Sunday, March 11, 2007
The Follow-Through Day: Myth or Money Maker?
Fans and followers of William O’Neil and Investor’s Business Daily have undoubtedly heard of the “Follow-Through Day” (FTD). O'Neil advises that market bottoms can be recognized when the downtrend stops and one of the market indexes is up about 1.7% or more on the 4th - 11th day of the rally. (The Successful Investor, p.9). Basically, a FTD gives a green light to buy after a market correction.
Because Friday, March 9th marks the 5 year, 5 month anniversary of the current bull market and the indexes are currently in a correctional period, I thought it might be interesting to see how accurate the FTD has been in signaling the end of a market correction.
William O’Neil has advocated using up to 2.0%, depending on volume and market liquidity, to determine when a FTD has occurred. This study will also use 2.0%, as the author believes that the figure produces fewer false signals. Also, the operational definition for “market correction” for our purposes is “A period of time when the index ceases to make higher highs, and makes at least one lower low.” When this condition is met, we will consider the market eligible to have a FTD.
On the first chart, we can see the huge moves starting on October 10, 2002, which signaled the beginning of the current bull market. The Nasdaq then corrects, makes a lower low, and becomes eligible for a FTD. On February 14th, a FTD attempt is made. The signal turns out to be false as the low for the FTD is undercut three days later. Exactly one month from the previous attempt, March 14th sees a gain of 4.8%, just one day after making a new lower low. This FTD launches a 10 month series of higher highs and higher lows, for a gain of over 1000 points.
The next chart shows October 2003-October 2004. The Nasdaq has rolled over and begins making a series of lower lows. On March 25, a 3.02% FTD occurs, but fails just a month later. Again in May, a FTD occurs, yet fails in July. Finally, on August 18, another FTD occurs. This one does not give a false signal, and the market runs until January 1st, 2005.
The third chart shows the Nasdaq make a lower low in March. After making a bottom in late April, the market continues to rally through July, never gaining even 1.7% or more in a day to qualify for a FTD. If one would have waited for a FTD during this run, his money would have stayed on the sidelines.
The 4th chart shows the Nasdaq coming down from its July highs, making a new lower low and then a 2.07% FTD in October. This run was good for 7 months and ~350 points. May 2006 brought another correction. In June and July, there are follow-through days, but both fail. The market comes off of its July lows, and rallies, without a FTD. This infamous rally lasts until the end of February 2007 without a correction.
The current correctional phase has not yet witnessed a FTD.
Conclusions.
False FTDs during study period = 5
Valid FTDs during study period = 3
Rallies started without a FTD = 2
Considering these statistics, one stands a good chance of buying a false rally, or missing the beginning of a rally entirely if he chooses to adhere to the Follow-Through Day concept as defined here and popularized by O’Neil and IBD. Limitations of this study are that it covers only a 5-year bull market period, and the author's operational definition of a market correction may differ from O'Neil's.
Because Friday, March 9th marks the 5 year, 5 month anniversary of the current bull market and the indexes are currently in a correctional period, I thought it might be interesting to see how accurate the FTD has been in signaling the end of a market correction.
William O’Neil has advocated using up to 2.0%, depending on volume and market liquidity, to determine when a FTD has occurred. This study will also use 2.0%, as the author believes that the figure produces fewer false signals. Also, the operational definition for “market correction” for our purposes is “A period of time when the index ceases to make higher highs, and makes at least one lower low.” When this condition is met, we will consider the market eligible to have a FTD.
On the first chart, we can see the huge moves starting on October 10, 2002, which signaled the beginning of the current bull market. The Nasdaq then corrects, makes a lower low, and becomes eligible for a FTD. On February 14th, a FTD attempt is made. The signal turns out to be false as the low for the FTD is undercut three days later. Exactly one month from the previous attempt, March 14th sees a gain of 4.8%, just one day after making a new lower low. This FTD launches a 10 month series of higher highs and higher lows, for a gain of over 1000 points.
