Wednesday, November 19, 2014

Dow Theory Special Issue: Schannep and I brainstorming about the last Dow Theory signal.

Was it a primary bear market signal or just a normal secondary reaction?


PART I



Readers of this blog know that the last primary bear market signal was not an easy one to discern (or at least was not so much textbook-like). After writing about a primary bear market signal on this Dow Theory blog, and seing that the Dow Theorist I respect most was not signaling a primary bull market, I contacted Jack Schannep of "thedowtheory.com" to further discuss the issue.

What follows is the result of our discussions. I think it is best to show our emails as they are without editing (save some highlighting). Since it was a complicated signal (if any), I feel the less I edit the better, and it is up to the reader to decide, after reading Schannep's reasoning and mine, which way to go in the future. 

In any instance, I find  that the thought process followed by Schannep and I is more important that the specific solution (which maybe does not exist, as technical analysis is not math). It has been a real tour de force which has resulted in a deeper understanding of the Dow Theory when applied in real time without the benefit of hindsight.

Since there are lots of text, and those really intend on understanding the Dow Theory are encouraged to read it very carefully, I have splitted the materials. Today I give you part I. Part II should follow soon.

My emails are in read color (I was a bear), whereas Schannep's are in blue (he was a bull).

 -------------- -----------------


Dear Jack,





Hope you are doing fine.



In the spirit of openness we have always had, I'd like to draw your attention to what I consider a primary bear market signal. It is true that the +3% rally has not occurred yet in the ongoing secondary reaction. However, Rhea (page 77, last paragraph, Fraser Edition) made clear that the violation of lows of the last completed secondary reaction (that is the previous one, not the current one) also serve to signal a primary bear market signal when jointly violated.



I have read said Rhea's paragraph at least 50 times, and I feel he is clear enough. Furthermore, it makes sense, since the last recorded secondary reaction lows was a valid "stop loss"" until the primary bull market was reconfirmed. If such a stop loss was valid until the Industrials confirmed higher closing highs, why should cease such a valid technical level to apply? It is like scaffolding or building a ladder (one new step, does not imply the previous step is to be removed). If we ignore it, we risk letting a falling knife fall until your -16% level is reached (ouch!!).



I hope I explain myself. If not, with pleasure I can try to deepen this issue. Of course, I am willing to be corrected, if I am missing something.



More details in my latest post:



http://www.dowtheoryinvestment.com/2014/10/dow-theory-update-for-october-15-stocks.html





Kind regards,

Manuel Blay





Dear Manuel,


Nice to hear from you and I hope all is well with you and your family, as well.  Your 'call' and information in your e-mail and 10/15 piece on your website are certainly thoughtful and thought provoking.  To start out you may be interested in two others who agree with you:






and




 



And for what it is worth, one who doesn't:






and this from Dow Theory Forecasts:



"The primary trend. Both the Dow Industrials and Dow Transports closed at all-time highs in mid-September. The primary trend will be regarded as bullish under the Dow Theory until proved otherwise, and the recent decline is consistent with a correction in a bull market. With rebounds above the September highs of 17,279.74 in the Industrials and 8,676.19 in the Transports, the bullish trend would be reconfirmed. For a bear-market signal, two things must happen. First, the averages need to rebound meaningfully without both reaching new highs. Second, both need to close below the lows established in the current correction".





But I realize you are not interested in what others say, but are looking for my insight into the situation, so here goes:





   Your reading (and re-reading) of page 77 is quite correct in saying it is "clear enough".  Notice, however, how wrong the opening sentence turned out (both averages penetrated the highest points previously (9/17)...the primary bull trend will continue for a considerable period of time (to 9/18 & 9/19) - NOT!  That failure, however, does not invalidate the final sentence to which you allude. The next paragraph does leave a little 'wiggle room': Occasional exceptions can be found, and it is proper that this be true....





   The definition in "Determining the Trend" (Chapter XIII) of successive rallies making new highs and declines holding above low points being bullish (to paraphrase) held true until the sharp secondary setback.  And there was not a failure to penetrate new highs and then decline to new lows which would have been bearish.  So it's not textbook.



