November 2006

Trading Tip:

Larry at the Expo
by Howard Arrington

The material for this article is from the seminar presentation by Larry Pesavento, given Nov. 18th in Las Vegas at the Traders Expo on the methodology he uses in his trading.   I have extracted the text from a video tape I made of his presentation.  The text has been edited to summarize the principles being taught in the seminar.  I also created my own examples for this article that mimic the charts shown in Larry's Power Point presentation.

Bias: Take money out of the equation and you will trade a lot better.  I have always said I would be almost infallible as a trader if you would take off the price, the time, and the name of any chart I was looking at, so I would not know if I was trading wheat, the Euro, the S&P, or IBM or Google.   Because then I would have no bias at all.  Because I have been doing this for 43 years, I have biases, and I have to fight those all the time.  Usually that is when I get into trouble, when I fight them, and so I do a lot of psychological work to try to get around them. AB=CD: This is an uptrend.  You have higher tops and higher bottoms.  What we are trying to do with a forecast tool is to find these spots (there a trend line meets a channel line.)

Point C might be a 61.8% retracement of the AB move, and CD might be a 1.618 expansion of the BC move.  We are trying to take these ratios of  0.618, 0.786, 1.272 and 1.618 and find the point in time where we are expecting a turn, match the price to it, put the trade on and see how it works.   A tool I use often is the AB = CD tool.   I am looking for the CD leg of the move to be equal to the AB leg of the move.   The BC wave will be a retracement that is around 61.8 percent of the AB leg.  Here is an example.

Neural Net Timing:

I use the Neural Net timing tool to tell me when to expect a change in trend, as in this example at Point C.   I look for a retracement to the 61.8% price level and put on the long trade at Point C.  The Neural Net is showing a high probability that the market will continue to rise the rest of the afternoon, which it does in this example.

Even if you did not keep this trade overnight to get to Point D, you still had a small profit buying at Point C and then closing out near the end of the day.  We are trying to find those points where the market is getting ready to trend and you have a nice Fibonacci number to buy off of.   It is that simple.  You do not have to have any oscillators, MACDs, moving averages, or any of those things to make it complicated.  Oscillators are lagging indicators based on things in the past.  With pattern recognition, we are talking about things that occur in the future.  We are looking to Point D for our trade.  Now we are going to be wrong some of the time.

Pesavento Patterns:

(Larry then showed a chart with just a zig-zag line).   You should learn to do a chart like this.  I am going to overlay the Fibonacci numbers on the swings of the chart.  I use the Pesavento Patterns tool to label the swings, like this example.

Look for the Fibonacci numbers given earlier.   You will learn to spot the AB=CD relationships.  In a downtrend, we want to sell into rallies.  In an uptrend, we want to buy on retracements.   Use the Neural Net timing tool to find places where it is showing the time for a bottom or the time for a top, and trade at the Fibonacci price levels.

We are trying to match the price with the time, which is an unusual concept because most people do not have TIME in the future.   The Neural Net curve is going out one day into the future and showing with a high probability when the market is going to turn.   I wish we had a chance to show live trading in today's seminar because you probably would not believe it when you see it for the first time.

Do I make money every day?  No, I do not make money every day.  But I make more money than I lose.  I have more profitable days than losing days.  When I do lose, it is because I misinterpreted what I am looking at, or the Neural Net is totally wrong.

When you are ready to do a chart for the next day, this is the type of analysis that you want to do.  You want to mark all the relationships on the chart that you can see.  Put on the Pesavento Patterns so you see all of the Fibonacci percentages.  You will start to see the bigger patterns of the AB=CD.  The numbers you are watching are the 0.618, 0.786, 1.272, and 1.618.  Any time you are getting ready for an expansion swing that is occurring into a forecast High, and the current contraction swing is arriving at a 0.618 or 0.786, that is when you are looking to be a buyer.

Now I can't show you everything in just one hour.  If you have interest in these types of patterns, begin by learning to mark up a chart with the swing relationships like the Pesavento Patterns.  That you can get right out of my book (Fibonacci Ratios with Pattern Recognition).  That is real easy to do.  But it is going to take you awhile to do it.  I have spent 15 years trying to find this, because I did not know what 1.272 was until 1986 and I had already been trading for 25 years.  So it was a real revelation when I found 1.272.  Then when I found 0.786 a little later.  That was when I wrote the book, and it was named 'Book of the Year' in 1997.  And, it has been pretty popular ever since.

