ROBERT PARDO

“The basic goal of everybody trading futures should be to make the most amount of

money with the least amount of risk.”

Robert Pardo is the author of Design, Testing and Optimization of Trading Systems,

one of the foremost expositions on how to realistically approach mechanical system

development. He helped pioneer many early commercial system testing programs

including Swing Trader—the first software to enable the construction and testing of

customized trading models, and Advanced Trader, which incorporated graphical

representation. His burgeoning system and software reputation led to money

management positions for large traders and corporate firms including Goldman Sacks

and Daiwa Securities of Japan.

For both companies, Robert developed proprietary in-house trading platforms that

featured significantly broader applications than today’s popular commercial programs.

Via their specialized features and in-depth allocation capabilities, Robert and his

assistants were able to create trading models that exceeded all initial expectations.

One of the Daiwa projects led Robert and team to assist in the development of what is

now known as the Chase Physical Commodity Index. Like its competitor, the Goldman

Sachs Physical Commodity Index, Chase tracks a basket of futures the way the S&Ps key

off given stocks.

Robert also endured some disappointing business developments in his career.

Notably, Daiwa had an abrupt management change just as he was about to get involved

in managing proprietary money, which may be part of why, in recent years, he’s backed

away from writing software for other individuals and institutions.

“Now if I do anything for anybody, it’s more the testing of individual ideas,” he said.

“It would not be of any real interest to me to have someone come up and say ‘build me a

trading system’. If I’m going to build them an original system that works, why would I

want to sell it to somebody? I’d just use it for my own trading.”

This wouldn’t necessarily preclude bringing others along for the ride, however.

Robert has recently been managing money under Pardo Capital Limited. It has been

documented in Futures Magazine and The Barclay Managed Funds Report as having a

return in excess of 300 percent since its June, 1999 inception. It was listed six times as

one of the yearly top five performers in Futures Magazine, attaining the number one spot
in 2001.

The interview took place in Robert’s Kenilworth, Illinois home. One obvious facet of

the conversation was how uninterrupted it was by market activity. There were no trading

screens or other market paraphernalia anywhere in sight. If they existed elsewhere on

the premises, Robert was not anxious about checking up on them or his trading progress.

He had the serenity of one possessed of total faith in his system’s ability to hum along

just fine without him.

Give us the simple answer--how does one make money trading?

If you’re going to get involved with this business, you’ve really got to do your

homework. You’ve got to realize that a lot of really smart people have spent a lot of

money and time on it. It doesn’t mean you have to be smarter or better. It just means

you have to find something you know works--something you’re comfortable with. Once

you have that, you’re fine.

Anybody who makes money in this has got some angle, some niche. You have to find

your niche, one that will survive and work for you. Systems are a great way to do that.

How did you arrive at that conclusion?

It may sound odd considering I trade for a living, but I’m rather risk-averse. I don’t take

unnecessary risks, I take calculated risks. I can afford the risks I take.

I had seen the whole gamut of traders. I have had clients who’ve done extremely well
trading on the floor. I started out working for guys who made millions trading cattle. I

saw those who drove the nice cars and had the big houses, so I knew, yeah, somebody

does it. Then I saw a lot of people trading marginally and they‘d make a little bit, lose a

little, and I saw a lot of guys who just lost. So I said, how does this work? A lot of smart

people lose money trading futures, and I decided that I wasn’t going to be one of them. I

was determined to be in this for the long haul, not for two or three years. I was not

interested in going into something half-baked.

When I first started getting into systems, I was persistent, objective and analytical.

I’ve always been willing to say what it is that I do know, and what it is that I don’t know.

If somebody said to me “this will work” I’d say, “well, why will it work? What‘s the

proof ?”

A thoroughly researched trading system will tell you that something works, where it

works, when it works, how it works, what your rate of return is, and what your risk is,

among other things. It allows you to trade a bunch of markets simultaneously which

wouldn’t be possible if you had to analyze them manually and in real-time. Once a

trading system has been perfected, it’s really no work to trade it--it strikes me as the lazy

man’s solution to a hard problem. It can quantify how much money you need to trade

with, and what to expect in the future. To a risk-averse person, that’s all very appealing.

