One of my favorite market indicators is our yield curve model. It has warned us several times in advance of significant correctionsthis year.
Warnings over the past few years have included:
July 16, 2018: The Yield Curve Update – We were a little early. SPX closed at 2798 that day, rose to 2940 before crashing 20% by Dec 26. The final 13% was signaled on Dec 5: The Yield Curve’s Warning.]
April 25, 2019: The Yield Curve Model Warns Again – SPX gained 21 points over the next four sessions before quickly shedding 226 points.
February 20, 2020: Buckle Up – SPX (which had topped out the day before) crashed by over 35% over the next month.
August 25, 2020: Update on AAPL – We were about a week early, but the model signaled a correction which saw SPX fall 11%, followed by another 9% the next month.
The recent breakout of the 2s10s is clearly a bearish signal – though it hasn’t yet paid off. Is the model still working? 
First, a little history. Among other things, the model holds that breakouts above significant resistance are bearish for equities.
If we plot the 2Y and 10Y together, we can see that significant sell-offs in stocks were marked by more rapid declines in 2Y yields than in 10Y yields (i.e., a widening of the spread between the two.)
The shaded areas below illustrate the period during which stocks experienced their most significant corrections between 2000-2013. Though the 2Y and 10Y both declined during these periods, the 2Y yields clearly fell faster.
But, as we saw in 2015-2016 and again in late 2019, not all corrections involved a steepening. These selloffs occurred without the yield curve model signal being triggered. 
Did the model stop working? Hardly. The decline earlier this year was a stark reminder of its predictive power. 
What made these corrections different? More importantly, what is the model signaling now, and how likely is it to play out?
continued for members…
We have discussed many times recently that the 2s10s has broken above the red TL marking horizontal resistance – yet no equity correction. Why not?
First, an overview of ES and SPX show they are still making new highs after seeing their channels break down. Note that what few downturns we’ve had since then have been very well defined, following tight channels and tagging TLs of support as if carefully managed.
The Nov 9 spike above the 3587 highs was attributable to good news on the vaccine front – but also had some help from other algo factors.
As we can see, this spike also tagged the rising yellow channel .786 line for the third time since Sep 2019. 
The next significant move should have been a test of the channel’s midline and, under normal circumstances, the channel’s .236 line and potentially its bottom. 
Although SPX never poked back into its broken white channel, it has made new highs as well after clearly breaking down. 
Its yellow channel is much less well-defined than ES’. 
We know from posts last month that Nov 9 is when VIX dropped down to tag its purple .886 line – thus precipitating the 10/20 cross a few sessions later. 
We also know that Nov 9 marked the day that CL broke out of a well-formed falling channel (below in gray) and topped its SMA200 for the first time since Jan 21, 2020.
Though RB had been occasionally bumping up above its SMA200, Nov 9 was the day it rose above it and remained above it. 
Nov 9 was also the day that USDJPY bounced an enormous 2.30 in one session after tagging its red .786 Fib.
And, it was the day that 10Y yields spiked up to .975 bps after languishing around .748 a few days before.
A close-up: 
Bottom line, it took all of these algo drivers working in coordination to push ES/SPX to new highs. Even then, the indices didn’t hold the previous highs. It wasn’t until Nov 23 that ES pushed back above 3587 to stay.
That was the day VIX gapped below its .886 Fib at 22.66, CL broke out, USDJPY popped back above its SMA20, etc. etc. etc.
It’s the same kind of nonsense that allowed this morning’s healthy decline reverse and break out.
VIX is threatening to drop through its SMA10. 
CL is back above its SMA10.
And, USDJPY is back above its broken TL from its Nov 9 lows and is threatening a rise above its SMA10.
As I’m typing these words, VIX has dropped below its SMA10…
…sending ES up to the top of a (new white) channel drawn using the recent lows instead of recent highs (relegated to purple.) It would make sense if the goal is, as I suspect, a backtest of the TL rising through 3658ish (where the SMA10 current stands) tomorrow.
Even the yield curve itself is cooperating. A reminder regarding the model’s rules as originally posted on Dec 5, 2018:
Reviewing the past year shows the carnage that happened when the yellow TL off the Aug 28 inversion broke down (Point 1.) when 2s10s bounced off the yellow horizontal support at Point 2, the downturn accelerated.
Point 3 represented a reversal off overhead resistance – theoretically bullish, but 2s10s subsequently broke back below the red TL – very bearish (a failed breakout, aka a breakdown) – particularly since it was followed by a breakout (Point 4.)
The retreat to Point 5 was bullish, as was the bounce there. We saw a minor breakdown to Point 6, which was bearish, followed by a breakout past overhead resistance to Point 7 – essentially the equal of the previous cycle high. Had it gone further, it would have been very bearish.
But, it retreated to the white TL where it bounced around from Jun 10 to Sep 3 before beginning a slow march higher – testing support at the rising yellow TL (Points 8 and 10) and resistance at the red TL (Point 9) along the way (bullish.)
Since its breakout above its June 5 highs, it has been carefully managed. Every time it appears to break out to higher highs, it is careful to reverse to its former support – particularly if stocks begin to falter.
It bounced at support, which was bullish (Point 11) and retreated quickly after threatening to break out again. You can see the slow, steady decline (Point 12) more easily in the chart below – this after a 33-pt decline in ES on the night of Dec 6.
Will 2s10s retreat from its latest threat to break out? Probably. But, one of these days – perhaps over the next 2-3 weeks or perhaps shortly after 2020 is in the history books – we are likely to see the breakout or breakdown that signals another downturn.
The key will be when USDJPY, CL and VIX are unavailable or have already played out.
UPDATE: EOD
As it happened, VIX plunged through its SMA10 – sending the all-safe signal to algos. ES came within 22 points and SPX within 12 points of its 1.272 Fib.
As we wait for DXY to reverse (or for confirmation that it already has) it’s worth noting that GC and SI reversed on target – GC at a backtest of its purple midline and SI at its white channel top and SMA100. Hmmm…