What you are trying to do is find the particular times for the turns, and the prices for those turns will be at a Fibonacci relationship.  Every day when I walk in, I have a place where I want to be a buyer and a place where I want to be a seller in the 10 major things I want to trade.  I wait for the Neural Net tool to tell me when the time is right.  Then I look to see if the price is right where I want to buy it.   Of those 10 market, I will probably have 4 or 5 orders to execute, and I might only get filled on 3.   I might miss the other 2 orders by just a little bit.

Human Nature:

(Audience question:  What is the time frame of the charts?  Are these daily charts?)   This particular chart happens to be a 5-minute chart.  The principles work on basically any chart.  If you took off the time scale and the price scale, no one could tell which chart it was.  The charts basically look all alike.  They have to because we are all human beings, right?  Do you know why you lose trading?   Because you are a human being and you want to avoid pain.  So when you are watching the monitor, you are focusing on the up ticks when you are long and the down ticks when you are short.  If you would stop looking at the monitor and just follow your trading plan you would be better off.  If you put a limit order in and put an alert on so it will either stop you out or alert you at your first price objective, then go back and look at it.  I look at the markets early in the morning for about a half hour just to see how close I am to what I expected.  Then I don't watch it.  I do other things.  If it gets to my price it will beep, and then I will look at it.  Isn't that an easier way to trade than to sit there all day long and agonize over everything that you are doing?  Agonizing is not trading.

When you see these young kids in a trading room, with 400 to 500 in the room, go back 6 months later and there will be a whole new group of young traders.  Because very few people get this right when they do it.  The statistics of people making money in this business, 90% of first time traders are going to lose.  Why?  They don't know what they are doing.  Now the person that stays with it over 3 or 4 years, is going to learn the rules and develop a trading methodology.  You are looking at someone who is very biased about what I do because I have been doing this pattern recognition stuff for a very long time.  I have not found anything that works any better.  I have a pretty good idea there will be a top right there.   (Larry points to a point in the future on the chart which completes an AB=CD pattern at a 1.272 ratio at a Neural Net timing point.)   I am not waiting for an oscillator or anything to go.  Now I have a Neural Net tool that gives me the timing of when to enter the market.

Complexity:

(Audience question:  How you keep yourself from adding more to your methodology?   Are you continually learning and trying new things?)  I learn something all the time.  All I have to do to learn anything is just call Howard.  He will teach me something because he knows a lot more about this stuff than I do.  My stuff that I do is really simple.  And I don't understand computers very well.  I am totally computer paranoid.  I literally panic, so I keep it really simple.  I haven't done anything different than what I have been doing for so many years.  I have looked at other things, but I just don't do anything different.  I see these same retracement patterns every day, and then I use the Neural Net tool to try to give me my timing.  I try to be right 3 days a week, break even 1 day a week, and I'll lose 1 day a week.  

My biggest problem that I have is because I am a strategist for a hedge fund, and I have to give predictions of where I think things are going to go in seven or eight weeks, like when crude oil was topping out.  I got that one right.  And I have to make an opinion, and I have to realize that my opinions are wrong a lot.  So I have to protect myself against myself.  Usually if you are in something 3 or 4 periods (bars) and it is not working out you will be wrong.  Usually this stuff works out right away, I mean almost instantaneously, within 20 to 30 minutes.  If you are not profitable after 20 or 30 minutes then there is probably something wrong, and you should get out of it and look at something else.

Christmas Present:

I'll give you a Christmas present.  If you are really interested in trading, buy Mark Douglas's book 'Trading in the Zone'.   Its a $25 book.  If you will read that book and then reread 3 pages a day for the rest of your trading life, you will do great.  What the book will do is teach you that you know nothing about the markets and that it is not necessary to know everything about the markets.  You will learn that it is about probabilities.  Even though this is a trade, it is just 1 trade out of a hundred trades.  As soon as you lose you have to go right into your next trade and follow your trading plan.  The only thing that prevents me from making money would be if I were sick or they close the markets.  That's the only way I am not going to make money.  I will sit there all day long and trade live in front of audiences.  That's the fun part when you are at a show and can show people in real-time what this is doing.