How easy was it for you to get into the actual mechanics of programming?

Even in the early days, I could use BASIC. I had the elements that would read the data,

write the reports and analyze the trades. I could type in a new trading idea in a couple of
hours and test it in a bunch of markets on an Apple II. I figured if you could do that, why

wouldn’t you do it? Why wouldn’t you want to know if it’s going to work?

It seems so logical, and yet people are so resistant to the concept. They avoid doing

the testing work at all, or they or they do it in incomplete or foolhardy ways.

I had a customer in the early ‘80s who had my software and who had a system he

thought was brilliant. I include some of this dialogue in my book where I have a

fictitious dialogue of the programmer talking to the trader. He told me the idea as best he

could and I asked him all the questions that would ensure it was working as he said it was

supposed to work; that it was consistent and there were no errors. And it just lost money

in all these markets.

So what did he do? He got really, really angry and said “you must be doing

something wrong!” So I said “ok, maybe I did” and so we went over it again with a fine-

toothed comb two or three times. It was what he said it was supposed to be, but it just

didn’t make money.

They call this sort of thing cherry picking now. So many people, when they’re

looking at an idea by hand will say, “oh, it worked here, it worked here, it worked there,

and boy, did it work great!” They ignore the fact that it had seven losers before this big

win, and three more losers before that big win. They’re maybe small, but they do add up.

They need to be included in the equation.

What other bad behaviors do people tend to bring to the table?

 A lot of people come to trading because they’re interested in gambling or excitement.

They come for all the wrong reasons, and then they lose money.

They’re unprepared and unrealistic. They think trading is something they can just

jump into. (W.D.) Gann had a great statement in one of his many books. He said a

doctor goes to school for four years, then he’s an intern for four more years. A lawyer

goes to college and then four more years. Why does somebody think that just because

they have $10,000, they’re an expert in trading commodities?

It seems that many people look to futures as a way of propelling themselves quickly

into a significantly higher financial echelon, which is seldom possible. What is a

realistic scenario for someone that hopes to make $50,000 a year? How much do

they need for startup given what kind of return rate, etc?

If you had a decent system and you wanted to make 50 grand a year, you probably need

to trade with between 250 and 500 thousand dollars. If you’re going to make 10 percent

a year, you’re doing pretty good. Fifty thousand off $250,000 is 20 percent a year. My

compound rate of return for the last three years is 40 percent, which makes me the fourth

highest performing advisor for the past three years.

People hear stories about a floor trader who makes $300,000 a year trading with

maybe $100,000. They figure, “well, if he can do it, so can I.” It’s not the same thing.

Floor traders can leverage their clearing houses in a way that isn’t possible for the

average person trading as an outside retail customer. .

 I have a friend who does marketing for the software application Advanced GET. He

teaches classes in how to use the program, and he’s a good trader in his own right. He

says guys are trading with $10,000 and they want to make $50,000 a year. Well who

wouldn’t? You’d have to be unbelievably, unbelievably good, and probably pretty lucky

too. I suspect most professional traders would just laugh at the notion. The kind of skills

required to make that kind of return are far in excess of what it takes just to make

respectable money trading futures.

Anyone trading futures should have a realistic expectation of what they can make.

You need to make sure that the trading capital you’re going to trade with is adequate to

handle the most extreme risk that may and eventually will come your way

Please give us a rough idea of what your program is and how it trades.

It’s similar to the volatility breakout and range breakout ideas, but it’s used and

formulated in a unique way. Then we have a proprietary weighting mechanism that

determines how much trading capital should be allocated to each particular market. It’s

designed so that our chances of losing 20 percent in a month are one in a hundred. That’s

all done with statistics.

Is your approach totally mechanical?

Yes. The only thing that’s not mechanical is when we do the re-optimizations each year;

there’s human evaluation of the results which is not done mechanically.

