Saturday, September 29, 2012

COT Data for Commercial Traders is in, can we confirm?



USD:

COT report shows that commercial traders have been more urgently moving to the long
side, as if they are in a hurry to get into position for an up move.



Also the report showed British pound futures, where the commercial traders are net short the pound (long the dollar) in a big way.

Oil:

Strong correlation between crude oil prices and the exchange rate between the Canadian and US dollars. Using the COT data for the Canadian dollar futures as a surrogate sentiment indicator for crude oil, the past week shows a fairly urgent move toward a bigger net short position in the Canadian dollar, even as oil prices have already been pulling back. Note in May 2012 the same thing happened and it resulted in a drop of more than $20/barrel. Are currency traders betting on a pre-election release of oil from the strategic petroleum reserve, in hopes of skewing the election results?

Gold:
COT data also indicates a big drop coming for precious metals prices. Gold futures, where the commercials have jumped up to the biggest net short position since just before the big top in 2011.  They were a bit early of the final price top, but they did get proven right when gold prices declined almost $300/oz drop.

The bottom line, an urgency to go long the $USD after the $USD or US Dollar Index already had a significant move this week. Remember, the $USD trades primarily inversely to most risk assets, a strengthening $USD typically means lower oil, equity and historically before Central Bank Intervention, the precious metals. 

There also seems to be a round-a-bout move to get short oil in front of the US elections on thoughts there will be a release from the Strategic Petroleum Reserve, which is kind of crazy as oil is near the top end of its historical storage range, gas prices are not nearly as correlated to oil stocks as common sense would have it, so whether a SPR release will have an actual effect, no effect or a placebo effect, Commercial Traders are betting on a SPR release-perhaps the October Surprise of election year politics?

Yesterday I said the market is trading as if QE3 was never announced, in a pullback scenario I expected to see or at least I was looking for large accumulation as pros aren't going to chase prices higher, they'll bring them down and accumulate and then send them higher, but a noted yesterday, there just isn't any consistent sign of large scale accumulation in to price weakness.

Gold has to be one of the strangest assets as there's little arguing with the wisdom of owning gold during a deflationary dollar cycle, however in the same report, traders are betting on a stronger dollar and thus far that's what we have seen, it seems almost as if QE3 was never announced or it was so front run and priced in that there seems to be little viable upside. I don't now what the reasoning is, I'm just floating ideas, but the charts and data are what I want to follow as I have for years even if I don't understand the reasons at the time, many times we have been rewarded for going where few go and common sense doesn't seem to apply, but in the end, the signals are usually correct and eventually we find out why, but if you wait for the reasons, you missed the trade.

So can we confirm any of this data?

* None of these charts are meant to be trading signals, they are too broad for that, we are looking at the longer term trends to see if we can verify the COT data which would give us a distict understanding of how QE3 is being viewed and acted on at least initially, they may change, but as I said the first day it was announced, my emotional side wanted to close all shorts and go full long just because that's the Pavlovian-like conditioning we've had toward QE, I resisted that urge and said we'd look at the data first and not make emotional decisions and good thing we didn't at least so far as the market has certainly not acted like it's in a mood for risk appetite as the broad averages move lower since.

Again, we can look for opportunities on a trade by trade basis, this is just to look at the Commitment of Commercial traders and see if what the data is saying can be confirmed with 3C.

First the $USD and Canadian Dollar, the COT data suggests a long $USD and short Canadian Dollar which will tie in to oil.

Remember, these aren't trading signals as they are too broad for that and this doesn't suggest a straight line move, there will be bounces and volatility, but if we get the big picture correct, we can use the smaller moves (bounces and volatility) to position ourselves with the least risk, the best entries and probabilities.

US Dollar Index -I can only look at day+ long data here so we'll also have to use UUP for intraday data.
 The Dollar Index-DXY0 on a daily chart shows the negative in 2010 leading it lower, then confirmation on the move lower, then in the summer of 201 a positive divergence sending the $USD higher and confirmation of that trend, in fact while 3C continues to confirm the trend, there's a large leading positive divergence as price is lower than the 2010 top, but 3C daily is significantly higher. This isn't what conventional wisdom would suggest in a dollar debasement environment.
UUP for intraday
 4 hour shows several negative divergences and recently around the QE3 announcement, the start of a leading positive position in 3C.

 The 60 min chart shows the same leading positive position as the COT data would suggest.

 The 10 min chart also shows the same leading positive divergence.

Canadian Dollar-COT data suggests the dollar long, the Canadian Dollar short.

 4 hour CAD with a distinct change in character as of the 13th or thereabouts, why was this in full uptrend confirmation and then suddenly go negative on the QE3 announcement, that's a big change.

 Same with the 60 min CAD, a positive divergence sending CAD higher and then a distinct 60 min 3C change to the negative.

 30 min CAD shows a relative negative at the area of the 13th, but more importantly the leading negative divergence after hitting new lows on the chart in 3C.

The 5 min chart shows the very distinct distribution after QE3 announcement and a new leading negative low in 3C.


 Even the 2 min chart shows the abrupt change at Qe3.


USO-Oil
 4 hour chart from positive to confirmation to a large relative negative to an even stronger leading negative-this doesn't mean short term USO can't bounce, but the trend here seems clear.

 60 min USO negative at the QE3 area,

 15 min USO negative at a head fake upside breakout right after QE3 was announced and downside confirmation since.

GLD/Gold
 4 Hour chart shows accumulation, confirmation of the move up and a distinct, clear negative divergence after QE3, a leading negative divergence that is very sharp.

 60 min GLD also goes from confirmation to leading negative right at the move up the day of QE3 and the 14th- new lows for the chart.

 30 min GLD negative at QE3 on relative negative divergences and a head fake breakout with distribution.

 I noted yesterday the 15 min leading negative divergence that formed the last few days, very strong and clear.

 10 min GLD leading negative position in the range.

 5 min looks like the 15 with a recent leading negative the last 2 days and the distribution at the head fake breakout on the 21st.


Even the 1 min chart shows the head fake move on the 21st as distribution, a move up short term on a positive intraday divergence on the 26th and again that same leading negative on the 5 and 15 min charts.

I can't explain to you why the market is acting the way it is, for me it's little more than a curiosity, I just want to be in the right place. With 3C confirming the COT data, it seems the underlying trade i much different than the conventional QE wisdom, at least for now.

It's one thing for 3C to be saying and showing us these things, but when the actual commitment of traders (Commercial) is telling us the exact same thing, I think it warrants a little more weight.