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 Frequently Asked Questions

Class Elliott Wave Fundamentals Psychology Quiz Studies Setups Software Tutorials More...
Candles Chart
Small Patterns
As of 03/22/2019
  Industrials: 25,502 -460.19 -1.8%  
  Transports: 10,052 -233.37 -2.3%  
  Utilities: 784 +4.91 +0.6%  
  Nasdaq: 7,643 -196.29 -2.5%  
  S&P 500: 2,801 -54.17 -1.9%  
  Tom's Targets    Overview: 03/14/2019  
  Up arrow26,900 or 25,200 by 04/01/2019
  Up arrow10,700 or 9,900 by 04/01/2019
  Up arrow800 or 750 by 04/15/2019
  Up arrow7,900 or 7,300 by 04/01/2019
  Up arrow2,875 or 2,700 by 04/01/2019
As of 03/22/2019
  Industrials: 25,502 -460.19 -1.8%  
  Transports: 10,052 -233.37 -2.3%  
  Utilities: 784 +4.91 +0.6%  
  Nasdaq: 7,643 -196.29 -2.5%  
  S&P 500: 2,801 -54.17 -1.9%  
  Tom's Targets    Overview: 03/14/2019  
  Up arrow26,900 or 25,200 by 04/01/2019
  Up arrow10,700 or 9,900 by 04/01/2019
  Up arrow800 or 750 by 04/15/2019
  Up arrow7,900 or 7,300 by 04/01/2019
  Up arrow2,875 or 2,700 by 04/01/2019

Written by and copyright © 2005-2019 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.

Last updated: 3/22/2018





Software, Patternz, Chart Pattern Indicator (CPI)

Can I have the Patternz DLL? No. One doesn't exist and I've no plans to create one.


Books and Magazines


See Also


Do I make more money from my books than from trading?
I have never made more money from anything in my life than trading. That includes my annual earnings as a hardware design engineer, senior software engineer, and software manager.
Where can I read your articles?
I have written for several magazines including Active Trader; Stocks, Futures and Options; Technical Analysis of Stocks & Commodities ( Some of them offer my articles for sale.

I've also written for Traders' ( and The Technical Analyst.

Which of your books should I buy/read first?
I am often asked what is the difference between my books, and which one should I buy first? Here's the short answer.
  1. Visual Guide to Chart Patterns. This teaches you how to find chart patterns.
  2. Getting Started in Chart Patterns. This gives an overview of how to make money using chart patterns.
  3. Trading Basics. Discusses trading basics like money management, stops, support and resistance.
  4. Fundamental Analysis and Position Trading. Teaches you about value investing and trading buy & hold positions.
  5. Swing and Day Trading. Discusses shorter term trading techniques and setups.
  6. Trading Classic Chart Patterns. Find chart patterns more likely to work using a scoring system.
  7. Encyclopedia of Chart Patterns. A reference book that contains everything you need to know about almost all chart patterns.
  8. Chart Patterns: After the Buy. Describes what you can expect after a chart pattern issues a buy or sell signal. How will the stock behave after you buy?
  9. Encyclopedia of Candlestick Charts. A reference book that contains everything you need to know about almost all candlesticks.

Here is the long answer.

  • Chart Patterns: After the Buy.
    Focus: Shows how stocks typically behave after a chart pattern appears.
    My book, Encyclopedia of Chart Patterns is to human anatomy as Chart Patterns: After the Buy is to human behavior. Chart Patterns: After the Buy shows how a stock acts after a chart pattern appears, so we can use that knowledge before buying to boost the odds of trading success. It covers chart patterns, but from a different perspective than all of my other books. Read an excerpt on double bottoms from the book (pdf: 1.58 mb). This book is meant as a reference.
  • Encyclopedia of Candlestick Charts
    Focus: Candlesticks and their performance.
    This book is the definitive reference book covering 103 candlestick patterns. This is one of the few candlestick books that tests each candle pattern and reports on that performance in a clear and concise manner. Compare the performance of one candle pattern to another and choose only the best to trade.
  • Encyclopedia of Chart Patterns.
    Focus: The internals of chart patterns.
    This is a definitive (and popular) reference book on chart patterns, reviewing 53 chart and 10 event patterns packed with performance information, identification guidelines, failure rates, breakout statistics, size and height stats, volume stats, trading tactics, and more. It peaked at #5 in Australia on
  • Top

