SELL S9/S12 COMBINATION
                           (C) 2010  www.tigersoft.com

      
The S9/S12 signals in combination can be very bearish.  They occurred right at the top in
            September 1929 and October 1987.  For there to be this much bearishness potential, we
            think that there has to be a lengthy period of a comparably weak NYSE A/D Line and
            multiple earlier Peerless Sell Signals.  

            S9/S12s in bull markets are more reliably bearish than S9/S12s in on-going
            bear markets. As a whole, they usually only promise a decline to the lower band
            or well-tested support.   Only 3 of the 15 bull market S9/S12s brought a decline
            of greater than 6.5% when reversed by the next Peerless Buy signal.  4 of the 12
            bear market S9/S12s brought this sized decline. However, 5 of the 12 bear market
            S9/S12s resulted in losses.  The S9/S12s are least reliable at a bear market bottom
            that forms quickly.

                   
BULL MARKETS S9/S12s
        
Number of closed out trades =  15
            Gains = 14 Losses = 1
           
Average Gain per trade = +11.4%
            There were 3 Gains of more than 30%.
            Leaving these trades off, the avg. gain was 5.1%

              BEAR MARKETS S9/S12s
               Number of trades =  11  Losses = 5
           
Average Gain per trade = +4.6%
            There were 2 Gains of more than 20%.
            There were 3 Losses of more than 5%


                                                          CLINCHED SELL S9s

            These signals occur on strength.  It takes some resolve and faith in Peerless to use
            them.  To be surer of an S9/S12 when the DJI is at or near an all-time high, when
            resistance is reduced, wait for NYSE declines to exceed advances by 50 or more
            on any day at the close after the Sell S9 or on the day of the Sell S9 for the signal
            to be "clinched" .   This would have avoided two S9 losses, one in 1987 and one in
            1999.

                                                             1929 CRASH
wpe1A1.jpg (48996 bytes)

            The more S9s and S12s on the chart, the more bearish is the eventual breaking
            of support.                                                                       
                                          
                            The Third S9/S12 Set Brought The Crash of October 1987

wpe1A0.jpg (52065 bytes)

                                          EFFECTIVE SUPPORT IN LIMITED DECLINES
                                                         FOLLOWING A SELL S9/S12

            The lower band sometimes represents how far down a decline will go following
            an S9/S12.  Below that, an S9/S12 often can be reversed at a line drawn through
            at least 3 low points or closings, especially when they are 8.5% to 12% below the
            previous peak.

            In a Presidential Election Year, it is more likely that the lower band and well-tested
            support will hold.   See this in the case of 1972 below here.  But this rule is not
            absolute.  2008 was a clear exception.  From this it may be inferred that when support
            IS broken in a Presidential Election Year, a much steeper decline is likely.  The Sell S9s
            in 2008 had their bearish effect merely postponed when the P-Indicator turned positive
            for more than 3 days.  
                                                             2007 - DJIA
wpe5E03.jpg (62316 bytes)

            In the tables below, we only consider the simultaneous S9s and S12s, when they
            occur on the same day.        


                                                    POSTPONED DECLINES AFTER
                     P-INDICATOR AND ACCUM. INDEX TURN BRIEFLY POSITIVE


            When the P-Indicator and Accumulation Index turn positive for 3 or more days
            after a Sell S9/Sell S12,  there is commonly an extension of the rally for a brief period
            before the   bearishness of the signals' first experience is borne out.  A 3% to 5%
            additional DJI rally is typical that lasts 3 to 6 weeks.  The 1959 example shows
            this.  About half the S9/S12s have the key internal strength indicators turn positive.
                                                                   Sell S9/S12
DAERA59.BMP (1068054 bytes)

                                                      BULL MARKET S9/S12s
             

             It is helpful to study these combined S9/S12 signals when they occurred in bull markets,
            before there has been a 12% sell-off.  Since 1928, there have been 15 combined Sell S9-S12
            signals in such a bull market environment.    Assuming one shorted the DJIA on these
            S9/S12s, 14 of the 15 were closed out profitably at the time of the next Buy.   Leaving
            the three biggest gainers off, the average drop would still have been +5.1% on short
            sales taken with these signals.  In 9 of the 15 cases, there was no paper loss.  In one
            case, there was a paper loss of more than 15%.  In all the others, the paper loss was less 
            than 3%.  

            The one loss,  in 1987, would have been avoided if we also required that the NYSE
            declines exceed advances at the close either on the day of the signal or on a subsequent
            day before there is a Peerless Buy.  Today  there were 555 more down than up at the
            close.  So, the S9 is now "clinched". 