The next chart shows October 2003-October 2004. The Nasdaq has rolled over and begins making a series of lower lows. On March 25, a 3.02% FTD occurs, but fails just a month later. Again in May, a FTD occurs, yet fails in July. Finally, on August 18, another FTD occurs. This one does not give a false signal, and the market runs until January 1st, 2005.
The third chart shows the Nasdaq make a lower low in March. After making a bottom in late April, the market continues to rally through July, never gaining even 1.7% or more in a day to qualify for a FTD. If one would have waited for a FTD during this run, his money would have stayed on the sidelines.
The 4th chart shows the Nasdaq coming down from its July highs, making a new lower low and then a 2.07% FTD in October. This run was good for 7 months and ~350 points. May 2006 brought another correction. In June and July, there are follow-through days, but both fail. The market comes off of its July lows, and rallies, without a FTD. This infamous rally lasts until the end of February 2007 without a correction.The current correctional phase has not yet witnessed a FTD.
Conclusions.
False FTDs during study period = 5
Valid FTDs during study period = 3
Rallies started without a FTD = 2
Considering these statistics, one stands a good chance of buying a false rally, or missing the beginning of a rally entirely if he chooses to adhere to the Follow-Through Day concept as defined here and popularized by O’Neil and IBD. Limitations of this study are that it covers only a 5-year bull market period, and the author's operational definition of a market correction may differ from O'Neil's.
Comments:
<< Home
Thanks! If one would take the bear market period following the tech bubble, there would be even more false signals generated. I need to go back and re-read O'Neil's writings and see if he specifies how much of a correction is needed before the FTD technique becomes viable.
Great job Wood.
For your free services, I will email you an illegal mexican to clean your computer.
Gapping:
We are not getting crushed on FXY. It's a currency for Christ's sake. What the fuck, do you think it's Vegas?
For your free services, I will email you an illegal mexican to clean your computer.
Gapping:
We are not getting crushed on FXY. It's a currency for Christ's sake. What the fuck, do you think it's Vegas?
Nice job. I remember about 12 or so years ago, the market was undergoing a correction, and O'Neil took up a page in his newspaper to personally state in big bold letters that we were now officially in a bear market.
I believe it was the next day we began rebounding. From then on, I've always thought of him as a fool.
I believe it was the next day we began rebounding. From then on, I've always thought of him as a fool.
Daytrader should watch gold stocks (ABX,GG,AEM) and South African Golds (AU,HMY,GFI) for buys-doing well in spite of a drop in crude. Also SMH should be traded as a buy on dips. Watching AMGN for a possibleturnaround later.
A lot of your "follow through days" are NOT. They are first day of attempted rallies. You confused the two. A follow through day happens on the 4th through 10th day AFTER the first rally day that has a gain of 2% on higher volume. Some of your "follow through days" are marked on the 2nd day off the bottom and are really just rally days. A follow through day can't be on the 2nd day, it has to be the 4th day after a rally day to be considered a follow through day.
You also wrote on a chart in 2003
"Lower low now eligible for follow through day" Lower low does not mean the very next high volume day that has a gain is a follow through- it is not a follow through day. Remember you need a rally day FIRST. Then 4 days later a follow through on higher volume. Read O'Neil's book again - lots of examples.
In order it goes like this:
1] Market is correcting down.
2] Rally day=Day 1. One of the major indexes has a 2% gain on higher volume than the previous day.
3] Wait 3 more days. This is the 4th day and you can now look for a follow through day [2% gain on higher volume than the prior day] from this 4th day through the 10th day.
4] follow through day happens on the 4th -10th day -start looking for stocks breaking out of bases
5]If after an attempted rally day or follow through day, the market goes below the low of
the FIRST rally day, then the rally is negated and you must wait for a new rally day and a new
follow through day.
The best way to find out the correct information is to go to Investors.com Ask IBD and search "follow through day" and the book How to Make Money in Stocks
You also wrote on a chart in 2003
"Lower low now eligible for follow through day" Lower low does not mean the very next high volume day that has a gain is a follow through- it is not a follow through day. Remember you need a rally day FIRST. Then 4 days later a follow through on higher volume. Read O'Neil's book again - lots of examples.