  • Fundamental Analysis and Position Trading: Evolution of a Trader
    Focus: Value investing.
    A primer about value investing (P/E and PSR ratios, book value, cash flow, dividends, and so on) and adding market timing to a buy-and-hold strategy using position trading. I test each of the fundamentals to see how they perform. The book tells what to look for when selecting stocks to double your money and to find 10-baggers (stocks that rise by 10 times the investment).
  • Getting Started in Chart Patterns
    Focus: Chart pattern information and trading the stock market for those with some knowledge of both.
    This book is a popular, low cost choice for both the novice and expert trader wishing to tune-up the basics. Written as a narrative (not a reference book), it discusses chart patterns and includes details on many of my trades, so you can get a feel for how I use chart and event patterns. This book is more advanced than the Visual Guide to Chart Patterns described below.
  • Swing and Day Trading: Evolution of a Trader.
    Focus: Swing and day trading.
    The book explains how to use chart patterns to swing and day trade, including major reversal times for day traders, plus the opening range breakout and opening gap setups. This book zeros in on the shorter time scales. It also includes a chapter on horror stories, which is an interesting read all by itself. I interview traders and discuss their botched trades.
  • Trading Basics: Evolution of a Trader.
    Focus: Information you'll need when trading or investing in the stock market.
    This book covers subjects like how much money you'll need for trading/investing, position sizing, scaling in and out of positions, leverage, dollar cost averaging, portfolio composition, how long to hold a stock, with entire chapters dedicated to stop placement, and support & resistance. It has a chapter on 45 tips every trader should know. This book tests each idea to see how well it performs and provides an in-depth look at what you need to invest and trade stocks.
  • Trading Classic Chart Patterns.
    Focus: a scoring system for chart patterns.
    The book introduces a scoring system to help you select the most popular chart patterns that lead to big gains.
  • Visual Guide to Chart Patterns
    Focus: A beginner's guide to chart patterns for those new to chart patterns
    This book is an entry level guide to chart patterns, taking you from how to recognize them, what are minor highs and lows, constructing trendlines (Part I), and into the many varieties of chart patterns (part II), ending with buy setups (part III) and sell signals (part IV). This narrative book has lots of color charts and also includes quizzes, which are difficult but fun to take.


Where can I buy your books?
The larger bricks and mortar stores carry them when they are first released. After that, you can order them from the store. The e-retailers such as and others carry them. Click here for more information on my books.
What other books do I recommend?
I have a page dedicated to my favorite books. Click here to visit them.
Would you tell me how to trade the Forex market?
Based on the number of emails asking about the Forex market, it sounds like many of you are losing money or at least getting into trouble. I don't trade the Forex market so the only help I can offer is to say that many of the techniques described on this website (such as chart patterns and candlestick patterns) are applicable to the Forex and other markets. Other than that, I don't know how patterns behave in the any market (currency, futures, options, commodity, etc) outside of the stock market.
Do chart patterns work in the Forex market?
This is the most asked question I receive, probably because I wrote a book review on the Forex market. The question is a good one but I do not know the answer. I do not trade the Forex market.

The reports from other traders that ply the Forex market suggests technical analysis is well suited to Forex because it tends to trend. Chart patterns appear and can be used as buy and sell signals. However, false breakouts from both chart patterns and congestion regions are common, as in any market. Thus, you need to take care when using any trading technique.

I have read that Fibonacci retracement levels and extensions work especially well, so bone up on those subjects. Look into moving averages, oscillators, candlesticks, and Bollinger bands. All of those indicators tell a story and once you know how to read the story, the message becomes clear. And I don't know if candlesticks work better or worse in the Forex market, either.


Why don't I write a book on patterns in the commodities or Forex markets? and How much money do I make writing books?

Writing a book, especially the kind I write because of the intensive research required, takes a long time and requires a lot of effort, much of it boring mouse clicking (to catalog the thousands of chart patterns).

I wrote my first book, Encyclopedia of Chart Patterns, not for the money but to discover how chart patterns work and to share that information with others. Having been published, it is now a question of motivation. I can use my website to share information, which I do -- for free -- so why write a book? I don't have a business to peddle that would benefit from the publicity, and I do not want to be famous, so what remains is money.

But how much money can you make writing a finance book? My book editor gave the following guidance (this is for hard cover, non fiction, moderately priced ($50). Paperbacks and the fiction world are different, so keep that in mind).

  • An unsuccessful book will sell 1,200 copies.
  • A moderately successful book will sell 5,000 copies.
  • A successful book will sell 10,000 to 12,000 copies.

All of those numbers are net sales over 3 years. In other words, after your book hits the shelves, three years later you will have sold from 0 to 12,000 copies. I have found these numbers to be accurate and very tough to reach, especially since the transition to e-books. It's as if people stopped reading.

