Amanita Market Forecasting

Bradley-Siderograph archive 2007 - 2009

(30.06.2009)

to the archive 2003-2006          to the article BS since 2009

6/30/09

Surprising insights into the Bradley siderograph

„Bradley" and „siderograph" continue to be the most frequent search terms on http://www.amanita.at/, so the interest is still very high. Since the latest free comment on the siderograph dates back more than a half year, I'd like to dedicate this newsletter to the Bradley and present, among other factors, a small quantitative study with surprising results. The siderograph can be found on many web pages but Amanita Market Forecasting is the only place to find this type of groundbreaking research!

The past turning dates were (standard window +/- 4 calendar days, sometimes up to +/- 1 week - basis S&P 500 close):

• Dec 14, 2008 (low, important): 12/16/08 short-term high for 2-3 weeks (deviation: +2 days), in early January the indices went somewhat higher
• Jan 20/21, 2009 (high): short-term low 1/20/09 (deviation: to the day)
• Feb 8/9, 2009 (low): short-term high 2/9/09 (deviation: to the day)
• June 3, 2009 (high): intermediate-term high June 2-12 (deviation: none, extended topping out within tenths of a percent over 10 days)
• June 26, 2009 (low): intermediate-term low June 22, 2009 (deviation: -4 days)

In the past 12 months we had the following 5 major (multi-month) equity market reversals:

• (1) 7/15/08 low
• (2) 8/11/08 high
• (3) 11/21/08 low
• (4) 1/6/09 high
• (5) 3/6/09 low

Boy, oh boy! None of the 5 was caught by the standard model (although the heliocentric model did a much better job)! So the Bradley recognized only minor reversals in the past year. Especially painful and irritating for many is that the bear market bottom in early March was missed. The premium subscribers of Amanita Market Forecasting were informed just in time, however, as 2 days after the March 6-9 low an update was sent out suggesting that probably a low of ‘epic importance' (low of 76.6 year cycle) had been formed, and a monster rally would be due. And as a matter of fact, we have meanwhile witnessed the biggest advance in 71 years. In a nutshell, the Bradley siderograph is grossly overestimated (in its traditional interpretation). That's why in my work this model is just one of 20+ factors in the CSQN model.

Remaining turning points in 2009 (the Bradley 2010 is released in late 2009):

• Jul 14/15 (important)
• Sept 14/15
• Oct 22/23
• Nov 9 (important)

Of the past 5 turning points, the polarity was right twice (i.e. Bradley high = index high, Bradley low = index low) and 3 times wrong. This confirms the warning repeated all of the time, "The Bradley predicts turning points ONLY and not the polarity, i.e. a high in the chart may also be a low and vice versa."

However, that doesn't mean that the Bradley isn't saying anything at all about the direction of the financial markets, let me share some astonishing findings. I have calculated the Pearson product-moment correlation coefficient (r) of the siderograph with oil, EUR/USD, SPX, gold, and what (latest data point: 6/12/09). Below please find the results for the geocentric and heliocentric model (geocentric means from the perspective of earth, heliocentric from the perspective of the sun).

In the 2^nd column please find the average for the periods "since 2007, "since 2008", and "since 2009", the reason is to give the newer data points more weight. At the bottom please find the mean for each market, e.g. the average heliocentric correlation r (CL, SP) = 0.823. Interpretation:

• Crude oil: The 93.4% correlation for CL in 2009 (r²=87%) is almost too good to be true, especially if you take into consideration that over the past 180 trading days the correlation of Nasdaq 100 and Dow Jones was a mere r=55% (http://www.mrci.com/special/correl.htm). Convincing is not just the stability (the lowest correlation was still a good 73.5%) but also the rise each year geocentrically, as well as the agreement of the geocentric and heliocentric Bradley (average correlation in both models 82%). Thus the siderograph is extremely important for the price of oil and, at the time being, mostly a model for crude but not for stock indices!
• euro/dollar: A bit of all right also the correlation of 83% since 2008 and still 75% since 2007 (N=891) for the euro compared to the geocentric Bradley. In addition, the correlation is rather stable over different time frames and quite similar geocentrically and heliocentrically, so the Bradley is good to forecast the EUR/USD rate.
• stock indices: The correlation with the stock markets is unstable and fluctuates between a useless r=43% and a high r=88%. Forget anything below r=50-60% (r²<36%). The correlation with the rising BRIC countries is better, a sign that they are attempting to take over global leadership.
• gold: Even less useful is the Bradley for gold where the correlation is below 50% on average. Interestingly, gold has reacted much better to the heliocentric (r=54%) than to the geocentric model (r=40%).
• wheat: This correlation is very weak, inconsistent and not usable in forecasting, especially in the heliocentric model with r=28% (r²=8%). In contrast to gold, wheat is more leaning towards the geocentric siderograph.

