Thanks to reader Billy for pointing out (in response to our prior post on gold’s correlation divergence) that there is a seasonal bullish tendency in gold that begins in September and for providing the following links:

http://www.321gold.com/charts/seasonal_gold.html
http://www.tradersnarrative.com/gold-seasonality-turns-positive-2909.html

We did our own analysis and confirmed this is indeed the case (see chart below).  Over 20 years, there is an expected push beginning in early to mid-September that tops in mid-February of the following year.  We can see that the predicted seasonality for the past year (red) was changed very little after adjusting to add the current year’s data (white), so seasonal factors should definitely be given weight in our analysis.  However, it is very difficult to time markets based on seasonality.  For instance, last September had a huge rally that was reversed in October, only to be reversed again (to the upside) into February.  So, while the seasonal pattern held, there was much consternation for the small and short term traders.  Accordingly, we believe it takes at least two weeks of data to confirm a seasonally-induced move.  Today may simply be the beginning.

Does all this change our previous analysis?  Only slightly because we must still wait for confirmation, but we now have a slight bias in the three three possibilities we previously outlined:

1)  Recent correlations hold and gold is simply forecasting another inflation-led [stock] rally that will begin in the next few days (meaning yesterday’s correction is not the bearish bellwether everyone thinks it is), or

2) Recent correlations hold and the move today in gold is a bull trap that will be reversed in the next few days, or

3) Recent correlations have broken and gold is decoupling from other markets, including equities.  Because correlation-changing paradigm shifts are rare, this is the least likely scenario; however, should it end up being the case, the significance is tremendous.  A break above 1,000 on a decling stock market would be definitive evidence for this scenario.

If today’s surge in gold has follow through and is the beginning of a seasonal move, then scenario 1 is most likely, which means the current stock market correction will be shallow and we will have another short covering rally that has characterized this year’s run up.  However, the if’s must be confirmed and, as we said before, all we can do is watch closely for now.

11:43 am EDT:  We expected rotation around the 995.25 highest volume point of control, and that is what has happened so far, with the ES having first traversed 5.25 points lower than 995.25, then 4.25 points higher.  The range should stay tight until the 2:00 pm FOMC minutes are released.  If a strong move develops on the release, we won’t fight it.  However, if the reaction is calm, we would still fade short 1004.00 to 1006.00 and would now fade long 988.00 to 992.00.

Gold has broken above its consolidating wedge and previous swing high of Aug 6 09.  The significance is that it is in the face of a declining stock market.  A strong distribution day in equities suggests deflation, which should push gold down.  Gold was our only leader yesterday that did not move as expected (according to correlations that have held for the last year).  Markets needn’t move in lock step even when they are correlated; however, today’s action must be paid heed.  To break an important technical level in a direction when the reverse is expected is significant.  The three likeliest explanations are:

1)  Recent correlations hold and gold is simply forecasting another inflation-led rally that will begin in the next few days (meaning yesterday’s correction is not the bearish bellwether everyone thinks it is), or

2) Recent correlations hold and the move today in gold is a bull trap that will be reversed in the next few days, or

3) Recent correlations have broken and gold is decoupling from other markets, including equities.  Because correlation-changing paradigm shifts are rare, this is the least likely scenario; however, should it end up being the case, the significance is tremendous.  A break above 1,000 on a decling stock market would be definitive evidence for this scenario.

We’re assigning equal weight to scenarios one and two for now, watching closely gold and the S&P 500 for clues as to which will play out.

The Precise Take – Expecting consolidation today

Time Profile:  For days on which the prior day closed down 2% or more, the Time Profile for the March rally (page 2) suggests a bullish bias throughout most of the day, with the final 45 minutes going either way.  However, the Time Profile going back to October 2007 (market top), for which we present down to the 15 minute level, has mainly a bullish bias only in the first 15 minutes (possibly because of gap ups) and from 2:30 to 4:00 pm.  If we are indeed beginning a material market correction, the Time Profile for March 09 onwards may be less relevant than the longer term one.

Trading Today:  Yesterday, the ES reached our first downside target, which was the highest volume market profile point of control (995.25), and which was the low of the day.  Overnight, the ES has traded below this level and we expect oscillation around it today. Accordingly, there should be a range day with maximum low of 984.00 and maximum high of 1006.50.  FOMC minutes are released at 2:00 pm today.  With the Fed making fewer waves on announcement day and, instead, electing to make policy changes outside the meeting (such as yesterday’s Agency and TALF announcements), the minutes could be more important than the actual announcement.  However, we don’t expect the same volatility in the markets as on…

Continue reading here.