A first-time author will receive a royalty of about 10% on the wholesale price of the book. The wholesale price is a bit less than half the list price. If your book has a list price of $50, you will earn royalties based on a price of about $24.

Let's say that your book is "moderately successful" and you sell 5,000 copies after three years of being published. Your earnings will be about 10% x $24 x 5,000 or $12,000. Since that represents three years of sales, you will have made $4,000 per year or about $1.92 per hour, assuming you worked on writing the book for 40 hours per week for a year.


Let's say you have a successful book and sell 10,000 copies, your hourly rate doubles to $3.84, well short of federal minimum wage of $7.25 per hour (rate as of July 2016). To earn minimum wage, you'd have to sell almost 19,000 books, and that would only make you as much money as flipping burgers.

The higher the price of the book, the fewer sales you will have. The broader the audience for your book, the more sales you are likely to see. A book about making money in the stock market has a higher sales potential than one about chart patterns in the Forex market. The more esoteric the topic, the fewer the sales. If you have a well-known name like Suzie Orman or Oprah Winfrey, you can put your name on sticks of used chewing gum and they would sell, but who ever heard of "Bulkowski?"

If I cannot make minimum wage by selling a book about chart patterns in the stock market, what makes you think I can make more by targeting a smaller audience, such as the commodity or Forex markets? It is not worth the effort to write such a book because the money just isn't there. It's not worth a year's worth of effort to sell 1,200 books, despite how good the book may be. When you factor in the poor customer reviews on that even the best authors get, why bother? I can mow people's lawns and make more money with less stress.

Where can I get your industry list?
The list of industries (for industry relative strength) and stock list can be accessed from the home page. You will find it in a spreadsheet format (.xls), middle bottom of the page, called Industry performance detail (.xls). Just above that link on the home page, you will find "Industry relative strength." From that page, you can find the industry, stock, and other downloads. All of those pages are updated weekly and they do not change much anyway.


How do I assign stocks to an industry?
When I first started building the list of stocks I follow daily, I used Value Line to assign stocks to an industry. Now, I use yahoo. After selecting the stock in a yahoo industry, I will read the profile (also an option on yahoo!finance) and even visit the company's website to verify that they are a big player in the industry. Sometimes, brokerage reports will list competitors, so you can find stocks for an industry that way.

Yes, there are industry codes that help with the assignment of a stock to a particular industry, but you will find that the composition of an industry varies from source to source. Over time, the company will sell assets or buy new ones, shifting the focus of the company and the industry that best represents them. Some sources may not have the latest information and companies compete in many industry segments anyway, so often it is difficult to assign stocks to just one industry.

What do you know about bitcoin and cryptocurrency trading?
My forte is stocks and the stock market. I don't know anything about bitcoins, altcoins, or cryptocurrency markets or trading them. I don't know whether chart patterns which may appear in the cryptocurrency markets work, how well they work, what they look like, and so on. Sorry.
How much money do I need to get started trading?
I started with $2,000 and bought shares in stocks valued between $5 and $20. Over time, I added more capital and increased the buy range to stocks valued up to $100. Stocks valued below $1 tend to go bankrupt or are in financial difficulty and make poor trading candidates. A stock below $5 makes me worried, too.
How much money can I make trading?
Your annual income can be unlimited, but don't be deceived. An article in the September 2005 issue of Active Trader magazine discusses an accounting firm that handled the tax returns for over 1,000 stock traders. In 2004, 36% of the traders showed a profit and just 4% made over $50,000. If you expect to make $100k in your first year, you had better think again or start with a very large portfolio.
How can I "retire" from my day job by 36 like you did?
Save your money and become debt free, invest wisely, and get lucky. That's what I did. If that doesn't work for you, marry wealth or inherit! Buy low, sell high. If a stock doesn't go up, don't buy it (paraphrased from Will Rogers).


Patternz won't download the indices (indexes). Why?
Patternz WILL download any security symbol, providing you give it a symbol the quote provider understands. If you get an error, it's likely a problem with the symbol choice and quote provider choice (google, quandl, etc) and not Patternz.

Check with the quote provider and ask if they support the symbol you're looking for. Many free quote providers do not allow the downloading (by patternz) of indices. Google and quandl don't offer them and yahoo only offers the S & P 500 index. If you want the Dow industrials, for example, you'll have to visit the quote provider's website and download it manually from there.

For example, if you ask yahoo to download symbol ^GSPC, it will download the S & P 500 index for you. Send the same symbol to google and it will reject the request because it doesn't understand it. Another example. NBL is Noble Energy on yahoo, but try to download NBL from google and it'll cough up fur balls. On google, you have to use NYSE:NBL as the symbol.

