Today’s Subject:

So You Want To Be A Millionaire? – Money Management Makes Trading a Business and Not a Boondoggle - (Part 2A)

Random Processes and Gambling Games

In the first installment of this series, I wanted to convince you that failure in business, whether it is trading, or starting a manufacturing business, is more than just a function of luck. Certainly, there can be failure, and indeed, most people who start businesses without a real business plan or at least an understanding of managing their capital WILL fail. But if you do have a plan which incorporates risk factors, trading, just like building a manufacturing business, is a doable thing. It takes hard work, discipline, research and patience, but it IS DOABLE.

Now, what I want to discuss with you today is that trading, if managed properly, is NOT gambling, but is rather a stream of proper risk management decisions. What is required for successful trading is:

1)     A plan, that is, a strategy involved in making trades with strict rules of engagement.

2)     A strategy that has, at least with some statistical backing, a fairly well defined win-loss ratio and per trade probability of success.

3)     Sufficient capital to make the trading system function properly over time.

4)     A set of money management rules to use in order to scale up or scale down position size, based on the total amount of capital one has, and the most recent wins and losses.

I am going to use some information from a very informative book (which may or may not still be in print) by an individual who is expert in the field, Ralph Vince, from his book Portfolio Management Formulas (First Edition, Copyright 1990, by Ralph Vince, Published by John Wiley and Sons).  In this book, Ralph discusses the differences between random events (which trades happen to be) from events which are skewed in favor of the entity running an event, such as a casino, in a gambling game. I certainly will not be as eloquent as Mr. Vince, but I will attempt to bring the information provided down to a usable level (not that his is not usable, but one has to work with it awhile if one is not familiar with mathematical approaches).

Trading can be described as a random process with independent outcomes.  A random process is a process in which the actual outcomes cannot be predicted, but which can be reduced to a probability statement.

A coin toss is a random process. Unless you are pretty lucky, it would be impossible to perfectly predict which side, heads or tails, will show up on each toss, but if two or more tosses are made in this random process, one would know that the probability of obtaining heads is 0.50 and the probability of obtaining tales is 0.50. I made no statement about the order of these processes; all I did was reduce the event into a statement about probabilities. Each toss is not dependent on the previous toss. For that reason the outcomes are independent outcomes.

On page 16 of the above referenced book, Mr. Vince shows what a fair game of coin flipping (that is, win one dollar if you are correct, lose one dollar if you are wrong) and assuming that the coin is balanced, the probabilities of outcomes would end up this way after 100 tosses.

Standard Deviations                     Fair 50/50 Game      

From Center                                Total Wins (Losses) in Dollars

+3                                                 +15

+2                                                 +10

+1                                                  +5

0                                                     0

-1                                                   -5

-2                                                   -10

-3                                                   -15

What does all this mean, well, first, let’s quickly define what the standard deviations mean.

First of all, this kind of population is defined as a normal distribution. What that means, basically, is that the outcomes are evenly distributed, and in this particular case, the probabilities of these outcomes are all the same, which leads to its even distribution.

Standard deviations are defined as the variability of a population. In this case, our “population” is 100 random coin flips. At one standard deviation, 68% of the variability in the outcomes is supposed to be defined; at two standard deviations, 95% of the variability in the outcomes is supposed to be defined, and at three standard deviations, 99.73% of the variability (almost the whole banana) is defined.

As you can see, one is evenly likely to win a lot or lose a lot in a fair game, because as one drifts from a large negative standard deviation to a large positive standard deviation, the outcomes are evenly distributed.

But now, let’s get nasty with this game. Let’s assume you can run your own legal game of “Bettor’s Coin Toss” where only real men and women can play (I love that macho talk from the old “Bowery Boys” movies (which immediately defines my age).

Let’s say the new rules are that if you miss your guess (it’s tails when you called heads), the house (me) gets $1.05 and if you win, you get a dollar. I now have just a 5% advantage over you, since this is my gambling house (only hypothetically of course) and my game.

Now let’s look at the distribution of outcomes after 100 tosses (also from page 16):

Standard Deviations                     Unfair 5% Advantage for the House

From Center                                Total Wins (Losses) in Dollars

+3                                                +10

+2                                                +5

+1                                                +0

0                                                 -5 

-2                                                -15

-3                                                -20

Now do you wonder why so many people enter Vegas in $50,000 Lexus cars and leave in $500,000 tour buses?

As you can see, when the house can take 5% more than the player when the house wins, the player has to operate at almost 2 standard deviations (that is, he is operating with about a 5% chance (100%-95% at two standard deviations) of winning anything substantial. At the 1 standard deviation level, where the breakeven is, one would have (assuming an equally likely distribution of from -3 standard deviations to +3 standard deviations of outcomes) only a (100-68) %/2 (to cover both good and bad halves of this distribution) or about 17% (it’s closer to about 16 % when you do not round the numbers, but you see the basic picture).

You are a guaranteed LOSER in most gambling scenarios, even with baccarat and blackjack, where one has the highest odds (though less than 50%) of defeating the house. The reason is that the odds are already built to favor the house across a broad distribution of outcomes.

Trading can be that way if the strategy used is at odds with market movement or does not properly use pattern recognition or price trend in the proper way. If a strategy produces a greater number of losers than winners, even if the probabilities of a loss and win are equal (that is 0.50 probability of a win or a loss), you are eventually going to lose all of your capital. That scenario of total loss has a probability of 1.0.

There is no way to determine what the exact outcome of each trade will be, but there is a way to determine how such strategies should perform so that the odds are in YOUR favor. It doesn’t GUARANTEE profits, but it can guarantee, as long as one measures one’s performance as one executes one’s strategy, that one is heading in the right direction. If discipline can be added to that equation, then the trader over time can become successful, and grow his capital stake.

In part 2B, of this series, I am going to attempt to show that even those of you who only attend the church of “buy and hold” are also actually trading, and that your risk of ruin can be substantial. (Don’t worry, I buy and hold some things too, within limits). The same sorts of statistics apply to all forms of portfolio management. In fact, one very famous doctor won a very important academic prize when he defined the risk elements in portfolio management.

After that, I am going to demonstrate a trading model with a specified overall win/loss outcome and apply some of the same statistics that Mr. Vince provides, to show you what the dispersion of outcomes can be even with good trading systems. Then, I will discuss how the trader can maximize his chances for success.

You need to learn how to build stakes and increase positions. Otherwise, trading will never be a business. It will only be a headache, and perhaps heartache, if one does not learn how to manage positions. I think this series will show an aspiring trader how to do so.

Stay tuned.