Conclusions from this little study:

• timing (window +/- 1 month): The correlation with the siderograph is rising in bad times because of the synchronization of the markets (the highest in 2008) and declining in calmer times. One has to allow a fairly large window of +/- 1 months (or even more): even oil with its high correlation still diverged for weeks from the siderograph.
• differences between markets: the Bradley works best as a trend intensity indicator for oil & euro, as opposed to the popular wisdom the correlation with the stock market is barely useful. Gold and wheat don't pay attention to the Bradley. At any rate, the siderograph should never be used as a stand-alone tool.
• geo & helio: one should always monitor both models in a holistic synopsis as they are almost equally powerful with r=61-62%, i.e. the standard model alone is not sufficient for the analysis.

Bradley 1852-2040

After this short-term view now to a very long-term one. Below please find the siderograph chart 1852-2040 (calculated with the „Market Trader" by Alphee Lavoie):

The formula assigns harmonic angles (60°, 120°) a positive value and disharmonic angles (90°, 180°) a negative value, the sum of all values is the chart as you know it. The lower, the worse the (economic) situation on average. The all-time low was in 1931 during the Great Depression, the model was also very low in 1875 after the crash of 1873, and in 2010 we are approaching the level of 1875. Isn't it interesting that the two historic models for the current situation are the years around 1931 and 1875? The Bradley low of the past 500 years was in the 1640s by the way, at the end of the Thirty Years' War (1618-1648) and after the first speculative bubble in history had burst (1637 tulip mania in Amsterdam). The reason for the low in 2010: the massive tension angles between the slow planets (T-square, partly even Grand Cross).

11/13/08 

 This is the standard model for 2009 (December 2008 - January 2010):

It has the following turning points:

• 12/14, 2008 (important)

• 1/20-21, 2009

• 2/8-9, 2009

• 6/3, 2009

• 6/26, 2009

• 7/14-15, 2009 (important)

• 9/14-15, 2009

• 10/22-23, 2009

• 11/9, 2009 (important)

Some Bradley model analysts would also include "micro spikes" but I strongly advise against that because that's not significant enough and should be interpreted as "white noise". For unknown reasons there are even minor differences between the different software programs that all use the original formula of Donald Bradley. While major turns in the Bradley chart are more or less identical you would get entirely different results if you zoom in too much.

The Bradley hasn't been discussed for over a year, what was the impact of the Bradley dates since 2007?

1. 12/22, 2007 low: no intermediate-term reversal but short-term secondary high
2. 3/8-9, 2008 high: intermediate-term low SPX 3/10/08 (deviation: + 1 day)
3. 4/7/08 low: no intermediate-term reversal
4. 4/27/08 high: no intermediate-term reversal SPX but intermediate-term top EuroStoxx 5/2/08 (deviation: +5 days)
5. 5/24/08 low: intermediate-term high SPX 5/16-19, 2008 (deviation: -5 days)
6. 6/7/08 high: no intermediate-term reversal SPX but intermediate-term top Nasdaq 100 index 6/5/08 (deviation: -2 days)
7. 9/9/08 low: intermediate-term high SPX 9/2/08 (deviation: -7 days)
8. 9/20/08 high: no intermediate-term reversal

We see that 3 of the 8 (38%) Bradley turns marked intermediate-term reversals in the main benchmark SPX, and another 2 Bradley dates were major reversals in another key index (EuroStoxx resp. NDX), which gives a total of 5 out of 8 (62%). However, these turning points were rather imprecise, 4 of the 5 turns came within +/- 5 days, once even the extended window +/- 7 days was used.