            In 40% of  these S9/S12 cases, the market rallied 3% or more before falling to at
            least the lower band.    What typically distingushes these cases is that the P-Indicator went
            positive for more than 3 consecutive days.  In these cases,  the decline to the lower
            band is usually just being postponed for 3-6 weeks.

              Bull Market S9/S12s
                     =================== 

        4-Year Cycle
 
         Presidential Election Year (PYE)            
           1962 + 1.2%  

         PYE + 1 (1929, 1933, 1937, 1941...)
           1929  +3.6%  
           1929   +5.0%  
           1929   +40.0%
           1981  +3.5%

         PYE + 2 (1930, 1934, 1938, 1942...)
           1978  +11.7%
           2010  ?
         PYE + 3 (1931, 1935, 1939, 1943...)
           1959 +1.9%  
            1975 +4.8%
            1987 +5.2%             
            1987 +5.2%
            1987  -4.0%
            1987 +30.3%
            1999 +6.4% 
 Simultaneous Peerless Sell S12s/S9s: 1928-2009
                Date          DJI           Gain
-------------                 -----------   --------
#1
...S9...S12    1/ 30/ 1929    312.6          +3.6%
1.023      .492    P=-22      -22      -.006  -415   .189
(la/ma  ann.roc    P-Ind  P-change    IP21   V-I   Opct)
Bull Market variety.
Fell to lower band after 3% advance and paper loss.
P-Indicator and Accum.Index stayed negative until DJI fall.
                              Sell S9/S12
wpe1A1.jpg (67984 bytes)
-----------------------------------------------------------------------
#2
...S9...S12    4/ 22/ 1929    315.3          +5.0%
1.036   .027  P=-39    17    .038   -576   -.016
Bull Market variety.
Fell to below the lower band and support after 3% advance and paper loss.
P-Indicator and Accum.Index turned positive for 2 days until DJI fell
to support at 300.
                                   Sell S9/S12  
                                  
Sell DATW2829.BMP (1075254 bytes) 
The chart above shows that the full daily data, where a daily high
ans low were kept, did not start until October 1928.  This data
is needed to produce a Sell S12. This makes the October 1928 S12 
unreliable. It is not counted here.
                                          
 
-----------------------------------------------------------------------
#3
...S9...S12    6/ 14/ 1929    313.7          +40%  
1.021   -.221 P=-59    11    -.035   -542   .162
Bull Market variety.  16% paper loss.
P-Indicator turned positive for more than 3 days with the
DJI at 331.70 and then saw the DJI rally to 381.20 three
months later before crashing back to 198.70 on 11/13/1929.
The speculative bubble was so great then that even when
the internal strength indicators did turn negative again,
the DJI continued to rally for two more months.  The falseness
of this rally is clear from hindsight.  At the time, it would
have seemed that the indicators were wrong and were not
predictive of a decline.
                      Sell S9/S12
wpe5DFF.jpg (67149 bytes) 
-----------------------------------------------------------------------
#4
...S9...S12    9/ 4/ 1929     379.6         181/380 or +47%
1.043    .889  P=-16   -2    -.059   -370   .133
Bull Market variety.
P-Indicator and Accum. Index stayed negative.  DJI broke support at 320
and then crashed to 198.70 on 11/13/1929.  No paper loss.
See how the support at 325 appeared at first to hold.  But the
rally that followed was weak.  The internal strength indicators
kept falling and the rally stalled exactly at the resistance of
the 65-day ma.  The rally proved to be a right shoulder in a 4-month
long head and shoulders pattern.
                                    Sell S9/S12
wpe5E00.jpg (67109 bytes)
-------------------------------------------------------------------- 
#5
...S9...S12    4/ 2/ 1936     160.4          +7.4%
1.026   .282  P=-27   12    -.089  -231     .201
Bull Market variety. No paper loss.
The P-Indicator and Accum. Index never turned positive
before the DJI caved in, breaking the lower band and the 
three times-tested 150 support. It fell to the 12% level
below its previous peak. 
(...S9...S12    4/ 6/ 1936     161.9          +8.2%
1.034   .236   -37   -1    -.152  -249     .085)
                                         Sell S9/S12
wpe1A2.jpg (63318 bytes)
-------------------------------------------------------------------
#6
...S9...S12    7/ 1/ 1959     650.2          +1.9%
1.03   .24  P=-2   38  -.062   -177  .166
Bull Market variety.  4% advance and paper loss.
The P-Indicator and AI turned positive for more than a few days.
This typically brings 2%-5% rally for a month to six weeks and 
then a decline to at least the lower band. This is what happened. 
The DJI peaked at 678. Its decline took it 9% down from its peak 
to the earlier 615-620 support.
                           Sell S9/S12
wpe5E01.jpg (65999 bytes)

----------------------------------------------------------------------
#7
...S9...S12    5/ 22/ 1972    965.31         +5.6% 
1.023  .019  P=-89   12  -.031  -2  .127
Bull Market Sell S9/S12 variety. No paper loss.
Declined directly only to lower band. A limited decline 
to the lower band or well-tested support is more likely after a 
Sell S9/S12 in a Presidential Election year.

