Analyzing OEX Components for Insight into Large Cap Participation
By
Rennie on Tuesday, May 4th, 2010 at 9:36 pm
Stock indices gapped lower at the open Tuesday and never looked back, fulfilling the short-term negative signal from last Friday but at the same time triggering another short-term sell given the elevated number of new 20-day lows. Historically, when new 20-day lows exceed 2,000 (as was the case Tuesday), there’s usually a bit more downside in store either immediately or after a short-term bounce. In the table below is every instance over the last six years in which new 20-day lows exceeded 2,000 for the first time in at least three sessions…
New 20-day Lows Top 2,000 First Time in Three Days
05/04/10… ???
02/04/10… Lower S&P close two sessions later
01/27/10… Lower S&P close one session later
01/22/10.,.. Lower S&P close four sessions later
10/28/09… Lower S&P close two sessions later
07/08/09… Lower S&P close two sessions later
02/17/09… Lower S&P close one session later
01/15/09… Lower S&P close two sessions later
11/13/08… Lower S&P close one session later
10/24/08… Lower S&P close one session later
09/29/08… Lower S&P close four sessions later
09/05/08… Lower S&P close two sessions later
06/20/08… Lower S&P close two sessions later
06/10/08… Lower S&P close one session later
03/17/08… No lower close within next four days
03/03/08… Lower S&P close one session later
01/16/08… Lower S&P close one session later
01/04/08… Lower S&P close two sessions later
11/19/07… Lower S&P close two sessions later
10/22/07… No lower close within next four days
08/10/07… Lower S&P close one session later
07/24/07… Lower S&P close two sessions later
02/27/07… Lower S&P close three sessions later
06/08/06… Lower S&P close one session later
05/15/06… Lower S&P close one session later
10/12/05… Lower S&P close one session later
10/06/05… Lower S&P close two sessions later
04/15/05… Lower S&P close three sessions later
03/23/05… Lower S&P close one session later
08/06/04… Lower S&P close four sessions later
07/22/04… Lower S&P close one session later
05/07/04… Lower S&P close one session later
04/29/04… Lower S&P close one session later
Note that in 30 out of 32 cases, or 94% of the time, the S&P500 posted a lower close (below the setup day’s close) within the next four sessions, significantly above the 72% random chance in the same time frame.
It wasn’t that surprising to see stepped-up volatility given the recent pair of outside weeks on the weekly SPX chart, a sign that volatility was likely to stay elevated (see this April 28th column). At the same time, assuming the bubble theory still has legs, downside potential should be limited from current levels. The S&P is trading just above long-term trendline support in the SPX 1168 area, which should continue to hold assuming there’s still some significance to that trendline channel. Also encouraging is that while the OEX closed under its lower bollinger band, few components of the OEX closed under their lower bands. Here’s a quick look at each instance since the beginning of last year in which the OEX initially closed under its lower bollinger band and the number of component stocks that also closed under their lower band. While we can’t look further back in history until we finish the tedious process of recreating the old OEX, I think it’s safe to say today’s -12 reading is a fairly low number…
05/04/10… -12
01/22/10… -49
12/17/09… -10
03/02/09… -36
02/17/09… -30
01/20/09… -31
01/14/09… -29
A chart of this indicator is scheduled to be added to the nightly rotation starting tomorrow (here’s an early look). The line on the chart is the net difference between the number of OEX stocks that closed over their upper bollinger band and the number that closed under their lower band. As of Tuesday’s close, no components closed over their upper band while 12 stocks closed under their lower band, resulting in a -12 close. Significantly, that’s below the -16 reading recorded on April 27th, suggesting we may have made a momentum low back in late April and we’re now following up with a lower price low on waning momentum. If true, we should see this bollinger band spread indicator continue to hold above that -16 level as the S&P stabilizes around current levels.
The concept of momentum highs and lows preceding price highs/lows has been a recurring theme at TraderFeed when Brett references his Demand/Supply numbers and DSI (demand/supply index) – see this March ’09 post for a good overview. This represents our latest effort at creating a similar momentum indicator based on a basket of stocks in relation to their volatility envelopes. Here’s a longer-term version of the above chart since January 2009. Note that when price makes a higher high or lower low in a short period of time while the indicator fails to follow suit, it’s usually been a good indication of a significant trend change. We’ll be making more history available in the coming weeks to see if this tendency persists.