 I do this in conjunction with Dunn Capital Management, one of the top three C.T.A.s

in the world. They manage about 1.5 billion dollars. At this point in time, I’m managing

their money and some of their customers’ money. There’s an article in Futures Magazine

describing how they’ve cultivated strategic alliances with other C.T.A.s and I’m one of

them.

Are you doing the programming yourself now or do you have partners?

It depends on what stage I’m at. When I get an idea, I used to work it up in my own

program called Advanced Trader, or I’d use other commercially available programs. I

first want to see that it’s actually doing technically what I expect it to do.

I almost always prototype things in the S&P market, because I like it better than any

other market. This is good and bad. It’s good because if I get a model that trades the

S&P well, that’s fine with me. Because the S&P is so volatile and dynamic in its own

right and swings so much, though, what works there may not work in other markets.

That’s the downside, but that’s the way I tend to do it.

Initially, I’ll look at how the idea performs in a small sample. If I’m observing

something that’s supposed to be trading every couple of days, I would expect the S&P

model to trade every couple of days and make a certain amount of money. If it trades

every couple of days and does not make anywhere near what I expected it to make, I’ll

look, for example, to see where the biggest wins and biggest losses were to see what

those were about.

I have a C programmer that I’ve just re-hired. For the more complex stuff, I go to him
or another guy who’s done stuff for me over the years. When we get to the stage we’re

selecting our portfolio and the portfolio weightings, Dunn will code it up. They actually

do the trade management. They’ll do the testing in a somewhat different way than the

way I test. They then code it up for their real time monitoring in their own proprietary

system.

How do people misuse optimization?

A lot of people will say, “Let’s try some moving average idea, and optimize and see

what we come up with.” They may optimize and find a few models that look really good

and completely ignore the fact that most of the rest look pretty bad. I don’t consider

optimization to be the way to find if a model is good. I won’t optimize a model looking

to improve its performance unless its performance is pretty decent within what I

intuitively consider to be a worthwhile set of parameters to begin with.

Do your ultimate systems tend to approach all markets the same way?

It’s the same system for every market, but there are different parameters for different

markets, and there are different weights for each market based on risk. The basic goal of

everybody trading futures should be to make the most amount of money with the least

amount of risk. It may sound easy, but that’s why we go through this procedure of

weighting markets. It’s so we can achieve optimal returns, in other words bet the most

we possibly can with what we consider to be a safe level of risk.

Please expand on the function of weighting markets in a portfolio.

We determine how much to allocate to different markets based on their volatility. We

also look at how much to commit to each model based on its risk characteristics.

You can have a good system and still lose money if you don’t know how to trade it

with the appropriate amount of capital. The minimum thing you have to do with money

management is make sure you don’t overtrade; that you have enough money to weather

the storm. The best thing you can do is figure out the ways to leverage your money to the

maximum degree and still have enough to survive the worst market downturns.

One of the really important things about evaluating weightings and risk is that it is

your prime determinate of how much money you need to trade a system. Most amateurs,

and even some professionals, don’t do that correctly. One of the most common failings

of the amateur is that they way, way overtrade. They take on positions that are way

beyond what they can comfortably afford. It’s fine as long as you’re winning, but if you

get one of those second or third standard deviation losers, then you’re weighted way up

and just looking to get wiped out. It happens a lot.

It’s probably the most common reason why people lose trading commodities. Think

about it. If you buy or sell off the flip of a coin, and you know how to manage your risk

and your profit, you should be able to make something. I have a former client who had

determined, prior to starting with a system, that he would some make money even with a

random trade selection. He called it a money management system—I’m not sure that’s

what I would call it.

 At any rate, you have to ask yourself why people lose so much money trading futures

and stocks when by random selection, you should be able to get a fairly even mix of wins

and losses. Most people don’t ask that question, but it’s a good question to ask.

What markets does your fund include?

Pretty much everything. We trade bonds all around the world...in Asia, Australia,

Europe, Britain and the U.S. We trade short and long term interest rates, stock indices in

Japan Germany, Hong Kong, Britain and the U.S. Also currencies, some metals. Most

of the energies in both London and here. Some of the “exotics” such as coffee and sugar.