The Bradley chart is always marked with the hint that the dates only show turning points and not polarity, i.e. a siderograph low has almost the same odds of nailing a stock market high or a low (the opposite is true for siderograph highs). Since late 2007 the same 50:50 pattern can be detected.

However, that doesn't mean that the siderograph is telling nothing about direction, even though the exact rules can only be determined with the aid of rigorous statistical testing. Since 1950 the Pearson correlation coefficient r (geo, helio, N=21.459) is at 2-3% which means that the correlation of the Bradley with the Dow Jones is zero, i.e. in general the daily direction (up or down) of the Bradley is not indicative for the stock market. The rule is that r can be between -1 and +1, with +1 meaning a perfect match and -1 a perfect inversion. Everything larger than r>0.7-0.8 (r²>0.5-0.6) is useful in principal, although one should focus on correlation coefficients of r>0.85 (r²>0.7).

However, everything depends on a moderating variable: whether the index is in a bull or bear market. Since 3/2008 r (geo, helio; N=210) = 76%, i.e. the Bradley has been quite useful in predicting the direction (up or down) of the Dow Jones. From 1/2002 until 6/2003 we had comparably emotional markets as since 2008 and the correlation was almost the same:  r (geo, N=520)=78.4% and r (helio, N=520)=67.6%. In the bear market 1973-74 we saw r (geo, helio, N=701) = 0.76 and thus again in the same region. In contrast, during the calm time 6/2003 - 3/2007 we had a quite negative (!) correlation: r (geo, N=1362) = -0.43 and  r (helio, N=1362) = -0.62. A negative correlation coefficient means that the siderograph and the Dow Jones tend to trade just inverted.

Somewhat exaggerated one could say: the dates of the Bradley do work regardless of the trend (bull or bear) whereas the direction of the Bradley only works in bear trends (in bull markets there is a weak bias for the stock market to do just the opposite than the Bradley). Because stock markets & commodities are both controlled by liquidity (equities are leading, commodities are following), the Bradley is getting an indicator for global liquidity in times like that and thus a indicates the direction of the stock markets in general.

How can it be explained that the market is paying more attention to astrological constellations in bad times than in good times (this observation is not restricted to the Bradley)? The more people are governed by fear (of financial losses, of economic troubles, unemployment and so on), the lower their consciousness and the lower their consciousness, the more they are governed by the constellations. In contrast, the more you are centered in yourself, the higher your consciousness and the more independent and free your become, which is true both for individuals and the crowd. Enlightenment can be defined as the theoretical condition (that may not be reached in reality) where you are only connected the divine and no longer with the stars.

In that article I want to introduce the Bradley siderograph for 2008 and discuss this darling of many readers. By the way, the search engine Google lists amanita.at as the 1^st entry for the search term "bradley siderograph".

Most Bradley "fans" do overestimate the importance of the siderograph, in the Amanita forecasts it accounts for hardly more than 5%, there are a great many other (more) important factors. First I'd like to review the results of the standard model for the year 2007 (in the protected Amanita area there are more models and analyses on that topic):

(the charts are based on the calculations of the software „Market Trader" by Alphee Lavoie)

Since May we have had only 2 reversals:

1. 6/14/07: the benchmark S&P 500 (SPX) did set an intermediate-term top in the standard window +/- 4 calendar days (top intraday 6/15/07, close 6/18/07)
   ---
2. 8/26/07: Neither in the normal (+/- 4 days) nor in the extended window (+/- 1 week) did we set a trend change of intermediate-term magnitude (= a high or low for 1-3 months), not even a good short-term reversal for weeks. Now I have to explain an interpretation concept that was first introduced in 2006 in the protected area: the big Bradley turns (1-4 each year, printed bold in the chart) often have to be interpreted with a larger time window, namely up to +/- 2 weeks. The most important low of 2007 was Thursday 8/16/07 which was 6-7 trading days away from Sunday 8/26/07. The siderograph is not sufficient as a stand-alone timing instrument for at least 2 reasons:
   (a) You need additional methods because the siderograph is only able to predict a part of the major stock market reversals (about half or less)
   (b) Even if the Bradley nails a stock market high or low, you still need further precision tools because a window of some weeks is simply too large. Claims of some Bradley "hobby analysts" that the turns are consistently correct to the day (= the Bradley date is also the peak or low date) are definitely wrong. Only mini-sized trends (lasting for days) are consistently reversed within a +/- 1 day window of the Bradley dates.