The Ensign Windows program has a playback system where you can actually go back in time and pull up some charts and update them as if they were live, and that's fun to do too.  But in one hour today I can't do this.  What I am trying to do in this hour is show you what the AB=CD formation is, hopefully you will look at the four Fibonacci levels we talked about, and maybe even take a look at the Neural Net timing tool.  If you are trading currencies and take a look at this, you will not trade without it.  It will get you hooked because it is a real edge in the markets that you have never seen before.  But it is not infallible.  It is only a probability and it is only a tool.  There is nothing incredibly mysterious about it.  It is just mathematics and looking at patterns.  These numbers are in the market everyday and that is why they repeat so much and gives you an edge.  But most people do not do this type of work.

Repetition:

I walked through the Expo hall yesterday watching all the exhibitors showing their charts with oscillators.   There was nobody doing this kind of thing.  They use some Support and Resistance, but no one was looking at all the patterns and how they relate.  Once you see how this swing repeats again here, here and here, you see this basic vibration in the market repeats over and over again.   That is all the markets are ever doing, repeating over and over again.  It only goes up, down or sideways.  It is not the hardest thing in the world.

My little daughter when she was about six, we would get Commodity Perspective every Saturday morning by special delivery because we did not have computers back then.  I would get my charts on Saturday morning and she would help me draw the lines.  I would let her mark up charts like cocoa and coffee which I did not trade.  I would ask her which direction the market was going.  She would look at the chart and say, this one is going up, or this one is going down.  If the market was going sideways she would put her thumb in her mouth and shake her head.   That is what most of us do anyway when we don't know what the market is doing.

(Audience question:  What is your daughter doing now?)   She is a doctor in Denver, Colorado.

(Larry then showed several examples of the Neural Net curve for the next day, and then added the bars to the chart so the Neural Net curve and the bars could be compared.   Correlation was high, but not perfect.  Some examples were shown where the Neural Net curve had a better fit when plotted inverted on the chart.  Please read my 2002 Trading Tips article about Neural Networks.) 

Always Use Stops:

The Neural Nets do not work this well all the time.  That is why you have to use stops.  Let me tell you about my personal trading.  Whenever I put a trade on, 2 things happen.  I don't use electronic trading and I know you are not going to believe it, but I still pick up the phone and call an order desk.  I tried a couple of electronic trades and I screwed them both up.  So I am still picking up the phone can calling my broker.  I am probably one of twelve people who still do that.  My broker will not take the order if I do not give the corresponding stop loss order.  The two reasons why you lose in this business is that you put your stop too close because you don't know what is going on, and the second thing is you don't put a stop in immediately.  If you don't put a stop in when you place the trade or shortly thereafter, just send the check directly to me.  Then you won't have to worry about commissions you pay to your broker.  You have no chance if you don't protect yourself.  You don't know what is going to happen next.  Even with the Neural Nets you do not know what is going to happen next with 100% certainty.  No one knows that.

(Larry then showed a Swiss currency chart where the Neural Net correlation was high for the first few hours of the day, and then the currency broke down when the Neural Net curve was turning higher.  This was a good example where having a stop was necessary.  The Neural Net curve ascended the balance of the day and the actual market went sideways.)

Face Value:

Currencies are the best thing to trade because you are trading the most pure thing in the world.  There is no supply and demand to worry about.  It is just people moving money.  That is all it is.  You do have to worry about when people talk about the money, though.  People like George Saros use the markets to manipulate their position.  You might find that hard to believe, but it is true.

Do you folks know what my educational background is?  I have a Masters degree from Harvard, a Doctors of Jurisprudence from Yale, and I worked for Chief Justice Burger for 2 years before I became a Navy Seal, and I won the Congressional Medal of Honor twice. (Audience chuckles)   Now what I just did to you is what CNBC does to you each day.  They sit there and lie to you.  (Audience laughs)  Well, they do.  No, its true.  They are passing on to you all this information they are getting from these self-serving sources.  After a while you would think they would get the picture they are feeding them this stuff.

