All donations go towards web site maintenance for all of Technical Watch,
keep it free of charge, and may be tax deductable as an investment expense.


PayPal Verified
Join our market chat sessions every Tuesday and Thursday at 4:00 pm Pacific time!
More information on subscriber services can be found at
http://www.technicalwatch.com/subs.htm




Support Technical Watch!
Buy from our Book Store!
Register Calendar Latest Topics
 
 
 


Reply
  Author   Comment  
mortiz

TW Patron++
Registered:
Posts: 1,054
Reply with quote  #1 

This topic has been beaten to death over past year or so, but with another 90% down volume day posted on Friday, 9-7-2007 (per the WSJ), defined as Down Volume/(Up + Down Volume), it dawned on me we have had quite a cluster of 90% down volume days over the past couple of months, and as mentioned by Remo on the Main Board, we've also had three 90% up volume days in that same time window.

Over the past 50 trading days, I count six 90% down volume days and three 90% up volume days.  Looking at the action from the end of World War II, at least six 90% down volume days along with at least three up volume days within a 50 trading day window has occurred only five times, including the current clusters.  Since August of 1945, there have been a total of 224 90% up volume days and 286 90% down volume days, so we can see there is an understandably slight bias toward 90% down volume days per the well known characteristics of herd psychology.

So what follows, with respect to price, a 2:1 ratio of 90% down volume days versus 90% up volume days within a fifty day trading window?  The historical ratio of 90% down volume days to 90% up volume days is about 1.28:1, so there has been a higher than normal bias over the past fifty trading days with respect to 90% down volume days.  I see many analysts and traders assuming since we have had such a plurality of 90% down volume days over up volume days of late, that is must be a bearish omen and the end of Western civilization is upon us.

Let's look at the facts with respect to price gains going forward with conditions similar to what we have seen of late (six 90% down volume days and three 90% up volume days).  Granted, these are rare events thus the sample space is small, but the data is what it is, and I'm sure the bears in particular will discount the results.



One fact of interest is four of the five up and down volume plurality extremes have occurred in years ending in "7", with 1967, 1977, and 1997 not being represented over the past 60 years.  Over the 18 months following such clusters of up and down volume extremes, the maximum drawdown was -6.72%, with an average drawdown of -5.80%, while the maximum potential gain was over 47% with an average potential gain of nearly 30%.

Take these results FWIW, but based upon history, buying the SPX last Friday will likely provide pretty fair gains going forward over the next 18 months.

For the nearer term, here is some potential fresh meat for the bears.  With SPX being down a little less than 1.5% last week, the retail equity options players took advantage of the weekly price dip for buying to open (BTO) equity  and ETF call options.  First the smallest equity options players (1 to 10 contracts per transaction) weekly BTO put-call (PC) ratio chart over the past few years.

The smallest equity options players' BTO PC ratio dropped to 0.47 last week from 0.61 the prior week... the drop in PC ratio approaching the overly optimistic zone.




The medium sized retail equity options players (11 to 49 contracts per transaction) were also buying the dip, on-balance, with their BTO PC ratio falling to 0.62 last week from 0.74 the prior week.  This group's BTO PC ratio is only in the neutral zone, but again, they were buying the dip.



The largest retail equity options traders (50 or more contracts per transaction) also showed more interest in long call positions last week than they have in several weeks, although their call buying was still less than their put buying. Overall, the largest retail equity options players have the best track record (on balance) of being on the right side of the market except at important price bottoms.  At significant price lows, the largest retail equity options players are as vulnerable to fear as any of the others.



In summary, there may have been a little too much dip buying last week to declare all clear ahead for the bulls, but the clusters of extreme up and down volume days over the past couple of months bodes well for the bulls in the longer term.

FWIW,

Randy

0
doc

TW Patron++
Registered:
Posts: 497
Reply with quote  #2 
Superb work Randy! You read my mind as I was wondering what the data would be like with the mixture we have had. Your data on the cluster of 3 90% up days from last year had the condition of no 90% down days interspersed within and that clearly did not apply here in todays market.

It is interesting that the other 4 examples occurred at times the RAAD was testing from above a significant upside breakout level. We have also recently challenged the 1959 RAAD breakout area. Perhaps a washout to lighten the load on the train, eh? Interesting to also note no CRASHES coming out of such data. Not so far... When you get a chance, could you update the RAAD chart you previously posted with the data going back to the 1930s. My observation will be apparent on that chart. Thx

Doc
0
mortiz

TW Patron++
Registered:
Posts: 1,054
Reply with quote  #3 
Hi Doc,

Thanks for the feedback!  I think there were a couple 90% down volume days in the summer of 2006 but I think they were prior to the three 90% up volume days... having trouble with Excel tonight so will verify later.

Here is the cumulative NYSE RA-AD line from 1926, with arrows denoting the approximate time frame when the other three 90% up volume and six 90% down volume days clusters occurred.  The 1946-47 events came after a RA-AD line all-time high, but the market was digesting its robust gains since 1942 and didn't get going in earnest for awhile.  The 1957 event was followed by a nominal new RA-AD line all time high, which happened to have been broken earlier this year.  The 1987 event occurred a few months after the NYSE RA-AD line hit a multi-year high... so the current situation with respect to the RA-AD line is most like the 1946-47 and 1957 events.

Regardless, prices in each case went on to higher prices with the 1957 and 1987 examples boasting stellar gains.




Randy
0
doc

TW Patron++
Registered:
Posts: 497
Reply with quote  #4 
Perfect Randy. Thanks for the chart. As for the summer of 2006, I believe you are absolutely correct in that the 90% down days were before the 3 up days. That is why you set the criteria for that study to not include events with interspersed 90% down days. That is a different study with different entry criteria. Different tools for different times. Thanks.

Doc
0
doc

TW Patron++
Registered:
Posts: 497
Reply with quote  #5 
Interesting study by Jason Goepfert. Scroll down to the March 2007 post on extreme down followed by up volume cluster using a 15:1 ratio...

http://www.hamzeianalytics.net/search/label/Jason%20Geopfert

Doc
0
doc

TW Patron++
Registered:
Posts: 497
Reply with quote  #6 
Well, looks like we have the 15:1 down vol day followed by the 15:1 up vol day here again in August. Same as in March. In March, and nasdaq also fit the mold, but this time, nasdaq was cloooose to 10:1's. More fruits for your studies on historical data. Also what about 2 occurances within a 6 month time frame like we had here in 2007. Has that ever occurred???




Doc
0
da_cheif

TW Member
Registered:
Posts: 137
Reply with quote  #7 

paralysis of analysis is the product of overanalysing the market.....ABC 2 down to the 16th of august..KISS analysis.....watch the sky.....snort

0
doc

TW Patron++
Registered:
Posts: 497
Reply with quote  #8 
Silly me, I enjoyed reading Randy's work and I was kind of glad I discovered a repeat of the 15:1 stuff and presented it to the board for consideration . Cheif, ya gotta give the younger guys a slap on the back once in awhile.
Doc
0
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.






Copyright 2000-2017 Technical Watch