A good example to demonstrate the sometimes very large windows is the siderograph turn date 5/20/06, the key high 2006 in the equity market was formed May 5-10^th which is (at least) 10 days from the Bradley date. Similarly, the Bradley point 11/28/06 roughly coincided with the key high of the 2^nd half of the year on 12/15/06 but was off by more than 2 weeks.

After analyzing the past and clarifying some important interpretation rules I'd like to present the 2008 model now, the chart below shows December 2007 through January 2009:

The 3 most important (bold printed) turning points are:

• 12/22/07

• 6/7/08

• 12/14/08

Other turning points are:

• 3/8/08-3/9/08

• 4/7/08

• 4/27/08

• 5/25/08

• 9/9/08, 9/20/08

Based on the work of Donald Bradley, Richard Schulz has developed the TAO indicator that - in contrast to the siderograph invented by Bradley - does predict the direction of the stock market (up or down) with a relatively high degree of statistical reliability (chance probability P<0.01%), i.e. not just turning points. A correlation of r=0.86 (r²=0.74) for the Russell 2000 is about the best you can expect from an isolated model.

The fact that the small cap index Russell 2000 is showing the best reaction doesn't surprise me. On the one hand, the small caps are less distorted by the PPT (Plunge Protection Team) market manipulation that is primarily acting over the SP futures. On the other hand, the small caps are a preferred playground for small investors and thus react more to the changes in crowd sentiment; the big boys are often not allowed to be exposed much to the small issues. Now the explanations by Richard Schulz, somewhat modified by myself:

Stock prices and astrological aspects are two independent variables, which qualifies them for a Pearson product-moment correlation study. I have modified Bradley's work:

1) I use a specific, uniform aspectarian for each planetary aspect. It doesn't matter whether it's Venus trine Saturn or Mars trine Jupiter- both trines receive the same positive value. The aspectarian is based upon a scientific formulation which sums harmonics and creates a zeroed oscillator.

2) I weight the various planetary aspects with a factor between 0.00 and 1.00. Mars in aspect to Jupiter has much more quantitative influence than Venus in aspect to Pluto. These weights are also specific and quantified.

3) I mathematically sum these weighted totals twice each week.

4) Then I take these biweekly totals and apply mathematical moving averages and summations to them. The major end result is the TAO (technical aspect oscillator). As a psychological indicator, when the TAO is positive and rising, people's attitudes and outlook are the most optimistic. When the TAO is negative and falling, the attitudes are the most negative.

5) I then correlate the stock indices with the TAO using the Pearson Product-Moment Correlation test. After this test is run, the result is the Pearson r, a value between -1.00 and +1.00. If the Pearson r is close to +1.00, then there is nearly a perfect correlation between the two variables. I get the following correlations for the TAO average (N=692): Dow Jones r=+0.73; S&P 500 r=+0.69, and Russell 2000 r=+0.86 (the quick TAO is considerably weaker). These Pearson r values are highly significant to a greater than 1 in 10,000 chance against being a random distribution. This demonstrates, mathematically, that it is very likely that the TAO is interrelated with stock market prices.

It is to be noted, however, that correlation is not causation. The correlation between the TAO and equity prices is just one useful observation among many. The stock market has, can and will move against the TAO for extended periods of time.

Current interpretation: the TAO is bullish from September through November 2007.

The Bradley siderograph was discussed the last time on October 5, 2006, so it's time to dedicate an article to this "darling" of many readers. To make my arguments understandable, I have to start with some general or even long-term explanations first. Let me stress that since 2006 I am probably THE biggest stock market bull on the planet. To my knowledge, no one else has given similar (let alone higher) price targets (that means in nominal terms). This radical re-orientation was caused by a research project in 2006 that examined the astrological and cyclical signifiers of inflations and state bankruptcies of the past 2,000 years.

