Subscribe to RSS feeds Bulkowski Blog via RSS

Support this site! Clicking the links (below) takes you to If you buy ANYTHING, they pay for the referral. My books...

Encyclopedia of Chart Patterns 2nd Edition book.
Chart Patterns: After the Buy
Getting Started in Chart Patterns, Second Edition book.
Trading Basics: Evolution of a Trader book.
Fundamental Analysis and Position Trading: Evolution of a Trader book.
Swing and Day Trading: Evolution of a Trader book.
Visual Guide to Chart Patterns book.
Picture of Bumper.
Picture of the head's law.
Bulkowski's Blog:
Class Elliott Wave Fundamentals Psychology Quiz Studies Setups Software Tutorials More...
Candles Chart
Small Patterns
As of 07/01/2022
  Indus: 31,097 +321.83 +1.0%  
  Trans: 13,289 +132.53 +1.0%  
  Utils: 993 +23.33 +2.4%  
  Nasdaq: 11,128 +99.11 +0.9%  
  S&P 500: 3,825 +39.95 +1.1%  
  Targets    Overview: 06/29/2022  
  Up arrow32,300 or 29,900 by 07/15/2022
  Up arrow14,000 or 12,500 by 07/15/2022
  Up arrow1,000 or 925 by 07/15/2022
  Up arrow12,100 or 10,750 by 07/15/2022
  Up arrow4,000 or 3,700 by 07/15/2022
CPI (updated daily): Arrows on 6/23/22
As of 07/01/2022
  Indus: 31,097 +321.83 +1.0%  
  Trans: 13,289 +132.53 +1.0%  
  Utils: 993 +23.33 +2.4%  
  Nasdaq: 11,128 +99.11 +0.9%  
  S&P 500: 3,825 +39.95 +1.1%  
  Targets    Overview: 06/29/2022  
  Up arrow32,300 or 29,900 by 07/15/2022
  Up arrow14,000 or 12,500 by 07/15/2022
  Up arrow1,000 or 925 by 07/15/2022
  Up arrow12,100 or 10,750 by 07/15/2022
  Up arrow4,000 or 3,700 by 07/15/2022
CPI (updated daily): Arrows on 6/23/22

Written by and copyright © 2005-2022 by Thomas N. Bulkowski. All rights reserved. Disclaimer: You alone are responsible for your investment decisions. See Privacy/Disclaimer for more information. Some pattern names are the registered trademarks of their respective owners.




Thursday 5/28/09. Symmetrical Triangle in the Financials. What Does it Mean?

A daily chart of XLF

Since I doubled my money in Conseco (CNO) about two weeks ago when the stock peaked, I have been watching it from the sidelines, wondering when was the right time to jump back in. I sold the stock at 3.47, point A in the inset on the chart. Price dropped dramatically after I sold.

In the last few days, I noticed that the stock had bottomed at 2.51 (two out of three times), so last night I placed a limit order to buy 2,000 shares at 2.52. That is above the round number 2.50 and a penny above the prior day’s low. Today, I received an email from my broker saying that 100 shares had been filled and my order at 2.52 made the low for the day. With a stock so cheap, the commission charged almost exceeds the cost of the shares!

Looking at the chart, I am glad that I received only a partial fill. Why? Because it shows a descending triangle forming. A descending triangle breaks out downward 64% of the time, which is not good news for owners of the stock. What other indications are there of a downward breakout?

I show the chart of SPDR financial select sector exchange traded fund (XLF). According to yahoo!finance, "The investment includes companies from the following industries: banks, diversified financials, insurance and real estate. The fund will normally invest at least 95% of its total assets in common stocks that comprise the relevant Select Sector Index. It is nondiversified."

I penciled in the red trendlines to form a symmetrical triangle. Notice the inbound trend is upward since price enters the pattern from the bottom. I wondered how often an inbound upward trend lead to an upward breakout? I remembered researching that, trying to settle the notion that continuations patterns outperform reversals (often, they do not).

I checked the website and three of my books before I found the answer in my book, Getting Started in Chart Patterns, pictured on the right. On page 178, the text reads, "A symmetrical triangle in an uptrend usually breaks out upward, so that’s the bet I made. Later I found out that the uptrend-upbreakout scenario happens only 55% of the time. That’s about random."

