It seems you have JavaScript disabled.

Ummm.. Yeah... I'm going to have to ask you to turn Javascript back on... Yeah... Thanks.

Stock Market Rocket Reignites at Trader’s Narrative

Stock Market Rocket Reignites

This is a guest post by Wayne Whaley, CTA:

The traditional Breadth Thrust Signal occurs over 10 days when the sum of daily advances outnumbers the sum daily declines by roughly a two to one margin. You can get some insight into thrust potential prior to the ten day period if the thrust is strong enough.

Over the last three days, my 3 Day ADT number is 78.2 (subject to minor revision). ADT = ADV/(ADV+DEC). There have been 14 instances since 1970, where the 3 day ADT has exceeded 78.0. The S&P 500 is a perfect 14-0 one year later with a gaudy 25.58% average return.

3 day ADT historical study May 2010

It would be convenient for the bulls to get confirmation from the 5, 10 and 20 day thrust signals over the next week, but this is a heck of a start.

In the short term, volatility contractions of greater than 21%, like the one we say on May 10th 2010, offer a reason to be cautious. At least in the short term:


VIX contractions SPX returns May 2010

VIX 20 percent changes May 2010

Enjoyed this? Don't miss the next one, grab the feed  or 

                               subscribe through email:  

13 Responses to “Stock Market Rocket Reignites”  

  1. 1 jezza

    Great Stuff Wayne, Martin Zweig would be impressed-outlook remains bullish medium term, good for long term equity investors.

  2. 2 Alex

    Hi Wayne,

    I replicate the same studies on VXO (the old Vix) using the -20% threshold, then optimized in time frame range from 21 to 180TD (holding period). The best fit is sell (yes, SELL) the Vix (so, the future) after it lose more than or equal to -20% in one day and buy it back 67TD later.

    100% hit ratio (9-0). The problem is … a huge adverse excursion :) So, probably the S&P500 goes down (have to check your data) … but the Vix doesn’t go up after 3 Months (67TD).

    I will scan my database to find the best portfolio, in that time frame, of world indexes (ETF) that better perform (outperform) is such cases. More on the subject tomorrow.

  3. 3 Alex


    these are the day on the VXO (Bloomberg Source) when it lose 20% or more in one day (only the first istance in a 67TD time frame):

    21.10.1987 -47.20%
    18.11.1991 -22.44%
    05.04.1994 -27.60%
    11.03.1996 -20.84%
    28.10.1997 -23.22%
    15.06.2006 -28.09%
    06.03.2007 -22.24%
    18.09.2007 -21.90%
    13.10.2008 -29.89%

    Some cool difference with yours … ;)

  4. 4 ww

    I’ll have some more today. There are a couple of things I want to study this morning. One of the problems with the VIX contraction numbers is that they came two days after an incredible one day VIX expansion. I’m a bit more likely to lean toward the thrurst numbers. But trying to resolve. Hope to have some new input in a couple of hours.

  5. 5 ww

    am really struggling to find a mkt that looks like May of 2010, where you are

    1) Days off a 12 mt high
    2) have a 5% intra day crack
    3) recover all of the losses within one week without a retest of lows.

    But take a good look at October 27, 1997.

    The market was 14 days off it’s yearly high and had a 6.89% selloff.

    The S&P closed at 876.98 that day, traded lower (855.27) the next morning, before

    staging an afternoon rally and then never looked back.

    The difference there and now, is that in 97, the sp at least opened lower the day after capitulation selloff.

    A bit perplexing, but everyday that passes, positive momentum is building and the train is leaving.

    A/D line is very strong again today on an unchanged price day

  6. 6 Tom

    Great stuff. Keep it up.

    The only question I have is with you expecting a pullback before marching higher. How many trading days should we see pass in your estimation before a resumption of the uptrend. Is there not enough sample data. 2006 sticks out like a sore thumb and 3 months seems like a fairly long time.

  7. 7 Ernest


    What software do you use for your analysis?


  8. 8 Wes


    Look at mid January of 98 in the NASDAQ. Wasn’t that the retest ?

    Is there some unwritten rule that a retest has to happen immediately ?

  9. 9 ww


    I ran my scan for days that had 5% intraday selloffs. Then scanned further for ones that occurs near tops. Then visually examined the handful. The 06 case could very well be similar, but did not have a 5% intraday selloff.

    I am struggling with whether a retest is in the cards or not. Normally, if a capitulation bottom occurs at the bottom of an extended move downward, there is a 90% probability of a retest. But given that this appears to be a quick sharp correction in a bull mkt, I can’t say with as much confidence. Very few data points to relate too. In the Oct 97 case above, which was associated with the Asian Currency Crisis, the bottom was put in at 876.73 on the capitulation day. The only pullbacks were to 903.68 on day 3 and to 905.96 on day 12, then we never looked back. Both of those pullbacks were sharp quick one day events with no follow through. And the tremendous momentum that I wrote about yday, may prevent a full pullback. maybe 20 points down is all we get, and then you run the risk of where does that 20 points start from.

    If I were an investor, I would be working into long exposure. If you are a trader, you have the quandry of what kind of pullback do you want to play for.


    I write my own computer software, old school, FORTRAN.

    Fridays in big up weeks

    As I type we are up 5.1%. Since 1950, there have been 22 weeks where Mon-Thursday was up 5.0%. Fridays are 14-8 in this spot. The one percent moves were 6-2.

  10. 10 ww

    Saw this on Yahoo finance page

    Todd Harris quoting Jason Goephart

    In the short-term, he notes the big rebound on Monday saw the S&P gap up more than 4% at the open. That’s something that has only happened six other times in recent history, Harrison says, citing the work of Jason Goepfert at

    “In every single [prior] case, the ‘gap’ was eventually filled, and usually very quickly,” he continues.

  11. 11 ww

    Four % gaps

    I scanned for four % opening gaps.

    There were six,

    Three from late Oct 87 and three in the Fall of 2008.

    I didn’t get much help from those, do you?

  12. 12 Wes

    While it’s too early to know exactly, it looks like we will be maximum overbought by Monday week (the 24th). Of course this depends to some extent on what happens between now and then.

    Right now, it looks like we’re getting oversold again with the oversold indicators at a higher level creating a bullish divergence.

  13. 13 jezza

    Gap appears likely to be filled, rapid Dow recovery did not feel right after such a long advance..shaping up to be a difficult summer.

Leave a Reply