This year we’ve added meats, beans and corn to our portfolio.

We review our portfolio once a year. We look for certain characteristics of liquidity

and volatility over a large universe of markets. If a market is not performing according to

our minimal requirements, we don’t trade it. We’ll add it or take it out of the portfolio

accordingly.

The core of our portfolio is always the financials. They’ve traditionally been the big

markets, and I’m sure will continue to do so because there’s so many of them and they

have such excellent liquidity.

We diversify in terms of systems, markets and timeframes. When all my models are

lined up in the same direction, the likelihood that I’ll make a lot of money is very high.

When they’re not, I’m slowly liquidating one side and entering another. I’m not really

getting hurt. Right now, for example, there’s a lack of clarity as to what direction the

markets are going to take. I’m long and short bonds, long and short stock indices, and I
have mixed positions in energy. With those kinds of mixed positions, I’m really almost

spread trading. When they resume lining up in the same direction, I’ll figure to have

winners. That’s the advantage of trading a portfolio with different time frames in

different types of correlated and inversely correlated markets.

Sometimes. though, when you have extraordinary world events, the markets move

in lockstep. A war crisis scenario, for example, usually won’t affect the indices

without also affecting the bonds, energies, currencies, etc. In other words, rather

than being diversified aren’t you sometimes just putting on several trades in the

same direction?

We reserve the right to liquidate positions if we feel something cataclysmic is

occurring. We’ve traded through some pretty cataclysmic events, and we’ve never

chosen to liquidate anything, so I’m not sure what that scenario would actually be.

You can decide that you’re going to trade $50,000 in bonds. Or, you can decide to

trade with one short term model, one intermediate model and one long term model. If

you’re trading your whole position size on one model you’re either going to be right or

wrong. If however, you’re trading a short, intermediate and long term model, it’s kind of

like you get into a third of your position when the first move starts to happen. If it

persists, you get into your second position when the intermediate term kicks in, and if it

persists more, your long term model kicks in.

It’s not that we’re doing that to build our confidence, because we’re very confident

that the models work. A lot of expertise went into building them, and they’ve been very
successful. But what it does do is to provide another very effective level of

diversification. If we just traded one U.S. bond model, our risk would be much higher

than trading different bond markets in different time frames all around the world, which

is what we do. The models on different time frames on different weightings are what

really adapt with a finer granularity to the economic climate. If things really don’t move,

the models won’t kick in as fully in the portfolio.

If volatility were to get progressively smaller and then kind of stayed in the smaller

range, the models would adjust their entries based on the existing parameters. It’s not an

arbitrary attempt to adjust to it. The models adjust because of the way they’re designed.

So what you have is more or less dynamically keyed somehow --to expanding-

contracting ranges, for example.

A better way to put it would be the nature of the model is dynamic. It’s easy to measure

volatility. What’s hard to do is to effectively integrate the volatility measurement. Some

years back, there was a great deal of interest in what was called equity curve trading.

People would keep track of a moving average of their equity curve. If the curve took too

much of a dip, they’d stop trading and then they’d wait for something else to tell them it

was time to take a signal.

I did a lot of work with that when it first became an area of interest and found that it’s

hard to integrate an equity curve with a model so that the tempos are in alignment.

Generally what tends to happen if you’re trading short term is, a big move, say in the

S&Ps, could be over in two days. If your equity curve trading has slowed things down to
where you don’t get a signal, or you don’t kick in until it’s halfway over, it tends not to

be productive.

Similarly, I’ve found that trying to make decisions on what to trade based on volatility

changes tends to get in the way of the models. With what we have now, I’ve been willing

to sell Swiss Francs for the last four or five days, but because it’s been so volatile, I need

a very big move for that to happen. [Because the volatility widens the entry levels]. It’s

not trending that much right now, so I’m not in it one way or the other. However, if it

stays quiet for four or five more days, the entries will get closer and closer to the market

and I’ll be more likely to get kicked in. The models trade dynamically because of their

design.