In the final analysis, the breakout direction does not matter. Price will either breakout upward or downward. If the breakout is upward, then I may buy more of Conseco. If the breakout is downward, I can tolerate a drawdown of my $250 investment. Unfortunately, I noticed the descending triangle in Conseco this evening, after the market closed, despite pondering the breakout direction for days. My guess is CNO will breakout downward with a drop to as low as $1.50.

As for XLF, since Conseco shows a bearish pattern, it becomes more likely that the ETF will also breakout downward. However, before shorting the fund, I would study other financial stocks to be sure, especially those that the fund owns (BAC, BK, CME, GS, JPM, MET, MS, TRV, USB, and WFC round out the top 10 holdings, according to Yahoo.

-- Thomas Bulkowski


Tuesday 5/26/09. Tutorial Tuesday: What You Don’t Know About Head-and-Shoulders.

The three views of a head-and-shoulders top chart pattern, show for symmetry.

Even if you do not follow chart patterns, you may be familiar with the head-and-shoulders top or bottom pattern. For head-and-shoulders tops, the pattern looks like the profile of a person from the shoulders upward. The bottom version has the profile flipped upside down. If you need more information on them, then click on the associated link.


The large image to the right gives an idea of what a head-and-shoulders top pattern looks like. I conducted research on shoulder symmetry and found some interesting results.

  • Head-and-shoulders patterns with an extended right shoulder tend to perform less well. The middle image in the figure shows an idealized example. The right shoulder is farther away from the head than is the left shoulder. The top image shows the two shoulders equidistant from the head. Patterns with an extended left shoulder (bottom image), tend to outperform.
  • Head-and-shoulders top chart pattern shoulder line
  • The more unsymmetrical a head-and-shoulders top appears, the better it performs. I like to think of this as ugly patterns work best, but it varies from pattern to pattern. This finding is a variation of the prior bullet item because it uses a range for symmetrical patterns. For details, read the study.

    In the study, I am comparing head-and-shoulders tops that appear symmetrical about the head (approximately equal distances from shoulder to head) to those not symmetrical. The farther the ratio of the two shoulder distances gets, the better the performance.

  • The more symmetrical a head-and-shoulders bottom appears, the better it performs. This is the same as the prior bullet item except that it pertains to head-and-shoulder bottoms. The results also flip. Again, I am using a ratio of the distances from the shoulders to the head. The ugly head-and-shoulder bottoms perform worst.
  • The smaller image shown above right compares the height of the two shoulder peaks in a head-and-shoulders top chart pattern. Patterns in which the left shoulder is above the right one shows better performance after the breakout. Shoulders that share the same price have the worst performance. Those patterns with a higher left shoulder see declines that average 25% compared to declines of 20% for higher right shoulders and 19% for even shoulders.
  • Head-and-shoulders chart pattern measure rule
  • For head-and-shoulders bottoms, the next image shows the configuration that works best. A higher left shoulder valley when compared to the right shoulder valley results in a larger rise postbreakout, but the difference is small. In a bull market those patterns with a higher left shoulder see rises averaging 39% compared to 37% for head-and- shoulders bottoms with lower right shoulder low, and 37% for those with even shoulder lows.


A neckline is a trendline that connects the two armpits of a head-and-shoulders pattern. The findings for this are simple and they relate to the slope of the neckline.

  • Head-and-shoulders bottoms with down-sloping necklines show outstanding performance, 42% versus 34% for trendlines that either slope upward or are horizontal.
  • Head-and-shoulders tops with horizontal necklines outperform, but the difference is minor, -24% to -23% (up-sloping) and -21% (down-sloping). The minus signs mean a drop after price stages a breakout.

These results can help you pick which head-and-shoulders pattern to trade, given that everything else is the same. Choose the configuration that has the highest performance. Your selection may be a dud anyway because you are dealing with probabilities. However, information such as this gives you an edge over other traders. Sometimes that edge is all you need to slice your way to higher profits.

-- Thomas Bulkowski


Thursday 5/21/09. Broadening Pattern in Wilshire 5000. What Does it Mean?

A picture of the Wilshire 5000 (^DWC) on the daily scale.

