Your pop up blocker may be preventing the MarketSmith tool from opening.
Learn how to resolve this issue.
This posting is in response to a previous forum thread titled “Guidelines for When 1st Day of Rally is Reset.” I looked beyond just the reset of the rally but how an attempted rally starts and ends. The data set used in this posting is for the S&P 500 from 11/1/2009 through 12/31/2011; all four major indexes were used for signaling the FTD days. The following are the attempted rallies for the S&P 500:
· 29 attempted rallies
· 16 of them failed (5 of the attempted rallies failed and made a new 1st day of attempted rally on the same day, effectively resetting the rally count).
· 13 had a follow through days (FTD) and became a confirmed uptrend.
The NASDAQ, NYSE Composite and DOW all experienced a similar pattern of failed attempted rallies, but do vary slightly. The number of FTDs is the same, as it only takes one index to generate a valid FTD for the overall market.
Once they have a FTD day in any index the market outlook is now a Confirmed Uptrend. You would now follow the rules for when a Confirmed Uptrend becomes an “Uptrend under Pressure” or a “Market in Correction”; a separate discussion.
You should not confuse the period of time from the start of the attempted rally to a FTD with the uptrend that occurs from confirmation day (FTD) to market top. They each have their own rules.
On five occasions, the attempted rally had reset its first day of attempted rally. The five days are: 5/21/10, 5/25/10, 5/27/10, 8/27/10 and 8/31/10.
The rules for the first day of an attempted rally are simple.
1. We must be in a correction and not in an Attempted Rally already.
2. For each index, is today’s close higher than the prior day’s close, if so, than that index is in an attempted rally.
3. It is possible to have an attempted rally start on the same day that an attempted rally fails. Effectively this just resets the attempted rally count. This would occur, when the market makes a new intraday low (at which point the previous attempted rally has failed), and then rallies intraday to close higher than yesterday’s close (making a new 1st day of attempted rally).
4. There is an exception to this rule on closing higher than the prior day’s close; if the day is a bearish reversal, undercutting the corrections low and rallying intraday on big volume to regain most of its losses. If this occurs, it can be counted as the first day of an attempted rally. This occurred on 7/7/2010 and is described in the following article. On this date, the markets closed down a small amount <0.2% on big volume, but intraday it was down as much as 1.7% and it closed above the prior days intraday low. The following article mentions this exception.
Once you are in an attempted rally, the market has only two ways to go. Either the attempted rally can fail or we can have a FTD. If we have a FTD the market becomes a confirmed uptrend, and the rules for when the market is in an attempted rally no longer applies.
An attempted rally fails, when
1. the Market Outlook is “Market in Correction”,
2. and, the index’s 1st day of an attempted rally is still valid;
3. And, today’s intraday low for an index undercuts its prior intraday low since the start of the current Market in Correction.
This is computed for each index. Not all indexes need to be in the same cycle in the attempted rally.
Once we have the FTD, all of the attempted rallies for all of the indexes get CANCELLED and the Market Outlook becomes a “Confirmed Uptrend”. No need to track the 1st day of an attempted rally anymore, as the rules for a Confirmed Uptrend are now in effect.
Two other Big Picture articles that may be of interest:
1. It takes only one index to flash a valid signal. On 9/11/2010 the Market outlook changed to Confirmed Uptrend when only the NYSE Composite Index had a FTD day.
2. On 8/27/10 the NASDAQ flashed a FTD. The attempted rally started on the same day that it undercut its prior low and previous attempted rally. Illustrating how resets of the day count for the 1st day of an attempted rally do occur.
Distribution Days do not come into play when looking at attempted rallies only when determining market outlook changes when we are in a Confirmed Uptrend or an Uptrend under Pressure.
A market that is in a confirmed uptrend has a number of conditions to signify the end of the confirmed uptrend and the start of either an uptrend under pressure or a correction. One of the rules is if it undercuts its lows, similar to the rule for attempted rallies. A confirmed uptrend never resets and will eventually end in a correction, ignoring for a moment situations when an uptrend is under pressure – which will be covered in a separate post. These rules are different than those used while the market is making an attempted rally.
I Hope this helps at least in answering the question on the guidelines for resetting the 1st day of a rally.
I would also be interesting in a discussion on weak or strong FTD’s. And looking at which ones fail before achieving meaningful results. Hopefully being able to identify weak FTD’s that do not produce any meaningful results and can be ignored. This might be best in a new posting with a new title. Of the 13 FTD’s I looked at, 2 lasted less than 2 weeks and occurred at the bottom of the 2010 correction while it was still forming a bottom. Besides looking at adjusting the parameters of a FTD, you might also look at the other confirming indicators; such as the actions of leading stocks, the action of the 85/85 Index, which industries are leading and some of the market sentiment and psychological indicators that IBD publishes on the market page.
Organizer of the IBD Meetup in The Villages, FL
I did have one question on the start of an attempted rally. Using the bullish reversal rule that IBD used on Day 7/1/2010, that started an attempted rally, to confirm the uptrend four days later on 7/7/10. See the big picture posted on 7/7/10 and a NYSE chart.
The positive reversal allows for a day to close below (but close to) the prior day's close. If that day occurred on higher volume, creating a distribution day, and pushing the distribution count too high, putting the uptrend in a correction. Can such a day also be counted as the first day of an attempted rally? It did occur on the S&P 500 on 11/8/2007 (The market date was 11/08, the newspaper date was 11/09).
However in this situation, the next day undercut the attempted rally which cancelled the it. So we don't know if it had created a FTD, what IBD would have done. I use this date as an example as I was able to confirm that IBD called the Market in Correction on this date. There are other occurrences before 11/08/2007, and it could happen in the future.
IBD any opinion?
Organizer of the IBD Meetup in The Villages
© 2013 MarketSmith, Incorporated. All Rights Reserved. MarketSmith® is a registered trademark of MarketSmith, Incorporated. All data provided by William O'Neil + Co. Incorporated unless otherwise noted.