Why do chart pattern indicator signals appear one day and disappear the next?
Before I explain this, be sure to read the WARNINGS for the indicator. Too many people are trying to trade using the signal and that's a mistake.

I use the NR7 chart pattern, but wait up to 7 days for a breakout to occur from that pattern for each stock. Let's say that the chart pattern indicator (CPI) says that the market is bearish on Wednesday with 150 NR7s completed but price has not broken out upward or downward yet (closed above the highest high or lowest low during the past 7 days). On Friday, let's also say that the market makes a huge move up, say 150 points. Most of those 150 NR7s waiting to breakout will do so on Friday when their stocks close above the highest high posted by the NR7. Thus, the bullish and bearish counts for the prior 7 days will be adjusted accordingly and that may well cause a bearish signal to disappear. That's why signals can appear or disappear for up to 7 days from the current day. It happens when the markets make large price swings either up or down.

Why do daily CPI readings not match those from the spreadsheet?
One answer is that I use data that is not current on adjustments for dividend payments. They are adjusted for splits, but not dividends. If your data includes both, then that could be the reason.

Before I explain this, be sure to read the WARNINGS for the indicator. Too many people are trying to trade using the signal and that's a mistake.

The daily chart pattern indicator readings are posted on most pages, at the top right of the screen as a date (clickable hotspot) under "Test Portfolios and CPI". If you click on the date, you'll see the bullish, bearish and waiting for breakout numbers. The waiting for breakout numbers are the key. If those waiting on the sidelines decide to breakout, then they can change the HISTORICAL CPI values. That means the daily totals will differ from those in the spreadsheet.

Simply put, the CPI totals can change for up to 7 days. That means the daily readings are snapshots for that day, but those numbers will change as more NR7s breakout. The new totals appear in the spreadsheet and even those will change (the most recent 7 days, that is. Ones older than that will not change).

The more complex answer is the same as to why signals appear or disappear. I use the NR7 chart pattern, but wait up to 7 days for a breakout to occur from that pattern for each stock. Let's say that the chart pattern indicator (CPI) says that the market is bearish on Wednesday with 150 NR7s completed but price has not broken out upward or downward yet (closed above the highest high or lowest low during the past 7 days). On Friday, let's also say that the market makes a huge move up, say 150 points. Most of those 150 NR7s waiting to breakout will do so on Friday when their stocks close above the highest high posted by the NR7. Thus, the bullish and bearish counts for the prior 7 days will be adjusted accordingly. The new totals will be reflected in the spreadsheet (and will not change if they are older than 7 days), but they will no longer match those posted daily.


How do I calculate the chart pattern indicator (CPI)?
My answer is this: Why would you want to? I update the website daily with the reading and also provide charts of the indicator each Friday. Plus, Patternz will calculate it for you at the touch of a button. If you want to use the indicator as a trading vehicle, you're making a mistake. If you don't understand why, or want to try building it anyway, then consult the following resources, arranged in order of importance.
  1. This is the main indicator construction page along with warnings about trading it.
  2. Here is the current list of symbols used in construction of the indicator.
  3. NR7 chart pattern. Understanding this will help.
  4. Download the Patternz program (windows XP). DON'T install it, just read the instruction manual (a .txt file) on the CPI section.
  5. CPI update page has pictures of the indicator results.
  6. If you want to use the indicator in non-US markets using Patternz, then create your own list of stocks to include. Place them in the folder where Patternz resides. Keep the quote data short, about 1 year in length or even shorter (3 to 6 months will do) to keep the processing speed up. I suggest using the new Patternz program and not the Windows XP version. The more stock files you process, the better the results. Before downloading every stock symbol in your markets, try downloading 100 or so. Check how long it takes Patternz to process that many. Then add more as needed and time allows.

Others that have tried to replicate the indicator have had success, but it's not an easy gig (nor terribly hard). Most trip over the "backfilling" of counts between the NR7's end and the breakout. Pay special attention to that. If you read the Patternz manual and the above construction page, you should understand how I count the patterns.

I want to trade using your chart pattern indicator.
I specifically warn against this because signals can change for 7 days. Usually, they are stable after 3 days but that is an observation.
Why aren't my Patternz patterns the same as the setups reported each Friday?
For Friday's blog post ("Patterns Trading Setups for the Weekend"), I manually find and catalog each pattern during the week and those are reported each Friday. Patternz finds patterns automatically. I may discard Patternz's patterns and add more of my own to the mix.