Rob asked in an email today why I had turned bullish? The answer to that comes in several parts. First, take a look at the chart of the Wilshire 5000 on the daily scale. I show a right-angled and descending broadening formation outlined by two red trendlines. The broadening pattern is characterized by trendlines that outline a flat top and down-sloping bottom. Premature breakouts are "very rare" according to my Encyclopedia of Chart Patterns, Second Edition book. The breakout is upward 51% of the time. That, of course, is no help at all to determine whether the market is bullish or bearish.

What I find interesting in this pattern is the partial decline, which I highlight in the figure. A partial decline correctly predicts an upward breakout 63% of the time. When searching for partial declines, the chart pattern must be established. By that, I mean it must obey all of the identification guidelines for that pattern. Mostly, this applies to the number of trendline touches. Price touches the top trendline and then drops but does not come that close to the bottom trendline before curling back up. When price touches the top trendline again, it often stages an immediate breakout. By immediate breakout, I mean price does not cross the pattern again. However, it may dance along the top trendline before closing above the line and eventually breaking out. Breaking out reminds me of pimples, but I digress...

If the Wilshire leads the other indexes by signaling an upward breakout, then that is bullish.

Another signal is the chart pattern indicator has flipped from bearish to bullish. I show the indicator at the top of this page, along with the diamond strength indication. There is currently just one diamond showing, so the bullish turn is a weak one. It could strengthen or revert to bearish in the coming days, depending on what happens to the markets.

If you believe in Elliott Wave, then the recent March low marks the end of a bear market motive wave. I do not follow Elliott because I find wave counting confusing and gave up trying to do it long ago. However, using the Dow industrials, I see a bear market extended wave 1 beginning from the high in October 2008. The ABC corrective phase began at the March low. Looking at the Dow chart or even the one above, the wave "A" of the ABC correction has just begun, so I see the upward correction of the bear market downturn continuing. Of course, I could be way off in my wave you can try your hand at wave counting and create your own prediction. It is possible that the March low has marked the beginning of a 5 wave motive phase and not a corrective phase. That would be delicious.

Finally, my blog entry discussed the Dow utility average. That measured move up followed by a return to the corrective phase looks as if it has completed, too. My guess is the utilities are headed higher, but that is difficult to tell from the utility stocks in my portfolio.

Since summer is about to start, the indexes could move sideways, waiting for September to come and bad new to arrive about the default rate in commercial mortgages (such as strip malls). That could take the indexes down to form a complex head-and-shoulders bottom, a pattern that I wrote about in an earlier blog.

So, Rob, those are some of the reasons why I think the market is heading up. Keep in mind that I am frequently wrong.

-- Thomas Bulkowski


Tuesday 5/19/09. Tutorial Tuesday: News About the High and Tight Flag

As you may know, a high and tight flag is one in which price climbs by at least 90% in less than 2 months (price is supposed to double, but I relaxed the rules, somewhat). According to my book, Encyclopedia of Chart Patterns, Second Edition, the high and tight flag has a 0% failure rate and an average rise of 69%. The 0% failure rate means that out of 307 patterns I studied, none failed to climb less than 5% after the breakout. The average rise measured 69% before price tumbled at least 20% (meaning, the trend changed). Out of 23 types of chart patterns in a bull market and 19 of them in a bear market, the high and tight flag was the best performing pattern.

Fast forward a few years. As prices dropped during the 2008 bear market and then bounced off the bottom, I saw a number of high and tight flags form. Then I saw almost every one fail. The doubling in price was just a retrace in a downward price trend. So, I decided to take a closer look at the chart pattern.

A picture of the failures after a hight and tight flag.

The figure shows various price tracks represented by colors. Segments A and B represent the flagpole and flag portions, respectively, of the high and tight flag. In 61% of the cases I looked at, price continued higher toward E. That move is how the chart pattern is supposed to work.

However, in 18% of the cases, price followed track D. That means price climbed above the top of the flagpole and flag combination before tumbling and closing below the lowest low in the flag portion of the high and tight flag (that is, below the low in segment B. Assuming a stop was placed a penny below the flag low, the drop from the top of the high and tight flag to the stop was 10%.

Those cases in which price followed segments D or E, the rise averaged 27%.

In 14% of the cases, price followed path C. These are the tracks that many of the high and tight flags followed when price bounced after the long downtrend in 2008. Although 14% is a small number, it comes from a study using over 2,500 patterns in 552 stocks out of over 1,000 searched since 1995.


