Saturday, January 19, 2008

The big picture ! - LT sell

Prior e-wave projection

Back in Sep 2006, on this blog, i had posted the e-wave projection for the final bull market top as SPX 1620. But given the 3-legged structure from Aug 07 bottom, i was still giving the benefit of doubt for another run-up to SPX 1620 in 2008, to complete the pattern. I was wrong. My LT system issued a sell on Jan 16, which now precludes any run-up to new highs anytime soon. The e-wave failure or truncation resulted in a bull market top at 1576. I missed my projection by about 40 points, which is insignificant for a LT type of projection. It's not about being right or wrong here. Even if i were dead right on my target, i would still have got a LT sell only about 10-15% down from the top. You don't get LT bearish confirmation at the tops.

LT sell





As i posted here on Jan 16 08, my system went to a long term sell after breaking the Aug 07 lows. Some recent history on this LT system.

Nov 2000 - SPX 1365 - LT sell. SPX declined to 768 subsequently.
June 2003 - SPX 988 - LT buy. SPX ran up to 1576 after this LT buy. About 60% run-up.
Jan 16, 08 - SPX 1370 - LT sell.

Now since the last LT buy back in 2003, SPX has had one heck of a bull run of about 60%. Now the LT sell has been generated at about 12% from the top. So giving up 12% of the profits from a 60% run-up is not a bad deal, if one views it from a LT perspective. The LT sell during 2000 was also generated from about 11% from the top.

If one were a ST trader, my ST sell would have gotten one out of the market right near the top on Oct 11 - ST sell posted by me on TT.

http://www.traders-talk.com/mb2/index.php?showtopic=77295&hl=

If one were a IT term trader, my IT sell would have gotten one out of the market at 1480 on Nov 11 - IT sell posted on traders-talk

http://www.traders-talk.com/mb2/index.php?showtopic=78887&hl=

So bottomline, it depends on the timeframe one is trading and how much drawdown one is willing to take. LT signals is of no use to traders. It's mostly for LT investor types.

This LT signal is based on three factors. All three conditions need to be satisfied.

1) My bull-bear indicator moving above zero for LT buy or moving below zero for LT sell.
2) Lower lows on weekly for a LT sell and higher highs on weekly for a LT buy.
3) A potentially completed e-wave pattern.

Amazingly, all the three factors came into play right about at the same time. That's what makes TA fascinating !


E-wave details






I was working with an assumption that we were in a Primary degree wave B from March 03 bottom. Given that i now have a LT sell and a potentially completed wave pattern, i have to now radically alter my wavecount, in the light of newly presented information. Waves are dynamic structures, which evolve over time. Now there is no way, i can call this structure a A-B-C pattern from the Oct 02 lows, without violating all the channel rules and e-wave time rules and compromising on the structural integrity of the wave pattern. I am not going to do that to justify my bias or to stubbornly prove my original thesis. Instead, i am changing my wavecount, that the wave from oct 02 was a Primary degree wave 1 impulse, with a wave 5 extension. So we in a primary degree wave 2 bear trend at this stage, until proven otherwise. The implication is that the Oct 02 bottom was a cycle degree wave 4bottom, which should not be violated for decades to come.

The first thrust (Wave 1) from the Oct 02 bottom generated a MACD reading of -14.3. The subsequent thrust (wave 3) generated a MACD reading of 39.58, which has never been exceed throughout the course of the bull market, even after the huge impulse from mid-2006. That pretty much confirms that 2004 peak was indeed the wave 3. Wave 5 generated a peak MACD reading of 37.27, which diverged negatively with the wave 3 peak, characteristic of wave 5. Now we have a situation where wave 1 < wave 3 < wave 5 which says wave 5 extended.

Price Projections Now typically, impulses once completed, retrace back to wave 4 of a lesser degree, which in this case comes around SPX 1170. The measured move objective for a potential double top at Oct 07 also generates an objective of 1170. So the preliminary price objective for the Primary degree wave 2 should be around SPX 1170 +/- 20 points. Whether further downside projections will be generated, will be determined if and when we get there.