----------------------------------------------------------------------
#8
...S9...S12    8/ 3/ 1972     947.7          +2.7%
1.022  .103  P=-82   -3  -.056   -2  -.131
Bull Market Sell S9/S12 variety.3% advance and paper loss.
Declined half way to lower band from 21-day ma. In this case,
the decline was postponed by a month when the internal 
strength indicators turned positive more than 3 days.

                        S9/S12               S9/S12
wpe5E02.jpg (70013 bytes)
----------------------------------------------------------------------
#9
...S9...S12    8/ 29/ 1975    835.34         +4.8%
1.024    .055 P=-139    23   -.102  -3  .008
Bull Market Sell S9/S12 variety. No paper loss.
DJI Fell directly to the lower band and well-tested support 11%
below the previous peak. The internal strength indicators
did not turn positive. 
                                     Sell S9/S12
wpe5E05.jpg (65600 bytes)
 
-----------------------------------------------------------------------
#10
...S9...S12    10/ 10/ 1978   891.63        +11.7%
1.02     -.219   P=-139  -19   -.136  -5  -184
Bull Market Sell S9/S12 variety.  No paper loss.
Fell directly below the lower band to well-tested support.
(...S9...S12    10/ 11/ 1978   901.42        +12.6%)
 wpe5E07.jpg (69549 bytes)
-----------------------------------------------------------------------
#11
...S9...S12    1/ 2/ 1981     972.78         +3.5%
1.029    -.264   P=-71   30   -.04  -5  -.017
Bull Market Sell S9/S12 variety. No paper loss.
Fell directly to the lower band.
-----------------------------------------------------------------------
#12
...S9...S12    5/ 6/ 1987    2342.19         +5.2%
1.021     -.328  P=-185  -9  -.079  -15  -.15
Bull Market Sell S9/S12 variety. No paper loss.
Fell directly to the lower band.
-----------------------------------------------------------------------
#13
...S9...S12    6/ 8/ 1987    2351.64                -4.0% LOSS
1.026     -.088  P=-32   23 -.045  -4   -.18
Bull Market Sell S9/S12 variety.  4% advance and paper loss.
This signal was never clinched as P-Indicator turned positive.

-----------------------------------------------------------------------
#14
...S9...S12    10/ 2/ 1987   2640.99        +30.3%
1.026   .18   P=-56    41  -.017  -7   .068
Bull Market Sell S9/S12 variety.  No paper loss.
This signal produced a decline below support that dropped extraordinarily.
----------------------------------------------------------------------
#15
...S9...S12    8/ 17/ 1999   11117.07        +6.4%
1.025  -.077   P=-425  58  -.025  -119  -.126 
Bull Market Sell S9/S12 variety. No paper loss. 
This signal produced a direct 12% decline below peak.
wpe5E06.jpg (72366 bytes)
----------------------------------------------------------------------
#16
...S9...S12    6/16/2010  OPEN
1.025  -.253  P=12    2   -.018   -136   -.206
Bull Market Sell S9/S12 variety.
P-Indicator turned slightly positive. 
A/D Line haa already broke above well-tesed downtrendline.
==========================================================
  Summation:    
          Bull Market S9/S12s
         Number of closed out trades =  15 
         Gains = 14 Losses = 1
         Average Gain per trade = +11.4% 
         There were 3 Gains of more than 30%.
         Leaving these trades off, the avg. gain was 5.1% 

 
                                                  


                      BEAR MARKET S9/S12s

wpe1A3.jpg (68689 bytes)
      
            Bear market S9/S12s were generally not as powerfully bearish as bull market S9/S12s.
            The average gain is only 4.6% and shorting the DJIA on these signals were
            unprofitable in 5 of the 12 cases.   Look for other signs of weakness to distinguish
            the most bearish situations.  In 1974, you can see steadily (red) negative readings
            from the P-Indicator and the Accumulation Index, a head and shoulders pattern
            in March 2002 and lower and lower price highs and lows in mid 1974 before the
            breakdown in August.

             The 1936 examples show how S9/S12s lose their potency as a decline unfolds.
              Breaks in the well-tested downtrendlines of price and the A/D Line should be
              considered judged Buys.  Originally, they were B6/B7s, back in the 1981 version
              of Peerless.