A picture of the four combinations of slope leading to a hight and tight flag.

When trading the high and tight flag, I recommend not taking a position until price closes above the top of the pattern. That may come as price climbs above the top of the flagpole or flag, depending on which is higher. Waiting for this type of breakout means you will not be caught when price drops following track C.

How many high and tight flags fail to rise at least 5% after the breakout? Answer: 19%. This result follows track D where price breaks out upward and then collapses. Note that this result is slightly higher than the 18% rate cited earlier due to the 5% rise limit imposed.

To help avoid these types of failures, look at the inbound price trend. I show the figure to highlight what to look for. You will find the best performance if the price moves in a shallow trend leading to the start of the high and tight flag. In this chart, the high and tight flag is represented by the red segment. That segment could be as long as 2 months.

Avoid trading high and tight flags in which the inbound price trend is very steep, either upward or downward. Those combinations result in inferior performance.

I show three links below which expand on the high and tight flag chart pattern. Most of the information presented here comes from the statistical analysis. Refer to that study for additional details and other new information.

  1. Discusses performance statistics, trading tactics and so on.
  2. Shows an example of a high and tight flag.
  3. A statistical analysis of the high and tight flag.

-- Thomas Bulkowski


Monday 5/18/09. Mental Monday: Set a Goal and Win!

A picture the strength indicator for the chart pattern indicator.

Before I forget, I added a strength reading to the chart pattern indicator. I show the new indicator in the chart to the right, which is a snapshot from the top of this page. The number of black diamonds varies from one (weak and unreliable signal) to three diamonds (strong and reliable).

The diamond indicator is experimental and not foolproof, but it should help you interpret signal changes in the chart pattern indicator. Click on the diamonds at the top of this page (not on the ones displayed in the chart) for additional details.

The chart pattern indicator (the red down arrow says that the bull run which started back in early March has transitioned into a bearish retrace as of May 11. The three diamonds suggest the bearish turn is not a false trend change.

# # #

This posting is now located here.

-- Thomas Bulkowski


Thursday 5/14/09. Easy Money Has Been Made. What’s Next?

A picture the Dow utility average on the daily scale.

I am using the Dow utility average as a proxy for what I see happening in the market now.

Price has climbed from the March lows in a nice straight-line run. Then the average went horizontal during mid March to the start of May. After that, price resumed the upward move, peaking 5 days ago. Since then, the utility average has dropped.

The green line shows the bottom of a support zone setup by that horizontal move I already described. I see the average dropping to this line and then rebounding. Since I am just guessing, the average could turn before it reaches the line.

If it does turn, then I expect the average to move up following the blue line. The apex of the symmetrical triangle will pose a challenge should the average rise that far.

I expect the other market averages to follow a similar course. The horizontal support zone is not as pronounced in the industrials, transports and other indexes, but it is there nonetheless.

Let’s take a closer look at why I expect a rebound.

From my study of triangles, I know that the apex -- where the two lines join in the future -- marks a reversal point 60% of the time or a minor high or low 75% of the time. If you accurately draw the red lines, and I have not in this picture, the apex of the triangle is directly above point A. Although that does not help us determine a future path, it is worth checking.

A picture the Dow utility average on the daily scale.

I show a picture of what happens after a measured move up chart pattern. The measured move is the A B C progression of price, just as the utility average shows in real life. Price rises in the first leg, A, goes horizontal in the corrective phase, B, and then moves higher in the second leg, C.

What happens after that? My book, Encyclopedia of Chart Patterns, second edition discusses the answer on page 518, along with Table 33.4.

The red line in the chart shows the typical course that price takes as it weaves up and down in a bull market. Refer to the book for the statistics in a bear market.

I studied over 800 measured moves to figure out the answer, and the adjacent chart shows the statistics. Here are the bull market numbers and what they mean.

  • Price remains above the corrective phase 19% of the time;
  • Price stops somewhere within the corrective phase 35% of the time;
  • Price stops below the corrective phase but remains above the bottom of the measured move up 31% of the time;
  • And price drops below the measured move up 15% of the time.