When I worked at Drexel, which is no longer in existence because of Mike Milken, they used to have stock broker meetings on Monday mornings and they would have portfolios of stocks that Rothschild who owned Drexel was trying to get rid of, and the joke was put lipstick on this thing and get rid of it.  And that is what the brokers tried to do.  That was back in the 70s.  Now you buy something and it goes up forever.  Back in those days it was difficult to be a stock broker.  You have to be real careful when anybody tries to tell you something.

Exceptional Timing Tool:

(Larry then went through several additional chart examples that illustrated the Neural Net timing.   Most of his examples were using 2-minute bars.  The Neural Net curves are available Monday through Friday for the hours of 8:30 a.m. to 3:00 p.m. Eastern time.   Creation of a Neural Net curve for a 24-hour day is a project that has not yet been accomplished, but might come to pass.)

In my Friday seminar, the Neural Net curves were so exceptional, I had to separate the bars from the curve so you see what the market was actually doing.  One lady said I know you are pulling my leg so she got up and left.  The Euro on Friday actually was so perfect that it tracked all of this.  There were little swings and the correlation was nearly perfect.  When you get days like that you walk out wishing it could be like that every day.  But, unfortunately, it is not.

(Audience question:  In nearly every chart you have shown, why does the correlation at the beginning of the day look better than the last couple hours of the day?)   Yes, the last hour of the day you will have less correlation.  The reason why is because you have gone through several hundred 2-minute bars towards the end of the day, and so the correlation is less.   But we have tested it early in the day, tested in the middle and tested it towards the end of the day.  It doesn't make any difference.  Once the probability curve is made for the whole day it is not going to change very much.

(Audience question:  What do you base the day upon for a currency pair?)  We use the opening time for the Chicago Merc opening time because that is how my research got it started.  Back in 1990, there wasn't any Forex currencies to trade.  Initial tests were done on the Merc products for the pit traded currencies.  Beginning of the day refers to 8:30 a.m. Eastern zone time for all currency pairs for the Neural Net curves.   The curves go through 3:00 p.m. Eastern time.

(Audience question:  Does the Neural Net work better with one currency than another?)  No, any 2-minute chart you want to look at, whether it Gold, S&P, soybeans, or Google, doesn't make any difference.  It is a probability based thing.  It is going to work some of the time and lose some of the time.  It gives you a probability of being right more than it is wrong.  In my personal trading, I am close to being right 70% of the time.  I want to be in markets where they are playing big with lots of volume and volatility.  It is not going to work on thinly traded stocks, IPOs, or stuff like that.  You want volatility. You want liquidity.  And that is what Forex offers you.

You have really great risk control.  You can view things overnight.  When you wake up you can put a trade on and be out of it by the end of the day.  That is an ideal situation without risking very much.  My problems as a trader when I have to make an investment decision for the hedge fund, when crude oil was making highs up there during the fighting between Israel and Lebanon, all the things I was looking at, how was I saying oil is not going to make $80?  $78 was it.  I agonized for months.  Every day they would call me and say why is it not breaking?  I said, 'When it gets above $80, call me.  I don't know when it is going to break.'  Now it is $56 and they want to know where it is going next.  I don't know.  All I can tell you is that was the top.  $55 may be the bottom but I am not sure.

Trade of the Year:

I do a trade of the year every year.  I have been doing it for many years and some of my trades in the past have been buying crude oil at $11, buying gold at $2.50, selling Treasury bonds when they were selling at $127, buying the Nikkei Dow when it was at 7000.  Out of the last nine years I have had 7 winners and 2 break evens.  And this year, I hate to say this, I have a very, very strong negative bias towards the stock market.  If I could stand up on a chair, I would yell and scream at you, but it probably would not make any difference.  But I see very bearish connotations. 

If the INDU goes below 11,500, which is about 700 points from where it is right now, you have a chance to lose a lot of money if you are in the stock market.  The reason behind this would take me 2 hours to go through.  The patterns are there just like we were in the year 1999 and 2000.  Very, very negative.  Whether it works or not, I don't know.  Believe me, I know nothing about fundamentals.  I never read a newspaper.  I don't listen to television.  This is all about what is on that bar chart.  That is all I know.