In order it goes like this:
1] Market is correcting down.
2] Rally day=Day 1. One of the major indexes has a 2% gain on higher volume than the previous day.
3] Wait 3 more days. This is the 4th day and you can now look for a follow through day [2% gain on higher volume than the prior day] from this 4th day through the 10th day.
4] follow through day happens on the 4th -10th day -start looking for stocks breaking out of bases
5]If after an attempted rally day or follow through day, the market goes below the low of
the FIRST rally day, then the rally is negated and you must wait for a new rally day and a new
follow through day.
The best way to find out the correct information is to go to Investors.com Ask IBD and search "follow through day" and the book How to Make Money in Stocks
Here is a comparison between the misinformed flyonwallstreet site and the correct way to view a follow through day by IBD.
WRONG WAY on here counts the second day of big gains on big volume as the follow through day.
http://bp3.blogger.com/_xHhMy_z72mc/RfS_A4ReD-I/AAAAAAAAAMI/JCvy0_odWXw/s1600-h/comp_02-03_ftd.bmp
INCORRECT WAY ABOVE
=======================
THE RIGHT WAY IBD's way
http://blogjam.typepad.com/blogjam/images/IBD_7-28-04.GIF
Follow Through Day is point #3 which is the 4th day [it HAS TO BE the 4th day or later to qualify as a Follow Through Day.]
http://blogjam.typepad.com/blogjam/2004/07/from_todays_ibd.html
This is basically what I mean when I talk about a "follow-through day." Sometimes you might get a variation on day 1, for example a powerful reversal from a morning selloff, but the rally just falls short of pushing the indexes into the green for the day. I would consider that the first day of a rally even though the indexes may have closed with a small loss. You can also get variations on the follow-through days themselves. The market is art, not science, and if you are too caught up in exact definitions of what this or that is then you are not using your brain to make judgments. Sooner or later you'll get your ass handed to you if you do that! Here is the article:
Wednesday, July 28, 2004
All Major Bull Rallies Begin With A Follow-Through Day
BY JONAH KERI
INVESTOR'S BUSINESS DAILY
You hear it so much, it's almost become the naysayer's mantra: "You can't time the market."
Short and punchy? Sure. And also completely false.
The market's price-and-volume action gives clear signs of the market's direction. A follow-through day gives you the biggest of head starts — timing the market's bottom.
A follow-through occurs at the earliest stages of a fledgling rally. After a significant market correction, the market will look to regain its footing. Any up day then counts as Day 1 of an attempted rally.
The next two sessions, Days 2 and 3, don't need to show much in the way of gains. As long as they don't undercut Day 1's low, the rally remains intact.
For a follow-through to occur, you want it to land between Day 4 and Day 7 of the attempted rally. On any one of those days, you're looking for one or more of the major indexes — the Nasdaq, S&P 500 or Dow — to rise 1.7% or more in higher volume than the previous day.
Though a follow-through in that span gives the strongest signal for a new rally, one that hits anywhere between Day 4 and Day 10 can work. Follow-throughs that occur after Day 10 yield lower success rates.
Though this method may seem esoteric at first, keep in mind it has decades of IBD research behind it.
To gear up for the next follow-through, study charts of past market bottoms. The Nasdaq flashed a follow-through in October 1998 which kicked off the final, furious stretch that carried stocks to huge gains.
Just remember: Not every follow-through triggers a huge, new bull market. But no raging bull has ever started without one.
After three years of harrowing losses, investors were starved for good news by the time March 2003 rolled around. On March 12 of that year, the Nasdaq fell to a low of 1253.21 before bouncing back for a 0.6% gain in swift volume (See point 1 on chart). That was Day 1.
The Nasdaq roared ahead 4.8% in massive trade the next day (See point 2 on chart). Exciting action? Sure. But you still wanted to wait for Day 4-7 to confirm the new rally.