Time Projections

Since there is no clear defined time relationship between wave 1 and wave 2, it's hard to come up with a time projection for wave 2. If it's a violent decline, we could see all this over in about 6-8 months. If it's a slow bleed with deep retracements, it could run all the way into 2010. I am not going to hazard a guess on that.


Signal Status

At this point all my systems from VST to LT remain on a sell. Remember in the context of this LT sell, we will have some large and vicious bear market rallies. But rallies should not make any new bull market highs, but a series of lower tops.

Implications of a LT sell

Now that we are in a bear market as opposed to a IT bear trend, this should lead to softening economic conditions going forward, and possibly a recession. Also, as long as the LT buy was alive, one could sit on drawdowns and hope for the LT trend to bail them out of the drawdown situations. One could also average down on IT term declines. Now that the LT is on a sell, drawdowns can lead to further drawdowns and averaging down can be a potentially losing strategy. Also now all signals will have to be intrepreted in the context of a bear market.

And again the "End of America" and the collapse of the Roman empire arguments will start to take centrestage as the bear market unfolds. The argument about emergence of China and India as the next economic superpowers will start to gain strength. The deflationary collapse and "Great depression" arguments will be accepted and dreaded. Those who hung out on the longwaves.net during the 2002 bottom know pretty well how convincing they look and how wrong they turn out to be. If anything this wave 2 will generate fear levels worse than the bottom of Oct 02. Those fear levels should present an opportunity of lifetime for the wholesalers to pick up the bargains once again, as has always happened in the history. But for the retailers with not that deep pockets, it's prudent to remain patient and let the bear play out and wait for the next LT buy confirmation.

Good luck.

Wednesday, January 16, 2008

LT sell



I will post a detailed update during the weekend. For now, it's suffice to say the break of the Aug lows and my bull-bear market indicator crossing below zero has signalled a LT sell.

Sunday, January 13, 2008

Oh my... !

The decline from July 07 top concluded as 3 waves. The next upleg from Aug 07 to Oct 07 was another 3 wave structure. The subsequent decline from Oct 07 is also a 3 wave structure so far. So no impusive action so far from July 07, from either a bullish or bearish perspective. That's why i give the benefit of doubt to a high level consolidation here.

The most striking feature is the irregular top on Oct 11. Irregular tops are bullish e-wave structures and they resolve upwards after the correction is over. The only way a 3-legged wave can be called the ultimate bull market top is if one subscribes to the truncation theory. If one theorizes that Oct 11 ended in a failure, then we could say that it was a major bull market top. But the truncation theory ends up wrong more than 9 out of 10 times. So it's hard to go with the truncation theory. A failure or truncation means the bearish forces were so overwhelming that the market could not reach it's logical conclusion i.e could not complete it's wave pattern. Now if it were true, then the subsequent decline from Oct 07 top should have been more violent than the rally from Aug 07 bottom. This is not the case as seen from the angle of decline or the time taken to retrace the the rally from Aug bottom. So the truncation theory is pretty hollow!

Now here's the "Oh my...." part.




My momentum indicator which demarcates the bull-bear market stopped right near the zero line as of this week's close. Oh my...

That also coincided with a climactic reversal in the form of penentration of the weekly bollinger bands and a reversal out of it. The BB penetration and reversal have market all the IT bottoms since the bull market begun in 2003. Oh my...

The decline from July 07 top to Aug 07 bottom took 5 weeks. The rally from Aug 07 bottom to Oct 07 top took 9 weeks. That's a total of 14 weeks for A+B. The decline from Oct 07 to date has taken 14 weeks. A+B = C. Oh my....

Can you call the bull dead ? You be the judge ! I am just presenting the TA case for a incomplete bull market.

My ST to IT indicators remain on a sell signal. If we get a strong rally accompanied with strong volume and breadth, then we could generate an IT buy signal next week. If we crap out here yet again, then the bear market case will start strengthening again. The market is currently wearing out both the bulls and bears. The violent snapbacks have killed the stubborn bears. The spectacular rally failures have killed the stubborn bulls. If one thing everyone is sure of in this environment is failures - both rallies and declines. Strong trending moves come out of this kind of environment. Be prepared !