(Audience question:  How strong was your opinion back in 2000?)  If you want to know, it was in this meeting in San Francisco, March 24th a Saturday, and I had 400 people in the room and the markets had been going straight up.  People were walking into the room with their equity runs of how much every one was making.  I was standing on a small podium, and I was showing them all these patterns, the Butterflys and Gartleys.  Everything was so bearish, much like it is now.  But it seems worse now because we have divergences.  The NASDAQ is not even 38% retracement from the high in 2000, and most of the stocks that were popular then like JNPR and CMGI, these stocks used to be $200 and $300 stocks and now they are nothing.

And I got up on my small podium, and I yelled and screamed at them.  I had 400 people in the room and I said if the NASDAQ goes below the low of last week, get out of all your stocks.  All you are risking is if it goes back above the high you can buy them back.  At least protect yourself.  That week, Business Week, U.S. News and World Report and TIME magazine had the same cover of a bull, one with NASDAQ 5000, Dow Jones 15000,  and the other with something about the NASDAQ going to the moon.  

On November 6th, 2006, on Barron's front page was the DOW 13000.  The Dow has a very poor track record.  You don't ever want to get your picture on the front of TIME magazine.  That is not good.  That is usually when it is over.  It is really not a very good thing.   One person out of 400 people called me to thank me for getting him out of the market.  And as you know, the NASDAQ has dropped 85%.

(Audience question:  Do you feel that same way now?)  I feel more so now because we have so much divergence.  The Transports, and the NASDAQ have divergence.  We do not have the euphoria like back then, but we have euphoria in other markets like the bonds and in real-estate market.  People do not realize how badly they can get hurt in real-estate.  But they will learn.

If you bought a house in Beverly Hills in 1929, you had to hold that house 36 years before it got back to even.   I am not saying things will be as bad as 1929.  When you see a piece of real-estate in New York that is 1200 square feet going for 2.3 million dollars, you know something is not right here.

(Audience question:  What is your prediction for the Dollar?)  I do not like to make predictions like that.  I give a 'Trade of the Year' based on patterns.  If the Dollar goes below 83, and it is at 85 now, there is a chance the Dollar could melt down.  That won't be good for anything.  It hasn't rallied very much, and if it breaks really hard, and the Euro gets above 130, the Euro could go to 150., the Pound could go to 2.50.  Remember, the Pound used to be $7, so a lot of these things could still move.

(Audience question:  If the markets break down, will Gold go to $1000?)  They are totally different markets.  I know you believe that the man who makes the rules, rules the Gold.  Each market is different.  I do not look at inter-market relationships.  I do watch for relationships among the currencies.  But gold investors are different than stock investors.  A gold investor is always a gloom and doomer.  He is always looking for the end of the world, and always has a little bit of paranoia.  But remember a paranoid person only has to be right once.  (Audience laughs)  

Bar Chart Tells It All

All I do is just look at the patterns.  I don't do anything else, don't look at fundamentals, or watch the Federal Reserve.  None of that stuff.  My Masters degree is in Business and my BS degree is in Pharmacy.  I don't know anything other than what is on those bar charts.  I believe the sum total of all buyers and sellers is in the bar chart.  They can lie to you, right?, like Enron.   They can cheat you like K-Mart, Wal-Mart and some of the others, but they can't hide from you.  If prices are going up there are more buyers, and if prices are coming down there are more sellers.  That's all you really have to know.  Sometime between when Enron went from $95 to zero, it had to cross the 200 day moving average, didn't it?  And if anyone ever talks to you about trading a 200 day moving average, don't walk away from them, run away from them.  That thing doesn't work.  That is the biggest crock of baloney.  They have been using that for 60 years, and it still doesn't work.  That's just my opinion, however wrong it may be.

(Audience question:  How do you decide what price to trade at?)  When I look at a particular time, I want to also look at the previous day and know support and resistance points.  If the market is moving down, I might wait for a rally to then sell short.  If it is not making a point I want to hit, I might pass on the trade, or wait for a retracement move to get in.  More often then not, because it is based on patterns, you are going to get a lot of places where the highs and lows are going to come in at your numbers.  It repeats over and over.  Just go back and look at old charts.  Patterns repeat over and over.  Do you know what it is like to wake up in the morning knowing you can beat the markets?  It is a great feeling.

(Expo staff opens the door and announces 5 minutes remaining.)  Oh, thought that was a margin call.  (Audience laughs loudly)