It's been a pretty boring and consolidative week for bond markets so far, so let's have some fun.  What follows is essentially a bullish case for bond markets, and you should be aware of some caveats before we begin.

1. The conclusion is based on past precedent, which isn't always a guarantee for the future

2. The past has been generally bullish for bonds for 30 years, give or take, and much of the current bearish case is based on the notion that 2012's low yields were generational

In other words, the past has indeed been a good indicator for the future of bond markets as long as we've been in this 30yr bull trend and today's bullish conclusion draws on past precedent.  If it's true that we put in long term yield bottoms in 2012, then we'd logically start seeing changes in the predictive value of what had been reliable fractals.

In OTHER other words, certain shapes on charts have repeated with better-than-random regularity.  This has made it worth mentioning any time those shapes pop up because the stuff that happened after those shapes popped up in the past has been more likely to happen in the future.  Some day that might change and those shapes might not mean as much.  If now happens to be that time--well...  sorry.

Even then, no example of past precedent is perfect, but for those of you more interested in going against the grain and maintaining some hope for a positive resolution to the recent pull back in rates (or put differently, those who see recent volatility as just that, but who don't think it affects the longer-term downtrend in rates), this chart's for you.  Analysis to follow.

2014-10-21 comparison

So what are you looking at here?  In short, this is a brief history of post-crisis rally spikes that meet similar conditions to our recent rally spike in bond markets.  Some of the conditions I looked for are as follows:

1. All of them had a day that was noticeably the most volatile and which contained the outright intraday low.  These are labeled as the "0" in the chart. 

2. They had to occur in an existing downtrend.  In other words, there were plenty of examples of big downward spikes, but many of them were simply the snap back after a recent, big upward spike. 

3. They all had to have a move higher in yield that was somewhat abrupt that also topped out and reversed within the preceding 45 days.  To systematize this, I simply looked for instances of yields moving from the middle Bollinger Band (those blue lines) or lower to the upper Bollinger Band or higher before embarking on a rally that ultimately crushed the lower Bollinger Band on the "zero" day referenced above.

To my surprise, there were only 2 other examples of remotely similar moves.  Actually, I shouldn't have been too surprised considering we know the recent move was a bit of an aberration.  Still, it's usually not that hard to find similar trading shapes in the past.

Now for the fun part of the conclusion.  In both past cases, bond markets consolidated for 5 days before resuming a longer term downtrend.  And today is day 5. 

MBS Pricing Snapshot
Pricing shown below is delayed, please note the timestamp at the bottom. Real time pricing is available via MBS Live.
FNMA 3.0
100-20 : +0-00
FNMA 3.5
103-23 : +0-00
FNMA 4.0
106-09 : +0-00
2 YR
0.3582 : -0.0083
10 YR
2.1930 : -0.0340
30 YR
2.9650 : -0.0350
Pricing as of 10/22/14 7:35AMEST

Tomorrow's Economic Calendar
Time Event Period Forecast Prior
Wednesday, Oct 22
7:00 Mortgage Market Index w/e 370.4
8:30 Core CPI mm, sa (%)* Sep 0.2 0.0