Analyzing OEX Components for Insight into Large Cap Participation
By Rennie on Tuesday, May 4th, 2010 at 9:36 pmStock indices gapped lower at the open Tuesday and never looked back, fulfilling the short-term negative signal from last Friday but at the same time triggering another short-term sell given the elevated number of new 20-day lows. Historically, when new 20-day lows exceed 2,000 (as was the case Tuesday), there’s usually a bit more downside in store either immediately or after a short-term bounce. In the table below is every instance over the last six years in which new 20-day lows exceeded 2,000 for the first time in at least three sessions…
New 20-day Lows Top 2,000 First Time in Three Days
05/04/10… ???
02/04/10… Lower S&P close two sessions later
01/27/10… Lower S&P close one session later
01/22/10.,.. Lower S&P close four sessions later
10/28/09… Lower S&P close two sessions later
07/08/09… Lower S&P close two sessions later
02/17/09… Lower S&P close one session later
01/15/09… Lower S&P close two sessions later
11/13/08… Lower S&P close one session later
10/24/08… Lower S&P close one session later
09/29/08… Lower S&P close four sessions later
09/05/08… Lower S&P close two sessions later
06/20/08… Lower S&P close two sessions later
06/10/08… Lower S&P close one session later
03/17/08… No lower close within next four days
03/03/08… Lower S&P close one session later
01/16/08… Lower S&P close one session later
01/04/08… Lower S&P close two sessions later
11/19/07… Lower S&P close two sessions later
10/22/07… No lower close within next four days
08/10/07… Lower S&P close one session later
07/24/07… Lower S&P close two sessions later
02/27/07… Lower S&P close three sessions later
06/08/06… Lower S&P close one session later
05/15/06… Lower S&P close one session later
10/12/05… Lower S&P close one session later
10/06/05… Lower S&P close two sessions later
04/15/05… Lower S&P close three sessions later
03/23/05… Lower S&P close one session later
08/06/04… Lower S&P close four sessions later
07/22/04… Lower S&P close one session later
05/07/04… Lower S&P close one session later
04/29/04… Lower S&P close one session later
Note that in 30 out of 32 cases, or 94% of the time, the S&P500 posted a lower close (below the setup day’s close) within the next four sessions, significantly above the 72% random chance in the same time frame.
It wasn’t that surprising to see stepped-up volatility given the recent pair of outside weeks on the weekly SPX chart, a sign that volatility was likely to stay elevated (see this April 28th column). At the same time, assuming the bubble theory still has legs, downside potential should be limited from current levels. The S&P is trading just above long-term trendline support in the SPX 1168 area, which should continue to hold assuming there’s still some significance to that trendline channel. Also encouraging is that while the OEX closed under its lower bollinger band, few components of the OEX closed under their lower bands. Here’s a quick look at each instance since the beginning of last year in which the OEX initially closed under its lower bollinger band and the number of component stocks that also closed under their lower band. While we can’t look further back in history until we finish the tedious process of recreating the old OEX, I think it’s safe to say today’s -12 reading is a fairly low number…
05/04/10… -12
01/22/10… -49
12/17/09… -10
03/02/09… -36
02/17/09… -30
01/20/09… -31
01/14/09… -29
A chart of this indicator is scheduled to be added to the nightly rotation starting tomorrow (here’s an early look). The line on the chart is the net difference between the number of OEX stocks that closed over their upper bollinger band and the number that closed under their lower band. As of Tuesday’s close, no components closed over their upper band while 12 stocks closed under their lower band, resulting in a -12 close. Significantly, that’s below the -16 reading recorded on April 27th, suggesting we may have made a momentum low back in late April and we’re now following up with a lower price low on waning momentum. If true, we should see this bollinger band spread indicator continue to hold above that -16 level as the S&P stabilizes around current levels.
The concept of momentum highs and lows preceding price highs/lows has been a recurring theme at TraderFeed when Brett references his Demand/Supply numbers and DSI (demand/supply index) – see this March ’09 post for a good overview. This represents our latest effort at creating a similar momentum indicator based on a basket of stocks in relation to their volatility envelopes. Here’s a longer-term version of the above chart since January 2009. Note that when price makes a higher high or lower low in a short period of time while the indicator fails to follow suit, it’s usually been a good indication of a significant trend change. We’ll be making more history available in the coming weeks to see if this tendency persists.