Does Patternz accept data from Metastock?
Patternz cannot read Metastock files. Patternz can only read data files ending in .csv (comma separated values) or .txt (text), and only one security should appear in each data file. Thus, use the Metastock program to covert the files into .csv format and then use the File Format page in Patternz to make sure the format is what the program expects. The new version of Patternz will download quote data automatically so you don't have to read in .csv or .txt files. Again, you should visit the File Format page in Patternz to configure the program correctly for the data files.

I have no intention of making Patternz compatible with Metastock data. Why? Because I don't use Metastock and trying to verify that the files work can be troublesome. Plus, if they change their format, Patternz will break. Lots of users will be upset and blame me. Then they'll want me to fix it. NOW! And they will want it done for free. Why should I bother with the hassle?

Why don't I charge for Patternz, the free program that finds chart patterns and candlesticks?
  • I do not want to be in the business of selling software.
  • I do not want to worry about people ripping me off by selling my software on the Internet and pocketing the money.
Can I have the Patternz algorithms?
No. They are too valuable to just give away. And promising you will not share them with anyone will not work either, regardless of whether you sign a non-disclosure form. I've already learned of one individual considering paying a firm to reverse engineer Patternz.
How does Patternz work?
First I identify the peaks and then I find the valleys in a historical price series. Once I have those, then it is just a matter of stitching them together. A triple top is just three peaks near the same price. The algorithms are simple 'if-then-else' constructs but the package is quite accurate and very fast providing you keep the data files short (a year or two long).

If you want a more technical approach, then create a software package (genetic algorithm) that when given a date range where a known-good chart pattern appears, it will determine the recognition rules and then find other examples, learning from its mistakes. That will take longer to build than the 3 months or so it took to write Patternz.

Can you recommend trading software?
I wrote my own trading software, which I do not distribute to the public. Since I have what I need to trade, I have not shopped around for other software. I have no recommendations. The bonus issue of the Technical Analysis of Stocks & Commodities ( magazine has a reader survey ("Readers' Choice Awards") that covers software. I would check there to see what other traders are using.
How can I get Patternz to work with Vista?
My computers run Windows XP and not Vista, so I am of little help. However, check the Patternz (win XP) page which shows some of the operations that users have been forced to undergo to get Patternz to work with Vista. If none of the solutions offered works, then visit the Microsoft website and see if they have another solution.

The new version of Patternz may work with Vista, but I haven't tried it. It does work with Windows 7-10.


Will I get Patternz to work with Vista or Windows 7-10+?
A new version of Patternz has been released. It works with windows 7 home premium and newer editions of the operating system.
Will I get Patternz to work on a MAC?
I have no plans to make Patternz compatible with non-Windows machines. I have enough trouble keeping it working with current versions. Sorry.
Will you be my mentor?
No. Why not? Because it takes too much time (up to 4 hours of research for each trade), and I have too many people asking for help.

That's why I wrote my books, and that's why I have a free website. The books are easy to read and understand. They are a lot less expensive than paying for a coach or mentor.

The Evolution of a Trader series of books explains the four trading styles and how I tried them all. It's a good place to start to get tips on buy and hold, position, swing, and day trading.

The Getting Started in Chart Patterns and Visual Guide to Chart Patterns are wonderful entry-level books for people new to chart patterns. However, they are not primers on how to buy a stock for those who know nothing about equities.

I wrote a mentor article to help you newbie traders, so be sure to read that. If you follow its guidance, it'll help a lot.

Visit the research page and tutorials on my website. Read my books. Use the site's search feature for common terms. If you can't find your answer there, then contact me.

Read the next question and answer to better understand what goes into successful trading.

What one chart pattern can make me a bundle?
The question is a simple one but the answer is not. It's a lot like asking a dealership, "Which vehicle is best for me?"

Then answer depends on whether you are in your midlife crisis and want a boy-toy corvette to troll for a trophy wife. Or will a bus suffice for the dozen kids you have? Or maybe you need an 18-wheeler to ship that cargo for your business. The correct answer depends on more questions, questions only you can answer, not me.

For chart patterns, the long answer is that learning to trade them correctly is like taking years to become a brain surgeon. Don't expect to be successful overnight, or even in a year. You'll have to know the answers to questions like these.

