ES Model

The night of October 17th, (last Thursday before all that strange behavior Friday the 18th) I had showed a CONTEXT ES model. Keep in mind that Capital Context is not a blog, not a website designed for people like us (that's why it's delayed 30+ minutes),  but rather it's an asset they sell to institutional or professional traders so they're not a joke, but after having watched them for a while and getting some information, we know a few things.

Here's the excerpt from that night's Daily Wrap with regard to the CONTEXT model for S&P E-mini futures (ES)...

"Tonight we are at -45 points. I recall an ES model (CONTEXT) that was something like -44 points or so, then ES pulled back to something like 39 points, it was amazingly accurate so long as they don't have to make some adjustment for a carry trade"

After having seen CONTEXT models at something like +40 points (rare in this market) or -40 points, the next day there's be some reset and CONTEXT would be at -5 points or something like that so I and a member did some investigating and found out that they re-calibrate their SPY Arbitrage model every week, I believe it's around 10 a.m. Monday mornings and they recalibrate the CONTEXT model on an "As needed" basis. Because the assets that go in to the make up of the model are "institutional in nature", they aren't typically traded by people like us, they include: Interest Rates and Curves, Credit and Credit Risk, Currencies and especially FX Carry trades, Commodities and Precious Metals. The idea is to understand what is a risk asset such as high yield credit or an FX carry trade and what is a safe haven asset which in the past could be precious metals in an environment where market participants have an expectation of rising rates, however in a stable interest rate environment gold takes on a different correlation. Treasuries have long been considered a Flight to Safety asset, however the F_E_D's treasury buying (QE) has caused some distortions in the correlation of that market as well.

All of this is taken and combined and weighted in to a single model and that model should reflect the fair value or anticipated value of S&P e-mini futures (ES) so arbitrage trading desks or traders can say, go short E-Mini's if the CONTEXT model is at a negative value and sell the E-mini if CONTEXT approaches fair value or above ES. This is actually a very novel indicator and I created something very similar a while before I had even heard of Context, some of you may recall "The Miners Trading System", which I'm going to pull up and run some back tests and see how it has performed.

The construction of my system was based on two asset classes you'd never think of. Most people think gold or the differential between the value of silver vs gold vs the geological value, meaning the proven reserves in the ground and what each should be trading at based on that, which would see silver much higher in value, but you can take that up with JPMorgan's manipulation of the silver market after they acquired a huge silver short with Bear Stearns that they've been defending for 4 years or so. No, my model was based on the two non-fixed costs mining operations face. You figure their land leases are known and fixed costs, the machinery and depreciation are fixed costs, labor for the most part (unless there's an uprising) is a fixed cost. What I figured to be their two largest non-fixed costs were energy and changes in currencies as a good portion of these are worldwide, but gold is sold in $USD denominations. So if the value of their foreign currency vs the $USD changed, so did the value of their sales, labor vs. sales and energy as it too is $USD denominated across the world.

In some scenarios with changes in currencies vs the $USD, not only are they receiving less for the gold they sell, it can cost them much more in Energy and a variety of costs, this is where I saw the biggest fluctuation in profit margins for gold miners as a group and thus what my model was based on and it's difficult to weight just those two assets ($USD and Oil) so I understand how complex the CONTEXT model must be. 

In any case, as I said on the 17th, " it was amazingly accurate so long as they don't have to make some adjustment for a carry trade" or any number of flip-flopping assets. As if I had predicted the re-calibration, a couple of days later the model was re-calibrated and at something like -7 points from -45.

Here's an easy explanation of why it needed to be re-calibrated, although I have no idea of which actual asset/s caused the re-calibration.

 This is the SPY (SPX) in green vs. TLT (20+ year Treasuries) in red. Normally Treasuries would trade the exact opposite of the SPX/market, it's typically a flight to safety trade, but as you  can see above in this most recent leg up from Oct. 9th, TLT is moving like a risk asset and almost in perfect sync with the SPY.

I don't think this was the actual asset, I'm just showing an asset that most of us identify with as being a flight to safety asset acting in the opposite way, that would cause the model to distort and the correlation would have to be re-calibrated.

Since CONTEXT was re-calibrated (it doesn't happen that often) here's what it looks like now...

As of about 5:30 a.m. today the model was around -12 points, not a huge differential, but big enough for a pullback. As of 3:42 p.m. today that grew to a -40.63 ES points negative. 

I found this ES chart this morning to be rather interesting...

Non-Farm Payrolls, the most important macro economic data in the US right now as the F_E_D has tied the unemployment rate to their interest rate hike schedule rather than the former calendar based schedule  (for those that didn't notice it, although the NFP missed, the unemployment rate dropped, that does not make institutional traders happy because the lower unemployment is, the closer we are to a hike in interest rates which sends a market down). We know the BLS has admitted to leaks in their data and that's why they are building a new data center so they can prevent inside trading on advance knowledge of the report.

I can't say whether there was a leak or whether there was a decision to run the flag up the pole no matter what as yesterday I clearly expected a move like this, in fact I predicated a Financial short position on this happening. 

If you look at the ES chart above and 3C, after doing nothing all night, we have a 3 hour, very flat range (typical of accumulation in this position) and 3C showing accumulation as the range went on, then up and away as the 8:30 NFP came out.

If you look at the CONTEXT chart...

The changes started around 8:30 as the differential expanded, but one thing I've noticed about CONTEXT is the angle of the green model. The differential can expand just because ES moves higher while the model is totally flat, in my view that does not suggest institutional selling in to the event, it suggests institutional assets remain constant and only the higher ES price creates the wider differential, today you can see that's clearly not the case as CONTEXT was heading down suggesting institutional assets were not only not buying this NFP move, they were selling in to it. However we could have guessed they were worried about the market just by the spot VIX being in the green +1.29% while the SPX was up +0.57%, the reach for protection.

In any case, I thought I'd point out this CONTEXT chart as it looks much more credible with institutional selling in to higher prices and a wide differential with a freshly adjusted model.