The super-bullish market environment since 2006 was anticipated by few and during the quick correction in March 2007 even dyed-in-the-wool bulls threw the towel and prepared for a "healthy consolidation". Needless to say that since March the indices have been skyrocketing. Admittedly, the bears have excellent fundamental reasons why the equity markets should turn down soon. According to the official numbers, the US economy has grown by only 1.3% in the 1st quarter 2007, in the preceding 3 quarters not much more than 2%. Of course, you have to take into consideration that the official US statistics are massaged heavily, according to the top analyst John Williams (link) the GDP growth rate is overstated by more than 4%, so you know that the US economy has been in a recession since the 2nd quarter 2006, since early 2007 even in a very serious recession.

Without doubt, the widely discussed collapse of the US housing markets contributes a lot since almost all of the "honest" growth of the largest economy of the world has been contributed by the housing sector. The ongoing recession is no surprise, I warned already of the Jupiter-Pluto depression cycle due in 2007/8 as soon as 2005 (article). This cycle has worked like a clockwork since the 19th century and always indicated very hard times for the economy. With high odds, 2006 was the onset of a depression in the US, e.g. some (less manipulated) leading indicators are in a worse shape now than in 2000 before the serious recession (and stock market crash) began.

Frequently, fundamentalists also look at the technical situation (especially sentiment) where some of them make a catastrophic mistake in their assessment that can be summarized as follows: "all others are bullish, I am the only 'lonesome' bear that remains rational". Even worse, sometimes top forecasters with an admirable track record are counted as "contrary indicators", and in the end these claims are marketed as an "analysis".

This fairly typical error in sentiment analysis is explained in the article "A criticism of the contrary approach". We must accept that the human perception is so error-prone that you have to rely primarily (or even exclusively) on quantitative-statistical indicators. Anecdotal stories (e.g. "forecaster X has turned bullish, so the markets must go down soon") are nice to read and often quite revealing but they must never be the focus of sentiment analysis.

I track a large number of statistical sentiment indicators and have to state that I have almost never seen a monster rally being accompanied by such a pervasive skepticism. Sentiment indicators have to be always interpreted in the light of the market action, and if a series of new all-time highs in the Dow Jones isn't able to turn the crowd bullish, then I don't know what will do it... At any rate, such an extra-ordinary set-up is very reliable, it's really a textbook example. I'd like to present a small sample of sentiment indicators used (link):

• The popular AAII indicator (American Association of Individual Investors) has the number of bulls close to the lowest number in 15 years.
  ---

• The odd lot short ratio at a 13-year high tells us that the dumb money is shorting as if tomorrow would be the end of the world. Odd lots of less than 100 shares are a very reliable sign of small speculators and so this ratio works as an excellent contrary indicator historically. In comparison, during the 2000-2003 bear market the odd lotters were primarily found among the buyers.
  ---

• The public to NYSE specialist short sales ratio at an all-time high is confirming that the dumb money is shorting as hell while the smart money (specialists) is accumulating.
  ---

• Calls are traded at a rather low premium compared to puts.
  ---

• The number of CBOE puts traded in March was as high as never before in history, not even after 9/11 did we see such a huge demand for puts.
  ---

• The latest Gallup polls show a rapidly deteriorating investor sentiment despite the ongoing explosion of stock prices; in 2000 the indicator was at circa 180, now it's only about 70.
  ---

• And last but not least, Mark Hulbert from "Hulbert Financial Digest" detects a very large number of bears among the editors of market letters (Link).

I could continue that list but I think my point is clear: this monster-sized rally is approached with an incredible skepticism, which of course is super-bullish from a contrary perspective. Still one should take into consideration that the notion of a "powerful bull market" is only true using nominal prices not adjusted for inflation. Using the true (unmanipulated) inflation numbers, the benchmark S&P 500 is in a bear market since January 2004, which is hardly noticed by anyone.

Now the 1-million-dollar question is how to reconcile the super-bearish fundamentals and the super-bullish sentiment indicators? I am convinced that the real reason is that we are in the early stages of a hyperinflationary depression. According to the "Privateer" edition #576, the Fed is already buying almost 20% of the US government bonds, a typical sign of a hyperinflation. Astute market observers recognize that the world financial system is on a journey with a "one way ticket only", we get deeper into irreversible territory every year.