How do you differentiate a large but acceptable losing streak from one that negates

your perceived boundaries? Have you ever had one go beyond accepted

parameters?

Yeah. The day that the U.S. said diplomacy is dead in the Iraqi situation we were long

bonds everywhere, short stock indices everywhere, long energy and short the dollar. We

had the worst day in our trading history. We just decided that that’s what happens once

in a while. We didn’t feel there was anything out of the ordinary.

Here’s what I think the primary consideration is. Let’s assume you did everything

correctly. You tested on as large a data sample as you had available, and you came up

with your max drawdown. Even going way back, a lot of people maintain that what you

should do is, assume that your max drawdown is understated. Assume that it’s actually
going to be at least twice as bad. Certainly in our S&P models, drawdowns have gotten

bigger over time, but then, volatility has also increased. The way you distinguish

whether the increase is acceptable is, the current drawdown verses the current volatility

should be somewhat proportionate to the original drawdown and the original volatility. If

the volatility and drawdown both doubled in size, your profit should also approximately

double. If that’s what you have, that’s ok in my book.

If your profits remain the same, volatility has not really changed, and you suddenly

have a drawdown that’s twice your original one, it doesn’t mean the model has failed, but

it does mean you’d better find out why. It’s a potential red flag .

Do you give equal weight to all your historic data, or do you think more recent

market activity is more valid than what happened in more distant history?

What we’ve discovered over our years as model builders is, to get optimal performance,

you need to know that your model is good overall on the biggest possible sample that

exists. But also, you want to pay a little more attention to what’s going on in more recent

times. Because really—who cares what the S&P did in 1983? The volatility is so much

higher now than it was then. Still, I personally wouldn’t want to see a model that made

tons of money in the last two years, but did not make money in the last 14 or 15 years.

That wouldn’t give me a lot of comfort.

So you’re kind of bridging the two. Recent price action is probably the most

relevant, but you’d still like to see the bias hold up to some degree throughout the
observable past.

Right, it’s a fine line.

Would you also like to see confirmation in other markets?

I’m a little bit different than a lot of people on that count. Let’s say somebody came to

me and said “I’ve got a bunch of different models that trade a bunch of different markets,

and they’re all different, but they’re all really great.” If I looked at the models and saw

they were sound, it wouldn’t bother me a great deal that they wouldn’t work in other

markets. Not as long as I felt they were sound in the market they were designed to trade.

Individual market characteristics do exist. When I consulted for Goldman Sachs, they

were—and still are-- very big in the energy markets. From this work, I learned there is

probably more information available about the fundamentals of oil than any other

market—probably more than you could begin to imagine. So if somebody had a model

that traded crude oil unbelievably well using information that was unique to crude, it

wouldn’t bother me as long as I understood the rationale behind the model and knew that

it was sound.

I don’t actually do that, but I know people at Goldman Sachs who were beginning to

work with those kinds of models. They used what I would characterize as technical

models built with fundamental data. I think that’s a fertile place for exploration.

So you acknowledge the possibility of effective systems keying off a given market’s
unique characteristics. But as you say, you almost never work with such targeted

ideas. Given your more universal approach, what signals red flags in your

research?

Since I know that our models are not based on the characteristics of any particular

market, it would not be a good thing in my mind if we tried them on 50 markets and they

worked in only five. I expect to see them be effective in all markets and perhaps better in

some than others. I think they do better in some markets rather than others not because of

the models, but because the better markets are the ones that are moving. If markets don’t

move, nobody makes money, except maybe the locals.

Getting back to the topic of your periodic re-optimizations: do markets change their

historic characteristics over time in meaningful ways?

Yes, they do change in meaningful and impactful ways. Up until 1990-92, trading

currencies was like picking fish out of a barrel. It was really easy for anyone with a

system that was worth anything. Shortly after, there was a big change in the markets and

trading currencies became really difficult. If you had tested your model on the last three

or four years of data in a couple of currencies, you might think, “well, this isn’t my road

to retirement.” It’s easy to fool yourself. That’s why it’s important to look at the bigger

picture—bigger time frames, more markets.

