Recent "window "dressing" and end of quarter action commentary from the Astrikos Tuesday and Thursday editions points to the possibility of some type of I/T top 6/23 - 6/24. Alot depends on today's action i think if deciding to sell the close...or look for more up early next week.
Regardless, a lot of stuff is adding up for a signifigant decline to begin very shortly, and possibly last2-3 weeks minimum.
Eliades new 10 Trin has crossed sheply below the .80 level, breadth has taken a considerable turn for the worse, VIX 10-21 crossover sell signal in effect so far, and cycles are in play here as well...etc etc.
...a potential market top could form in the 1-2 days
following triple witching. The reasoning is that the Tuesday following the
upcoming option expiration kicks off the last five days of the quarter, which
in the past has been a positive time for stocks due to so-called 'window
dressing'. Take a look at the S&Ps performance in the last five days of each
quarter between the beginning of 1999 and end of 2001...
Window Dressing Effect Then
12/21/01 - 12/31/01... S&P +0.3%
09/21/01 - 09/28/01... S&P +7.8%
06/22/01 - 06/29/01... S&P -0.1%
03/23/01 - 03/30/01... S&P +1.8%
12/21/00 - 12/29/00... S&P +3.6%
09/22/00 - 09/29/00... S&P -0.8%
06/23/00 - 06/30/00... S&P +0.9%
03/24/00 - 03/31/00... S&P -1.9%
12/23/99 - 12/31/99... S&P +0.7%
09/23/99 - 09/30/99... S&P +0.2%
06/2 /99 - 06/30/99... S&P +3.0%
03/24/99 - 03/31/99... S&P +1.4%
Since 2001, however, the market's really struggled during this supposedly
positive time frame for stocks, with the S&P500 falling in 4 of the past 5
cases...
Window Dressing Effect Now
06/24/03 - 06/30/03... S&P ???
03/24/03 - 03/31/03... S&P -1.9%
12/23/02 - 12/31/02... S&P -2.0%
09/23/02 - 09/30/02... S&P -2.2%
06/21/02 - 06/28/02... S&P +0.1%
03/21/02 - 03/28/02... S&P -0.5%
Given Monday's big gains, it appears the window dressing phenomenon has
already kicked into gear, which will most likely leave little buying power
left for the final week of June.
...And this from last nights commentary
Our research shows that the S&P has a tendency to rally in the 1-2 days immediately following triple witching. For example, there have been a total of 29 triple witching expirations since the beginning of 1996. In 24 of those 29 cases, or 83% of the time, the SPY (S&P500 tracking stock) posted a higher close within two days. All of these occurrences are listed in the table below...
Triple Witching & the SPY
06/20/03... SPY ???
03/21/03... SPY -2.15 two days later
12/20/02... SPY +0.03 one day later
09/20/02... SPY -2.04 two days later
06/21/02... SPY +0.52 one day later *
03/15/02... SPY +0.02 one day later
12/21/01... SPY +0.41 two days later
09/21/01... SPY +3.42 one day later *
06/15/01... SPY -0.06 two days later *
03/16/01... SPY +2.34 one day later *
12/15/00... SPY +1.75 one day later *
09/15/00... SPY -0.03 two days later *
06/16/00... SPY +1.88 one day later *
03/17/00... SPY +2.25 two days later
12/17/99... SPY +1.12 two days later
09/17/99... SPY -3.00 two days later
06/18/99... SPY +0.31 one day later
03/19/99... SPY +0.25 one day later *
12/18/98... SPY +1.66 one day later
09/18/98... SPY +0.78 two days later
06/19/98... SPY +0.53 one day later *
03/20/98... SPY +0.68 two days later
12/19/97... SPY +0.61 one day later *
09/19/97... SPY +0.53 one day later *
06/20/97... SPY +0.04 two days later *
03/21/97... SPY +1.09 one day later
12/20/96... SPY +0.35 two days later *
09/20/96... SPY +0.03 two days later *
06/21/96... SPY +0.13 one day later
03/15/96... SPY +1.23 one day later *
Note that I've put an asterisk (*) after those instances in which the SPY closed lower on 'triple witching Friday'. Interestingly, those times when this setup was clearly wrong occurred when the SPY closed higher on triple witching. If the SPY closes lower, odds are significantly better that we'll see a higher close Monday or Tuesday of next week. Of the fifteen times the SPY has closed lower on triple witching, it's proceeded to post a higher close in every case but two, racking up gains of over 13 points and only losing 0.09 (virtually nothing) on the two instances in which it was wrong.
used with permission
astrikos.com |
