of the most mainstream technical studies is the Relative Strength Index, or RSI.  This is an indicator of momentum that essentially measures how fast and by how much prices/yields are changing relative to past precedent.  It ranges from 0-100, but those extremes are never achieved.  Instead, horizontal lines are drawn at 70 and 30.  When the line on a chart touches or breaks past one of the lines, it's considered overbought or oversold, and increasingly at risk of a correction.

A slightly less basic, but still very popular application of RSI is for the purpose of identifying divergences.  These are instances where the chart of the security in question is doing one thing while the RSI line is doing something else.  The theory is that changes (or a lack of change) in momentum can be more revealing than the changes in price itself (or "yield" in the case of charting rates). 

For instance, if Treasury yields were falling, making new periodic lows, that rally would be 'confirmed' by momentum if the RSI line was also making new lows.  When that's NOT the case, it's known as a divergence and suggests a higher risk that yields would reverse.  This is essentially what's going on right now.

Now, of course if this was conclusive evidence of an impending bounce, markets would be broken and there would be no reason to trade.  Any ONE technical indicator really doesn't have much better than a 50% shot at accurately predicting the next move, but technicians best hope is that a consensus among multiple indicators will also jive with the fundamental (aka the opposite of 'technical') picture. 

In the current case, the fundamental picture is on hold while we wait for data or news.  Short term technicals are slightly negative for bond markets in the meantime, and traders are aware of that.  As such any additional weakness in today's session is susceptible to more 'snowball selling' as we saw yesterday afternoon.  Conversely, any firm strength not directly attributable to an event, makes a stronger statement about the staying power of the recent rally.