Around 1980, T-bills were an incredibly volatile market; a primary source of

speculative activity. You had a situation where the prime rate had hit something like 20
or 22 percent and short term interest rates were even ahead of that.

At one in point in time, the business community was saying, “we don’t care what

prime is, just make sure it stays there for six months or a year so we can do some

meaningful planning.” The market rallied about a thousand basis points in two or three

months with big volatility, yet the biggest pullback the whole time might have been 75

points. It made some people very wealthy very quickly. Years before the classic bull

markets in the Nikkei, S&P and Nasdaq, I would consider the rally in T-Bills in this

period was probably the mother of all bull markets.

After that rather staggering runup, the T-Bill market kind of flatlined. I think it still

exists as a futures market, but at some point it stopped trading in a meaningful way. I had

a business associate who was a floor trader and order filler during this period. He made a

record amount of money during the volatile years, but suddenly he was complaining that

he didn’t know what to do. I said, “If I were you, I’d go to the S&P pit. I don’t know

what happened, but the T-Bill market is dead.”

Some system traders try to anticipate all that in their models as well, through

volatility filters and other dynamic tools. Obviously, they’re trying to avoid the

human element even in the step of assessing viable trading environments. Do you

re-do systems as a direct response to changing markets?

With my approach, I re-optimize at periodic intervals anyway, but also, with the way we

trade now, that market would be beneath our radar screen for volatility and probably

liquidity.
That’s what I’m getting at. Doesn’t the system self-correct without you having to

make a decision as to whether a market has gone dead or become ineffective in some

way?

Actually, we deal with that issue in a number of ways. The systems use volatility as a

primary characteristic. If volatility changes and stays changed in meaningful ways, it’s

going to either increase or decrease the size of the entry point. But also, our second

approach is to re-optimize the model every year. If there’s been a significant change in

volatility one way or the other, we may change our parameters.

The third way we correct is, we look and see if the market tradable. We don’t care if a

market goes like this or like this or like this [traces various bull and bear trends in the

air]. We just don’t want to see one go like this. [Makes a flatline gesture]. If it goes flat,

there’s no economic interest in the market, and you really can’t trade it profitably.

Liquidity tends to follow volatility. If you have an illiquid market, you will have very

few players. So our last line of defense is, we’ll examine whether there is enough

movement in the market to warrant being involved, which is something we’ll look at in a

number of ways. There’s always a limited amount of available money. Is somebody

somewhere else offering a better return?

We more or less have enough to trade everything. We’ve never been faced with a

situation where we didn’t have the money to trade markets we wanted to trade, though I

suppose that’s a possibility. But the bottom line is, if Commodity B is marginal, and

Commodity A is looking really great, why trade Commodity B when you can trade
Commodity A? We use a number of check lines to arrive at that.

One reason the S&P has held so much interest over all these years is because it has

great volatility and great liquidity. Almost all the stock indices around the world have

similar volatility, but they don’t all have the same degree of liquidity.

Why do systems work? What market characteristics do you think they exploit?

There’s no generic answer to what characteristics in a market a system exploits because

different systems exploit different characteristics. The ultimate reason they work is

because they give you a mathematical advantage. A good system by definition has a

positive expectancy. A good system will work because it’s consistent. It will do the

same things day in and day out.

Does it have something to do with market inefficiency?

I’ve never believed the markets are efficient. but it kind of comes down to what you

mean by efficient. If efficient means that a market is always at the perfect price that it

should be at in any particular moment, I don’t think that’s true. The systems I use take

advantage of volatility, and the fact that there are persistent trends in the market.

A system can only work if it catches runs. Runs is a more meaningful term in the

intraday market. In the larger timeframe, you’re looking more at trends.

There are things that occur in intraday data that you won’t see in interday data. It’s

fairly unusual, for example, to find five up bars in a row in a daily or weekly chart. It
rarely happens, yet it’s so very common in intraday bars.

The trend in a daily bar may be up, down, up, up, down, down. You have a lot of

swings. Ultimately the only way to make money trading is to exploit some kind of trend,

regardless of bar size, and to hold onto it long enough to make money overall.