Back to page 77 he talks about a situation where "the implication is bullish for the immediate future, but not necessarily indicating a primary bull trend".  I see the current situation as the implication is bearish but not necessarily indicating a primary bear trend, therefore I'll use the completion of a more typical 'setup' for a Dow Theory Sell signal such as I show on the Subscriber's Home page before I'll join you in your Bear market call.  As Hamilton wrote "...It is much harder to call the turn at the top than at the bottom"(p.47).  AMEN ! 



BTW, I realize you used "the last recorded secondary reaction lows was a valid 'stop loss' until the primary bull market was reconfirmed".  But the primary bull market WAS reconfirmed on 1/17 so by my reckoning I would say that stop loss falls away. But then what do I know?

Best Regards,

Jack



Jack & Bart Schannep

Editor & Contributing Editor

for the Schannep Team

thedowtheory.com



Dear Jack,



Sorry for the belated answer. I was on business travel for 2 weeks, and even now time remains scarce.



I feel we have been confronted with a difficult market (from a Dow Theory standpoint) and, thus we all can learn more by analyzing it.

 

I have devoted time to reading all your emails and to read the sources you mention therein. It is a really tough issue.



I am sorry if what follows is too long. I feel, though, we are dealing with a very interesting matter, and as Rhea wrote, the more time it passes, the more we see small nuances and more useful becomes the Dow Theory.



It is true that I am more interested in what you say that in what others say because after having studied them all, I find your approach is the best (and by far). And not only because of performance but because of drawdown contention. However, I also like to know what others are saying and doing because one can always learn something from them.



Starting point:



I look at the markets with Schannep’s glasses.  In other words, I am not looking at three-week corrections (rather 10 days, as you rightly explain), nor at 33% minimum retracement (but just a confirmed declined exceeding -3%, etc. So in appraising the ongoing, and more importantly, the preceding secondary reaction, I do this with Schannep’s rules; not with Rhea’s or other dubious not so excellent Dow Theorists. Furthermore, I avail myself of three indices, not two. Thus, most Dow theorists couldn’t spot a secondary reaction (last August) because Rhea’s requirements for a secondary reaction were not met. On the other hand, Schannep’s requirements were met. This fact leaves you and I alone, as the August lows don’t qualify as a correction if we are to use Rhea’s Dow Theory, and, accordingly, the violation of such lows would not qualify as a primary bear market signal (for those who stick to Rhea).



Thus, if I were to appraise a secondary reaction strictly according to Rhea’s Dow Theory, I feel that what you and I label the preceding last completed reaction (August-September 2014) would not qualify as such: Nor in time (or barely at best) and certainly it did not retrace more than 33% from the previous advance.



Therefore, to declare the existence of a primary bear market on Oct 10th, the following requirements should be met:



1) Follow Schannep’s rules. If one were not to use Schannep’s rules, the last completed secondary reaction of last August-September would go undetected.



2) Be in agreement that the lows of the last completed secondary reaction serve as a valid stop in spite of stocks having made higher confirmed highs (which is contentious).



Since I am the only Dow Theorist (at least that writes publicly) that follows Schannep, no wonder other Dow theorists still believe we are in a primary bull market; and the few that have said that it is a primary bear market have said so by chance (really) but lack proper explanation (more on that later).



So this leaves us: you and I alone. Since we both agree that the preceding completed reaction was that of August-September, the only point to be clarified is whether what was a valid “stop loss” remains in force after making stocks higher closing highs or whether such higher closing highs “delete” or “negate” such stop, and we are left with the abyss below (until the market deigns to have a +3 rally, something which might occur at astounding depths).



I see some kind of circular reasoning when maintaining that higher highs negate the last completed secondary reaction lows as a valid stop loss point. If there had been a failure to make higher highs, without any shade of doubt, those “last completed secondary reaction lows” are our stoploss, since this is precisely one of the setups you describe in your book. In other words, if after the lows of August 2014 stocks had rallied but had failed to make higher (confirmed) highs, then under your flavor and Rhea’s we could not declare the secondary reaction as extinguished, and hence, the lows of such secondary reaction would remain a valid exit point.