Which breakout direction should I trade?
Do reversals or continuation patterns perform best?
Should I look for them at the end of long trends or at the start of trends?
Should the inbound trend be steep or shallow or does it matter?
How important to success of a trade is the industry in which my stock does its primary business?
How important is the market trend?
Is the market choppy and volatile, range bound, or trending?
How often will I get a throwback or pullback?
Should I wait to buy until after a throwback completes?
Should I trade busted patterns instead?
Am I a buy-and-hold investor, position trader, swing trader or day trader?
Am I planning to trade as a hobby or invest for the longer term?
Do I have enough money to last a downturn?
Is my portfolio diversified to spread risk or concentrated to make big bucks quickly?
Is the stock volatile, high or low (even negative) beta?
Is the stock a trend follower or a countertrend player?
What time frame will I be trading? 1-minute, 5-minute, hourly, daily, weekly?
Should I use the log or linear scale when viewing charts?
What time frame should I use to view my charts?
Do I have the patience to become a successful trader?
Can I keep emotion out of my trading?
Can I handle the pressure of trading alone or do I need a group dynamic?
Do I have the skills to do the research necessary each day before and after trading?
How will I handle loss after loss after loss?
Should I use leverage?
Will I blow out my account with too much leverage?
Should I trade options, stocks, forex, futures, or all of them?
Will listening to my broker make me broker?

That one simple question has turned into a complex answer. It's so complex that traders like me write books on the subject.

If you have had the patience to read through the long answer, I'll give you the short answer. Rectangles. Good luck with that.


I'm from India (or wherever) and I need help with our markets. Can you help me trade the Sensex (or whatever)?
Read the question about mentoring.

Your markets are different than the U.S. You have different rules and your trading style will be different, too. I'm not familiar with your markets so my advice either won't work or might be wrong anyway.

I also deal with stocks only, not currencies or other security types.

Use this website to learn tips you can use in your markets or find ideas to test on your securities. Fear and greed don't change from culture to culture, but only you can decide what applies to your style of trading and what will work where you live.

I tested your chart pattern/candle/whatever and got different results. Why?
The question is often phrased like this: "I tested your chart pattern or candlestick pattern and got different results. Bulkowski, you must be wrong."

When I was a hardware engineer, I used to blame software for all of my problems. When I turned into a software engineer, I blamed hardware. Either way, the fault was nearly always mine. As I matured, I decided to quit blaming others for my own mistakes and first assumed that I was the cause.

Here are the reasons why you can't reproduce my results.


  1. Different data. I dislike using one security like the QQQQs for my tests. Thus, my tests often involve hundreds of stocks (from 500 to thousands) and years of daily price data. If you don't use the same data, you'll get different results. Depending on how thorough your tests are, your results may or may not be close to mine.

    My data is split and dividend adjusted at the time of initial download, and then split but not dividend adjusted thereafter. Not adjusting the dividends will cause different results. This is one reason why the results will be close, but not exact.

    Data is a problem with Patternz when users try to reproduce my chart pattern indicator. Their stock symbol list is different than mine and they get slightly different results. No kidding! If you use a lot of symbols, then the results should be close enough. I tested my 650 symbol database and another trader who used the entire stock market, over 5000 symbols. The results were not the same, but very close.

  2. Different methods. If you use a different testing method than the one I use, then expect different results. If you decide to use stops and I choose to measure performance at selected intervals (how far price moves a day later, 3 days later, a week later, and so on), then yes, the results will be different. You should expect that. Think of it as you choosing to race a Honda Civic and I choose a Corvette. Both are cars, but don't expect your Civic to win every race.
  3. Different testing times. Frequently, I split the results into bull and bear markets because I think it better helps traders determine what will work best under different market conditions. If you combine bull and bear markets together, in one test, then your results will be different than mine.
  5. Different time scales. I use end-of-day data for my tests because that's what I have available. If you test using intraday data or weekly data then that may explain the different results.
  7. Different markets. My tests use securities based in the United States. I have seen a chart of the same stock from the US and from another country, in which the charts look different enough that I could see a pattern in his chart, but not in mine. Even comparing end-of-day quote data finds that yahoo!finance and google finance often disagree. Imagine what a candlestick on the one-minute scale will look like if your chart and mine are off by one second. In that one second, price can move dramatically, changing the shape of the candle. I found this out when day trading with another trader who lives 30 miles from me. He used a different data source and his candles were sometimes different than mine.
  9. Automated v. manual. For chart patterns, I prefer quality over quantity and that means I manually looked at each of more than 38,500 chart patterns when I wrote the second edition of my book, Encyclopedia of Chart PatternsEncyclopedia of Chart Patterns 2nd Edition book.. If you use an automated service, then recognize that they will not be as accurate as your eyes in locating chart patterns. And that includes Patternz (my free pattern recognition software). It makes mistakes.

    I use the engine behind my Patternz program to help find patterns, and use other technology to help catalog the various features of a chart pattern, but I make up to six passes on each pattern to verify that it is valid and to make sure that all features are accurately represented in my data collection.