Click To View Larger Image
On March 17, the follow-through hit on Day 4. The Nasdaq jumped 3.9% in sharply higher turnover(See point 3 on chart). Volume was also above average. Though the index stumbled a bit on March 31 (See point 4 on chart) and a few other times shortly after the follow-through, it never approached its prior lows. The Nasdaq would go on to gain 72% to its January 2004 peak.
Source: Investor’s Business Daily Newspaper and investors.com website.
July 28, 2004 in Mr Market | Permalink
Post a Comment
WRONG WAY on here counts the second day of big gains on big volume as the follow through day.
http://bp3.blogger.com/_xHhMy_z72mc/RfS_A4ReD-I/AAAAAAAAAMI/JCvy0_odWXw/s1600-h/comp_02-03_ftd.bmp
INCORRECT WAY ABOVE
=======================
THE RIGHT WAY IBD's way
http://blogjam.typepad.com/blogjam/images/IBD_7-28-04.GIF
Follow Through Day is point #3 which is the 4th day [it HAS TO BE the 4th day or later to qualify as a Follow Through Day.]
http://blogjam.typepad.com/blogjam/2004/07/from_todays_ibd.html
This is basically what I mean when I talk about a "follow-through day." Sometimes you might get a variation on day 1, for example a powerful reversal from a morning selloff, but the rally just falls short of pushing the indexes into the green for the day. I would consider that the first day of a rally even though the indexes may have closed with a small loss. You can also get variations on the follow-through days themselves. The market is art, not science, and if you are too caught up in exact definitions of what this or that is then you are not using your brain to make judgments. Sooner or later you'll get your ass handed to you if you do that! Here is the article:
Wednesday, July 28, 2004
All Major Bull Rallies Begin With A Follow-Through Day
BY JONAH KERI
INVESTOR'S BUSINESS DAILY
You hear it so much, it's almost become the naysayer's mantra: "You can't time the market."
Short and punchy? Sure. And also completely false.
The market's price-and-volume action gives clear signs of the market's direction. A follow-through day gives you the biggest of head starts — timing the market's bottom.
A follow-through occurs at the earliest stages of a fledgling rally. After a significant market correction, the market will look to regain its footing. Any up day then counts as Day 1 of an attempted rally.
The next two sessions, Days 2 and 3, don't need to show much in the way of gains. As long as they don't undercut Day 1's low, the rally remains intact.
For a follow-through to occur, you want it to land between Day 4 and Day 7 of the attempted rally. On any one of those days, you're looking for one or more of the major indexes — the Nasdaq, S&P 500 or Dow — to rise 1.7% or more in higher volume than the previous day.
Though a follow-through in that span gives the strongest signal for a new rally, one that hits anywhere between Day 4 and Day 10 can work. Follow-throughs that occur after Day 10 yield lower success rates.
Though this method may seem esoteric at first, keep in mind it has decades of IBD research behind it.
To gear up for the next follow-through, study charts of past market bottoms. The Nasdaq flashed a follow-through in October 1998 which kicked off the final, furious stretch that carried stocks to huge gains.
Just remember: Not every follow-through triggers a huge, new bull market. But no raging bull has ever started without one.
After three years of harrowing losses, investors were starved for good news by the time March 2003 rolled around. On March 12 of that year, the Nasdaq fell to a low of 1253.21 before bouncing back for a 0.6% gain in swift volume (See point 1 on chart). That was Day 1.
The Nasdaq roared ahead 4.8% in massive trade the next day (See point 2 on chart). Exciting action? Sure. But you still wanted to wait for Day 4-7 to confirm the new rally.
Click To View Larger Image
On March 17, the follow-through hit on Day 4. The Nasdaq jumped 3.9% in sharply higher turnover(See point 3 on chart). Volume was also above average. Though the index stumbled a bit on March 31 (See point 4 on chart) and a few other times shortly after the follow-through, it never approached its prior lows. The Nasdaq would go on to gain 72% to its January 2004 peak.
Source: Investor’s Business Daily Newspaper and investors.com website.
July 28, 2004 in Mr Market | Permalink
<< Home