When it gets tricky is what happens get stocks make higher confirmed highs. Do such highs imply that, until a new secondary reaction develops and a +3 rally ensues, were are left with no stop, looking into the abyss?



I have reviewed my blog and when stocks made higher highs in September, it was clear to me that the “old” stop loss (the completed secondary reaction lows) was not to be discarded. This post was prophetic:






“This means that if a new secondary reaction develops from these higher highs, it is likely that the resulting new secondary reaction low would lie at a higher level than the current one. All in all, we would have a tighter Dow Theory stop and being risk averse, I think this is not a bad thing. More about the Dow Theory trailing stop here.



What if the new secondary reaction refuses to put its lows at a higher level than the last recorded secondary reaction? Wouldn’t this mean that we risk a higher loss? No. Because, in such a case, the last secondary reaction lows continue working as a valid stop loss



All in all, I didn’t discard the “old” secondary reaction lows as a valid stop loss. Of course, if the “normal” secondary reaction followed by +3% rally occurs at a higher level, I would be very happy ignoring the “last completed secondary reaction lows” but this does not mean I am oblivious to it.



While what follows is not strictly Dow Theory, my trader instincts tell me that one should always have a trailing stop. If higher highs negated the hitherto existing stop (last completed secondary reaction lows) then we would have to “look down below."  (until the market undergoes a secondary reaction and a +3% rally on one index). Of course, we have the -16% Schannep’s stop loss (which is not strict Dow Theory but a very sensible and well tested “loss container”). But, could you imagine the typical Dow Theorist who is not conversant with your -16% stop? He would be left without any valid exit point after stocks make higher highs, if we were to discard the last completed secondary reaction lows as a valid stop loss.  What would happen to a “Rhea” Dow Theorists if stocks, after making higher highs, decline by 20% with no intervening +3% rally in at least one index?



Furthermore, we should bear in mind that according to Rhea’s Dow Theory secondary reactions don’t occur as often as with Schannep’s. All in all: Under classical Dow Theory (deprived of the -16% stop and with more infrequent secondary reactions), “deleting” the last completed secondary reaction lows as a valid stop loss because stocks have made higher highs, is playing with fire and opens the door to potential catastrophic losses (which may have never materialized in the past thx to market goodness, but we must look forward and we know our worst trade is always awaiting us). However, ignore the last completed secondary reaction lows (after making higher highs) in commodities markets (which are not so lenient with investors) and we are exposed to catastrophic losses.



Thus, we trail stops. As soon as a tighter stop appears (that is a new secondary reaction and +3% rally after higher highs), we forget the old, looser one; but if the tighter stop fails to materialize, then we stick to the last recorded stop loss.



I am (was) writing all these without the benefit of hindsight, as stocks are staging a nice rally off the lows. So it may well be that “the last completed secondary reaction lows” is a failed signal. However, even if it turns out to be a failed signal, I find my reasoning does not contradict Rhea, nor Russell (when he was an excellent Dow Theorist). And we know that any individual signal is not supposed to be always right. I’d rather take a small loss and “being wrong” rather than being exposed to a larger loss if the market does not oblige.



Furthermore, a principle of trading is to raise stops as stocks advance. If we are to ignore the “last completed secondary reaction lows” because stocks made higher highs, then we are left with the possibility of lowering our stop as prices advance (as it has just happened to you now: the ongoing reaction lows are lower than the preceding ones), which is something any trader abhors. Please cogitate this: If we are to only follow the “normal” rule, then our current stop loss would be the current secondary reaction lows (not those of the one finished in September), but such lows lie at a lower level. So our open risk is larger now than it was back in August-September. While this is not Dow Theory, I find this “trading rule” is useful. I don’t like to lower stops

to be continued...

Sincerely,
The Dow Theorist 

1 comment:

  1. Awaiting part 2 anxiously. This is getting intellectually interesting!

    ReplyDelete