If you add the percentages, you will find that price will remain at or above the corrective phase 54% of the time and above the bottom of the pattern 85% of the time. Having those kinds of statistics at your fingertips is why so many traders consider my books invaluable resources, but I will let you decide.

-- Thomas Bulkowski


Tuesday 5/12/09. Tutorial Tuesday: How I Doubled My Money in a Week!

A picture Conseco (CNO) on the monthly scale.

A trade like this one comes along too infrequently. The chart shows Conseco (CNO) on the monthly, logarithmic, scale. In some of my charts, when I discuss the target price, I have highlighted overhead resistance like the red line on this chart. The flat base makes for a tempting price target.

In this case, the relatively horizontal price structure has lasted for years, from 2003 to 2007, before price broke down. To me, that suggests the fair value of the stock should be at or above that line, or somewhere in the range of 17 to 25.

Reversion to the Mean

All of this implies that price will rise back up to that level. It is a concept I call reversion to the mean and it is borrowed from the math department. The idea behind reversion (or regression) to the mean is that prices way out of normal will eventually return to the pack. They will revert back to their average value. My recent buying is using that concept to find price targets for long term (2-3 years) ownership. During that hold time, I expect all of the stocks I buy to double or triple. If a stock shows less than a potential triple, I often throw it back into the market pond.

So, that is how I got my price target. I also know that the year after a bear market, price tends to recover dramatically (based on research using fundamental analysis). In the first year after the 2000 to 2003 bear market ended, price climbed 25% in the stocks I studied. A year after that, price continued moving up, but at a much slower rate. I am not saying that price will climb 25% this year. I am saying that the year after a bear market ends is when you will see a strong move up, just like we are seeing now.

Why Buy Now?

Why buy the stock and why this one? On my fishing expedition, I first noticed the overhead price target well above where the stock was trading, so I decided to take a closer look at the company. The research reports said the news was not good. S&P wrote in a report dated April 3, "Our risk assessment for Conseco reflects the potential for the company to file for bankruptcy protection. In addition, our assessment takes into account CNO’s weak capital position, the potential for further investment losses, and volatility in earnings. We also see risk associated with ongoing litigation and restrictive covenants." Scary stuff, all.

On the plus side was the insider buying. On April 2 and 3, insiders bought stock 14 times with additional buying stretching back into 2008. None were selling until as far back as October 10, 2008 when one insider sold 1,466 shares. Even as price dropped, insiders were buying and the heavy buys (200k, 25k, 20, 14.3k and 10k shares) happened in April 2009. That tells me the risk of bankruptcy is possible but overblown.

From a fundamental analysis perspective, the ratios were right were you want them to be: at the lower end of the 3 year channel for price to earnings, price to cash flow, price to book value and price to sales. On a value basis, assuming the fundamentals were still valid, the company was cheap. Of course, the fundamentals use historical data and anything can change in the back office kitchen, depending on how they cook the books.

I started looking at the stock on April 5 and noted that earnings would come out on May 4 according to yahoo!finance. That was outside the 3 week zone in which I will not buy a stock for fear of price tumbling.

Here is the criteria I used to find this stock.

  • Price shows a flat ceiling for years
  • Long-term price target ($17 - $25) is well above the current price ($1.64)
  • Fundamentals at or near low end of 3 to 5 year range (price to earnings, cash flow, book value, and sales)
  • Quarterly earnings announcement is more than 3 weeks away
  • Headline search and research reports turn up nothing scary about the company
  • Insiders are buying large amounts with little or no sales
  • The stock is cheap and represents good value

A picture Conseco (CNO) on the daily scale.

The Buy

When price climbed out of a congestion zone, I bought and received a fill at $1.64. I forgot about the earnings report. My intent was to hold the stock for the long term, to see it climb from 1.64 to 17 or even 25.

Each day price seemed to climb but I did not focus on it. If the stock market did well (which I expect), then the stock would also. Why? Because insurance companies take their outrageous premiums they charge you and me and invest it in the stock market. If their investments rise, the worth of the company rises and the stock should follow.

When I looked at CNO this morning (Monday). It did not dawn on me that the stock had gapped up 45% (from yesterday’s close of 2.69 to a high so far of 3.90). Then I went searching for the reason for the gap. Earnings. The company announced earnings this morning, something that I forgot about. Fortunately, they were better than the market expected.