To do this effectively, you have to keep your risk measurably constant in some kind

of way. Many people will, if a trade looks good to them, put on too much. They’ll risk

more than normal in the trade. If they’re nervous about it, they’ll put on too little even

though they can afford a lot more based on risk or volatility.

In a system, risk is uniform and constant. I re-optimize models periodically because

conditions and volatility change. You have to adapt to that to get optimal returns.

Generally, though, we’re risking the same tomorrow that we are today. Most people not

only will vary their risk a great deal, but they’ll get very skittish when they actually get a

profit.

Which of course violates a prime trading directive: cut your losses short and let

your profits run.

Exactly. I can give you a story that exemplifies the way the typical amateur trades.

Years ago when I was a broker, I had a customer who was a very bright guy. He finished

second in mathematics in his country, and got a PHD in chemical engineering. But he

was a horrible trader. We had a situation where he’d taken some losses in lumber based

on these haphazard risk and reward things that people engage in. Finally, he got a

winning trade on. Instead of letting the thing run, he kept on moving the stops virtually
every 10 or 15 minutes. He gave the market no room to swing. He had managed to sit it

out long enough to let a loser turn into a winner, but he was so anxious about it that he, he

squeezed it so hard that he wound up making $200 on it. His previous losses were like

$700, $800 and $900.

That’s typical of the amateur. They’re willing to take more risks than they actually

should, and when they finally get a winner going, it seems so unusual to them that they

don’t give it a chance to run.

In futures trading, one of the things that I have found is, in general, you’re better off

letting your profits run as far as they can even though that can be a little uncomfortable

sometimes. In contrast, some people maintain that the best thing you can do is to put up a

target order and take uniform profits. I have found that that kind of thing only works if

the system is highly accurate. Most systems aren’t that accurate.

So you’re not really in favor of profit targets?

With the systems I’ve used over the years, targets have made the models perform less

profitably. What you might find is, a model that trades a great deal and is highly accurate

could be made more accurate by targets. If you’ve got a model trading 60 percent

profitability and it’s a good model, the targets might add 5 to 10 percent to it. So then

you’re right seven times out of ten, and for a lot of people, that’s a big difference. They

really like being right that often. But that’s another reason people don’t trade well.

They’re more interested in being right than they are in just making money.

I’ve never really cared about accuracy in a model. A lot of people do, and obviously,
the more accurate the better, but I’ve never made that a primary focus in building a

model. The models that my current platform started out with were in the area of 45 to 48

percent accurate. They still made a lot of money because they let profits run, and the

losses were very manageable.

It’s a great irony to me that with no effort to get accuracy, my current models are in

the 60 to 65 percent accuracy range.

Have you always taken such a logical investing approach or, like most of us, have

you had experiences where you’ve learned hard lessons through trading in counter-

productive ways?

I once had a customer who traded bonds and options on the floor. We became good

friends, and I showed him models that I was very comfortable trading in S&Ps, bonds,

crude, soybeans, and silver. We agreed to pool some money and trade them.

We lasted about a month. In that time period, he insisted on picking the signals he

wanted to trade. He knew better than the system.

The long and the short of it is, we lost money in the account. I said “Tony, this is

stupid. You’re making my systems look bad with your stupid cherry picking. I’m not

going to do this anymore.”

The models, of course, made a ton of money in that time. They probably would have

doubled the account. At that point in time, I didn’t know enough to say to him “you can’t

do that.” I figured he was a successful trader, he should know better, but that was not the

case.
 I saw him a couple years ago and he still makes a comfortable living. He’s still kind

of playing games with systems though. I’m not saying no one can enhance a system,

because I know some people who, in fact, do. But they do it in a systematic way.

They don’t do it according to what makes them feel good.

Right. They’ll look at a buy signal relative to another indicator and figure the odds will

be a little better. It’s something they observe. I haven’t seen too many people with a

golden gut. That’s why I was determined that I would have a method that would be

sound and do all the things I wanted it to do. The rest would take care of itself.