            
Presidential Election Year (PYE)
          
1936  +7.4%
         PYE + 1 (1929, 1933, 1937, 1941...)
           2001 
-5.1%

         PYE + 2 (1930, 1934, 1938, 1942...)
          
1934  7.0%
           1934 
-8.0%
          1942 -5.2%
           1958  -2.4%
           1962 
+1.2%
           1966 
+6.9%
           1970 
+0.6%
           1974 
+27.3%
           1974 
+27.1%
           2002 
-0.2%

         PYE + 3 (1931, 1935, 1939, 1943...)
         
Table S12-D     Simultaneous Peerless Sell S12s/S9s: 1928-2009
                Date          DJI           Gain
-------------                 -----------   --------
#1
...S9...S12    8/ 22/ 1934     94.3          +7.1%
1.047    .437   72   41    .062   -8   .117
Bear Market variety.
P-Indicator was positive.  This signal properly cancels errant B12 
three days later even though P-Indicator is positive.
DJI fell below the lower band to 3x tested support.
--------------------------------------------------------------------
#2
...S9...S12    9/ 27/ 1934     93.4                -8.0% Big Loss.
1.029    -.105 -40   20    -.036   -80    -.19
Bear Market variety.
There is a A/D flat-topped price breakout at 97.60 on 11/7/1934 which
should have been used to cover any short sales. Loss then would be 4.5%
See how price channel and downtrending A/D Line trends are broken
just before advance.  This shows how S9/S12s lose their potency.
DATA1934.BMP (1084854 bytes)
-----------------------------------------------------------------------
#3
...S9...S12    5/ 11/ 1942     99.2                -5.2%  LOSS
1.027   -.062  -42   9     -.162  -53      -.028
Bear Market variety Sell S9/S12.
Break in A/D Line downtrend on 5/22/2010 at 99.20 -- No loss. 
DATA1942.BMP (1080054 bytes) 
----------------------------------------------------------------------
#4
...S9...S12    1/ 7/ 1958     447.8                -2.4% LOSS
1.027   -.049   -3   15  -.051  -192  .054
Bear Market variety Sell S9/S12.
A/D Line makes a new recovery high on 1/16/1958 with DJI at 445.20
P-Indicator turns positive a few days later, which typically
brings 2%-3$ rally and then decline.  This is what happened.
----------------------------------------------------------------------
#5
...S9...S12    7/ 10/ 1962    586            +1.2%
1.034  -.327  -451  -103   -.113  -139   -.147
Bear Market variety Sell S9/S12.
P-Indicator turns positive a few days later, which typically
brings 2%-3$ rally and then decline.  This is what happened.
----------------------------------------------------------------------
#6
...S9...S12    9/ 15/ 1966    814.3          +6.9%
1.029   -.143  -174  50  -.078  -13  -.146
Bear Market variety Sell S9/S12.
Immediate decline slightly below lower band.
----------------------------------------------------------------------
#7
...S9...S12    7/ 15/ 1970    711.66         +0.6%
1.023   .416   -87    16   -.016   -2   .193
Bear Market variety Sell S9/S12.
P-Indicator turns positive a few days later, which typically
brings 2%-3$ rally and then decline. This is what happened. 
----------------------------------------------------------------------
#8
...S9...S12    2/ 22/ 1974    855.99        +27.3% 
                                     Reversed at 621.95 on 9/27/1974 by B8
1.027  -.214  21   7   -.083  -2  -.08
Bear Market variety Sell S9/S12.
P-Indicator did turn positive and DJI rallied to 890 before zig-zagging lower.
-----------------------------------------------------------------------
#9
...S9...S12    6/ 7/ 1974     853.72        +27.1% 
1.033    .039  -77  39   -.074  -2  .03
Bear Market variety Sell S9/S12.
Btoke down far below lower band after two failed rallies. 
-----------------------------------------------------------------------
#10
...S9...S12   4/ 10/ 2002   10102.74             -5.1%  LOSS
1.031     -.128   -121  134  .053  -174    .095
                        ===
Bear Market Sell S9/S12 variety.
P-Indicator turned positive. A/D Line broke above well-tesed resistance 
on 4/18/2010 at 10615.83

----------------------------------------------------------------------
#11
...S9...S12    10/ 10/ 2001  9240.86            -0.2% LOSS
1.032     -1.049  -212  96  .053   -245   -.258
Bear Market Sell S9/S12 variety.
P-Indicator turned positive. 
A/D Line had already broke above well-tesed downtrendline.
DATA0102.BMP (1075254 bytes)
-----------------------------------------------------------------------
==========================================================
         Number of trades =  11  Losses = 5
              Average Gain per trade = +4.6% 
              There were 2 Gains of more than 20%. 
              There were 3 Losses of more than 5%