  10. Dividend adjustments. My data was adjusted for splits and dividends when I downloaded it. I update it each trading day and will manually adjust it for stocks splits, but not dividend adjustments. Thus your price data, if adjusted for dividends, will not match mine, nor will your performance numbers.

Before you blame me for any difference in testing results, consider how difficult it is to verify that aspirin can reduce heart attacks. Some studies indicate that it works, others say it doesn't, and the rest found no meaningful benefit. If any of the above listed items apply to your tests, then you can expect different results. It's possible that both of our tests are correct, but are specific to our market.


How many stocks do I follow daily?
I have about 620 securities which I review periodically. Among them are the major indices, over 100 ETFs (exchange traded funds), and lots of stocks.

What chart patterns are the most reliable and which do I like?

You can find them ranked here. I like descending triangles, symmetrical triangles, double bottoms, and head-and-shoulders bottoms.

When I go shopping for stocks, I like to see an upward breakout from a flat base (for an initial push upward) and a cloud bank (high profit potential) overhead.

What indicators do I use in my trading?
Price is the best indicator! It's the only one I use now.

Years ago, I used three additional indicators: Wilder's relative strength index (RSI), commodity channel index (CCI), and Bollinger bands (BBs). For the RSI, I looked for divergence (price moves one way and the indicator goes the other way), over bought/over sold signals, and failure swings.

I used CCI for divergence and for confirmation of short term trades.

For BBs, when price nears a horizontal band, it tends to bounce off it. Also, when the bands narrow, expect a breakout from congestion.


How do I use the measure rule?
The measure rule varies from chart pattern to pattern, but it usually involves finding the pattern's height and adding it to (upward breakouts) or subtracting it from (downward breakouts) the breakout price. The result is the target price. A better method is to compute the height as just described and multiply it by the "Percentage meeting price target." Click here to view the measure rule formulas and details.
Portfolio: How do I pick stocks to follow?
I have two stock lists. One is a list of stocks I own or am thinking of owning (my portfolio). The other list is a watch list. The stocks in the watch list are the ones I pull from and put in my portfolio. I don't go trolling for stocks in the general market. Rather, I just scan my watch list each day and stop on anything that interests me. Reviewing the same stocks each day means you get to know both the stock and industry. You know when a stock is cheap and when it is expensive. This entry describes how I build the watch list, not my portfolio list.

When I built the original list years ago, I used Value Line (available free in most libraries) and picked companies that I knew were solid fundamentally. I looked for price to have a heartbeat, meaning up and down movement over the years. I looked at the yearly price range. If it was just a few points wide, I moved onto the next stock. I wanted to see the low price was half the high price in at least two of the past 5 years. In other words, the stock doubled in at least two years.

I avoided selecting companies too expensive or too cheap. The preferred price range was about $5 to $ 20. That meant I could buy many shares without worrying about them heading for bankruptcy (like penny stocks trading below $1). I selected stocks with a positive price to earnings ratio (if Value Line shows NMF in the P/E ratio box at page top, I skipped the stock). I also selected stocks with a large 3-5 year price appreciation potential.

I built a list of at least five stocks in each industry, just to be sure I had good representation. If the stock was a utility, I looked at the yield first then safety rating and read about how safe the dividend was in their view. I looked at the payout ratio, debt load, and other fundamental factors before adding it to my watch list.

That was then. Since my list is built, when I want to fatten up an industry (by adding a company) or if I am adding a new industry to the list, I will look at yahoo!finance and scan their industry list (this may have changed because they revamped their website and got rid of the good stuff). I will read the profile of the company to be sure that it fits into my definition of the industry (in other words, if it mines gold and has a small subsidiary that makes concrete, I do not want to put it into the cement industry.

I will look at market cap because I want to have the big boys (large caps) represented and will add many stocks to the industry providing they are above about $100 million in market cap. Below that and their existence gets dicey. Also, I avoid thinly traded stocks, say, below 100k to 250k shares daily.


How do I know when the S&P 500 index (the market) has bottomed?
On the way down, bullish chart patterns disappear. For example, you do not see a double bottom confirming (price rising above the peak between the two bottoms) when a stock continues making new lows. When the market bottoms, bullish chart patterns appear, but that is not the time to buy. Those patterns will have upward breakouts, sure, but many will fail soon after when price reverses (heads back down). The market has not begun trending, and it is a dangerous time to trade. Bad news pushes the market lower, almost daily.