The Sale

Knowing that I bought the stock at $1.64 and it was now at 3.50 and dropping, I decided to pocket my winnings. I sold it and received a fill at $3.47.

What happens to the stock in the coming days is anyone’s guess. I think it will make a higher high tomorrow (Tuesday) and perhaps coast higher before rounding down. I consider this an inverted dead cat bounce.

I will say that I am not pleased with this trade. I gave up too much money (from the 3.90 high). I was worried about another trade that went bad, and that still colors my perception of this one. I also missed the earnings announcement, meaning that I bought just a few days before the announcement, despite having a checklist to prevent such mistakes. Of course, if I had followed my checklist, I would not have bought the stock.

Here are my reasons for the sale.

  • The likelihood of price dropping exceeds the upside short-term potential
  • I doubled my money in 6 days! Yippee!

I made about 110% on the stock in 6 days. Annualized, that is a return of almost 6,800%. That highlights the fallacy of annualizing returns, but wouldn’t it be nice to see that kind of return every week?

-- Thomas Bulkowski


Monday 5/11/09. Mental Monday: Have You Entered the Twilight Trading Zone?

This posting is now located here.

-- Thomas Bulkowski


Saturday 5/9/09. Saturday Supplement

A picture of Air Force 1 flying over the statue of Liberty

Attached is a picture of Air Force One (called that when the president is aboard) that so upset residents of New York City. Today’s national news reports said that Louis Caldera, who headed the White House military office, resigned over the incident.

Not living in New York and hearing about the event, my first thought was, "What did it cost?"

Answer: $328,835. Your tax dollars at work!

I think that those responsible should repay the taxpayers for the cost of such a stupid photo op. Losing their jobs is a plus, too.

-- Thomas Bulkowski


Thursday 5/7/09. Striking Oil in USO

A daily chart of USO, United States Oil

I use an exchange traded fund called United States Oil (USO) as the proxy for the price of oil. According to yahoo!finance, "The investment seeks to reflect the performance, less expenses, of the spot price of West Texas Intermediate (WTI) light, sweet crude oil. The fund will invest in futures contracts for WTI light, sweet crude oil, other types of crude oil, heating oil, gasoline, natural gas and other petroleum based-fuels that are traded on exchanges. It may also invest in other oil interests such as cash-settled options on oil futures contracts, forward contracts for oil, and OTC transactions that are based on the price of oil."

I show a chart of the fund on the daily scale. I prefer this fund over others because the volume is so large. Over 16 million shares traded just today.

The etf has been trending downward since mid July 2008 when it peaked near 120 (only a portion of the decline appears in the chart). It closed today (Wednesday) at 31.43. Taken on a wider scale, it looks like a rounding turn, a bowl shape, but the chart has fooled me before.

The chart shows a head-and-shoulders bottom chart pattern. The left shoulder is marked LS, head is, well, head, and the right shoulders is marked RS. A neckline joins the two armpits, which I show as a red line. When price closes above the neckline, it signals a buy but that is only for down-sloping necklines. If the neckline sloped upward, then I would use the price of the right armpit as the buy signal. Why? Because steep up-sloping necklines may never trigger a purchase because price cannot catch the line.

A head-and-shoulders bottom is a bullish chart pattern and one that says oil and energy related plays are worth taking a look at. If you do your homework on oil plays like the oil services industry, you will find that older rigs are being cold stacked (put in storage). Some companies are trying to sell some of their excess rigs. I cannot recall any delays in taking delivery of new rigs, though. Also, research reports say that land based drilling is becoming scarce, given that both oil and natural gas have been trending lower. The deep water plays are where the action is. Drilling in deep water is more expensive, so larger oil companies take multi-year contracts on the rigs to do that type of drilling.

What is clear from looking at some of these stocks is that 1) they are cheap, 2) it may take longer for them to recover than expected. If you have a long term horizon, at least 2 or 3 years but perhaps much longer, like 5 or 10 years, you can do well.

Even though the price of oil, based on the above chart, appears headed up, I would like to see some confirmation. That could come from the big players increasing their capital spending for the coming year. From what I have heard, the spending has been getting cut amid a world awash in oil. Food for thought.

-- Thomas Bulkowski


Tuesday 5/5/09. Tutorial Tuesday: Position Sizing.