The key to understand the situation is that in a hyperinflationary depression the stock indices rise faster the bigger (!!!) the economic problems are - because liquidity has to be created to "fix" the problem (presumably also over the offshore/ Caribbean activities of the Plunge Protection Team). To my knowledge, this paradox logic has been recognized by almost no one so far, let alone applied to forecast the markets. That's why the ongoing Jupiter-Pluto depression sends stocks higher almost in an "up-crash". Discussions with leading researchers in alternative medicine and healing have been offered me an explanation for the "global switching", i.e. what was formerly bullish is now bearish and vice versa... I am working on an article connecting many different areas, it appears that electromagnetic radiation plays a key role in this process.

So what has this got to do with the Bradley siderograph? Due to the hyperinflationary forces, the market has seen  only one intermediate-term or longer-term trend in the past year: UP, of course. In almost 12 months there was not a single correction that lasted much longer than 1 week. Something like that has almost never happened before in the history of the markets. Without doubt, these are symptoms of hyperinflation stage #2 where the powers that be are still successful in driving the excess liquidity into the "right" direction, i.e. to the stock markets (and partially also bond markets) which are driven higher and higher and higher and... The actual inflation calculated with the formula "monetary expansion minus GDP growth (or shrinking in this case)" already yields 2-digit inflation numbers for the US (somewhat lower in Europe). Then, in the final stage #3 the markets get out of control and rise almost daily. A historical example: at the end of the German hyperinflation in 1923 money bills were used to heat apartments simply because their monetary value was zero (50-billion mark notes were issued).

Unfortunately, as long as the stock markets don't return to the normal mode which they left in 2006 (as my statistical calculations show), intermediate-term market timing as supported by the Bradley model is of limited or no meaning whatsoever. The turning dates of the siderograph have to be interpreted with a standard window of +/- 4 calendar days and sometimes you have to give or take a full week, so it's clear that the Bradley can only be applied successfully for intermediate-term trends of 1 month or longer. In the current situation, trading is only meaningful on an intraday basis or very short-term (2-3 trading days) which requires intraday entry & exit signals - that can never be delivered by the siderograph.

Stage #2 of the hyperinflation has a very important impact on the markets: the totalitarian-fascist gleichschaltung (synchronization) on the political level emerging mainly from the US can also be detected  in the markets that more and more dance to the rhythm of liquidity, the individual market cycles are waning. The good thing about that is that since 2006 the Bradley has started to work better (!) for the other markets than for the stock indices. I believe the reason is that the other markets are still "normal" and not "crazy" and therefore have good trends in both directions, not just up.

Analysis of the past:

(1) Bradley turning point 11/28/06: this was the most powerful turning point in 2006, yet it didn't produce anything in the S&P 500, the benchmark for the Western stock markets. In contrast, the other 5 markets (precious metals, oil, currencies against US$, bonds, grains) all set intermediate-term tops holding for months in the days after 11/28/06.

(2) December 06 - February 07: no meaningful reversal in the siderograph, so the standard model was of no use whatsoever for a pretty long period of 3 months. This is a reminder that the Bradley can only be an assistant in market timing but does not perform a full job.

(3) Bradley turning points 3/5/07 & 3/14/07 (since these 2 dates are just 9 days apart and overlap, they are taken together): the SPX closing low 07 was exactly on 3/5/07; however, the only meaningful top so far this year (2/20/07) did not show up in the standard model (it was present in the other models though - which are reserved for Amanita subscribers). In precious metals & grains we had intermediate-term tops for months on 2/26-27 (deviation: 6-7 days), in the debt market there was a double top on March 7th & 13th (deviation: 1-2 days), and in JPY/USD a multi-month high was printed precisely on 3/5/07.

(4) Bradley turning point 4/20/07: in the stock indices no reaction whatsoever but in oil an 8-months high and in the euro a 2-year high on 4/27/07, in wheat a 4-months high on 4/26/07 (deviation: 6-7 days); on 4/16/07 a 2-months bottom in bonds (deviation: 4 days) and on 4/20 an intermediate-term precious metals top

(5) Bradley turning point 5/4/07: it's too early to make a reliable judgment but precious metals & currencies could form an intermediate-term bottom and stocks, bonds & oil an intermediate-term top (grains are mixed)

For 2007 only 4 key dates are left, please note that the amplitudes in the siderograph are larger for these turns, so they are more important on average (bold in the chart) then the dates of the first 5 months 2007 discussed above:

(1) 6/14/07

(2) 8/26/07

(3) 10/17/07

(4) 12/22/07