Eventually, price will level out. You will see many basing patterns, such as rectangles. Bad news that used to drive the market down by hundreds of points in a single day hardly budges it now. You will see many bullish chart patterns forming, such as double bottoms, head-and-shoulders bottoms, and triple bottoms, as if individual issues are just begging to move higher, but there is still some hesitation. Those with upward breakouts will have price continuing to rise in a nervous stair-step move. Before, one low followed another but now, price makes higher valleys and higher peaks in many securities. That is the time to buy.

Remember that the market looks ahead six months, so even though you hear bad news, the market may move up anyway, shrugging off the news and bursting upward on good news.

One other tip. I would avoid trading high and tight flags. After an extended market drop, or a short but severe one, stock prices will quickly double, forming a high and tight flag. Many of those will see price collapse instead of confirming the chart pattern. Even if they do confirm (a close above the flagpole or pattern high), they'll reverse quickly. You won't see price move up another 100%. You'll be lucky to see it move higher by 10% to 15%. Many will just drop immediately.

How do I set price targets for the indexes?
I check the direction of the chart pattern indicator. A trend in motion tends to remain in motion, so my guess will follow the direction of the CPI. Once I have the direction, then I look for the nearest support or resistance area. Those will be prior peaks or valleys, trendlines setup by prior peaks/valleys, and horizontal consolidation regions. I look for the nearest round number, but that's less important for an index because it's based on component parts (it's an average). Finally, I estimate how long it will take price to reach the target. If the target is close, it should take less time providing the index is trending. If the index is moving horizontally or if the target is farther away, then I will give it more time. All of this is just guess work. Once I arrive at an estimate then I announce the numbers so they appear under "Tom's Targets" in the grid at the top of most website pages.


Will you join us on our webinar/show/whatever?
No thanks. I am a private individual who does not own a camera hooked up to the internet. So I'd have to buy one, join skype (or whatever), and I'm not interested in doing that.

If you want to know how I trade, read my books (start with Getting Started in Chart Patterns (it's filled with my own trades) or read my written interview.

Can I have a business relationship with you.
I'm not interested in any profit sharing (such as using my website to host your ads) venture or collaboration (such as writing an API for my Patternz program). I'm just not interested in any business venture.
Can I contribute content to your website?
No. I provide my own content for the website and am not interested in hosting the content of others. This includes mortgage or loan interest calculators and other applications, and ads, even if you promise, swear to God, that you'll compensate me. Not interested.
I tried to reproduce your results but can't. Why?
My data was adjusted for splits and dividends when I downloaded it. I update it each trading day and will manually adjust it for stocks splits, but not dividend adjustments. Thus your price data, if adjusted for dividends, will not match mine, nor will your performance numbers.

If you want an exact match, then you have to use the same data, the same chart pattern types, the same number of chart patterns, the same time periods, and the same methodology. If you do that, the results will match exactly. Vary any one of those and the results may or may not match.

For example, if I am running a test in a bull market and you select a bear market, then we'll get different results. If I measure from the chart pattern's breakout to the ultimate high and you do not, the results will vary.

Think of it as trying to prove whether aspirin prevents heart attacks. One test says it does. One test says it does not, and a third test is inconclusive. If enough tests are run, then the consensus will trend toward the correct answer.

Many other people have conducted their own tests and found results similar to mine. The results from my tests also match those of other researchers. For example, in my book, Fundamental Analysis and Position TradingFundamental Analysis and Position Trading: Evolution of a Trader book., pictured on the right, I detail results of historical research then I present my own findings using vastly different methods. Since different methods were used, the numbers won't match but the trends do. When other researchers write that better gains occur from stocks with low book value, for example, that's what I found, too.

So if your results don't match mine, take an aspirin.

Do I have any preference in charting services?
No. I use my own charting package that I wrote. I use to view a stock I don't follow.
Do I have a consulting service?
Nope. And I am not looking to start one either.


Do I take phone calls?
Let's just say I enjoy my privacy. Send an email instead to . I'll be able to give you a more thoughtful answer. The best time to write is on the weekends. I have the most free time then.
Can you help me find a job?
No. That is because I have never worked in the securities industry. I'm a self-taught investor who does not lecture, peddle useless but expensive tapes and other junk to unsuspecting novices. I don't have personal contacts who I can share because my world revolves around email. Sorry.
How can I check on a broker?
The NASD offers a free check of any NASD-registered brokerage showing the background, business practices, and conduct. Here's the link:
I want to buy your courses, webinar, or Skype service
I don't have any courses that I sell nor do I have any webinars or Skype services. I have this website, which is free, and I have a number of well-regarded books on trading or investing in the stock market. If you want to learn how to trade or invest using chart patterns, then start reading.

-- Thomas Bulkowski


Written by and copyright © 2005-2019 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. Warning: I have gas and I know how to use it!