There are many ways to correctly size a position for trading. I am going to discuss three of them and then tell you what I use. Not covered is the one by Ralph Vince in his book, The Handbook of Portfolio Mathematics. It is on my favorite book list, but many of you will find the math a challenge. I emailed him and asked for a simplified equation, and he didn’t laugh, but it sounded like he did.

  1. Fixed Size.

    Let’s say you have a portfolio worth $100,000 and you are comfortable holding 10 positions. Divide the two to get the position size: $100,000/10 or $10,000 for each trade.

    For example, Let’s say that I wanted to boost the taxes I pay to the government for dividends and found an electric utility called DTE Energy Co. Holding (DTE). It pays 7% and earnings cover the dividend (I am not recommending this stock, just using it as an example, and no, I don’t own it, either). When I checked it, the stock was trading at 30.24. Dividing $10,000/30.24 gives 330.69 shares. Using a fixed size, you would buy 330 shares of DTE at an approximate cost of $10,000.

    That is how fixed size works.

  2. Percent-Risk.

    DTE on the daily chart showing stop location

    Percent-risk position sizing is based on risk to determine the size of each position. Staying with DTE, suppose I placed a stop loss at 27.53, which is below the prior minor low shown in the chart at the red line. Say I want to risk no more than $2,000 per position. How many shares should I buy?

    The formula for this approach is:

    DollarRiskSize/(BuyPrice - StopPrice)

    In this example, the DollarRiskSize is $2,000, the BuyPrice is $30.24 and the StopPrice is 27.53 giving a result of $2,000/($30.24 - $27.53) or 738 shares. That would cost you 738 x 30.24 or $22,317.12.

    Using this approach would allow you to buy about 4 positions for your $100,000. While $2,000 is just 2% of 100k, it represents too high risk in my view. If you cut it in half, risk just $1,000 per position, you could buy 8 positions. That would give you a more diversified portfolio providing you bought stocks from different industries.

  3. Percent-Volatility.

    This method adjusts the risk for volatility of the stock. Testing by Active Trader magazine says it performs much better than the percent-risk method.

    Here’s the formula.

    PositionSize = (CE * %PE) / SV
    Where CE is the current account equity (size of portfolio)
    %PE is the percentage of portfolio equity to risk per trade.
    SV is the stock’s volatility (10-day EMA of the true range).

    For example, if the current account equity (CE) is $100,000, the percent of portfolio equity we want to risk (%PE) is 0.2% (1/10 of 2%), and the stock’s volatility is $0.83, then the result is: ($100,000 * .002) / $0.83 or 241 shares at a cost of $7,288.

    I used 1/10 of 2% because using the full 2% chews through almost the entire $100,000 and I want a diversified portfolio. This method could provide about 10 to 12 positions for 100k. For simplicity, I also used an average of the true range over 10 days instead of the exponential average.

What I Use

I use a variation of the first method. When I started trading, I bought enough shares, rounded to 100, to eat $2,000. If a stock cost $19, for example, I would buy 100 shares.

In recent years, my portfolio has grown so I use $20,000 per trade. Sometimes I buy multiple $20k positions in one stock (often utilities). When the market started to show higher volatility last year, I dropped the position size to $5,000 and even $2,500 per trade. For each position, I divide the position size by the stock’s price. Using DTE, this would give $5,000/30.24 or 165.34 shares. I rounded that up to 200 shares at a cost of about $6,050.

These small positions mean I can buy a lot of different stocks, and I do, having no more than 3 stocks from any one industry (except for utilities, of which I own 5, but they are diversified across 3 regions from the United States). I keep buying stocks and often do not run out of money because some trades will not work out and need to be sold and others will hit their price targets or have moved up as much as I believe they can. They are sold, too. The turnover generates new cash that I use to buy more stock.

I am not saying this approach is the best or that it tested well because I have not tested any of these methods. I am just saying this is what I use.

-- Thomas Bulkowski


Written by and copyright © 2005-2022 by Thomas N. Bulkowski. All rights reserved. Disclaimer: You alone are responsible for your investment decisions. See Privacy/Disclaimer for more information. Some pattern names are the registered trademarks of their respective owners. Use the best: Linux for servers, Mac for graphics, and Windows for Solitaire.