Testing entries with random exits?

Paul King, March 27th 2007 

There has been some recent discussion on the trading forums I frequently read about random entries and exits.  Here is my opinion on this subject.

 

I believe that an entry's predominant characteristic is setting trade frequency and entering trades at times where there is the least chance of stagnation of price movement and most chance of continuing high volume. This helps to minimize slippage and also reduces the chances of a "flat line" trade that uses up capital, takes risk, and achieves no return.

If there is any predictive nature to entries in any market, one has to think about the likely time period over which the entry pattern is predictive of future prices. If this time period is shorter than the average duration of trades, it is practically insignificant to your overall trading results.

I would/could consider trading a random entry assuming it was chosen from a liquid universe of instruments and I had the choice whether to take the trade or not. I would never consider trading a random exit under any circumstances. This to me indicates my beliefs about the relative importance of entries and exits. Other traders may have a different opinion.

Attempting to test an entry by using some kind of random, fixed-bar exit is not really representative of real-life trading.  Further, a fixed bar random exit basically takes unlimited risk between entry and exit - price could go anywhere and you would not exit. This overestimates the returns of any trading method compared to "real" trading where one normally limits risk during the trade in some way (unless you want to blow up your account at some point).

Broker selection, liquid instrument selection, setup, entry, exits, and position-sizing all combine to create a particular instance of a trading system. In my opinion setup/entry influence your trading results the least out of all components, but that doesn't mean they have zero effect. Trade frequency is much more important in my opinion than some short-term predictive anomaly associated with an entry.

I know this goes against the grain with most traders, but isn't that what the fabled "search for the holy grail" of trading is about - finding so-called "accurate" entry signals? Since I believe that winning percentage is determined by exits not entries, "accurate entry signal" is an oxymoron to me.

 

Testing trading methods as a consistent "whole" that most closely matches the way you are going to trade for real is an effective approach that does not put too much or too little emphasis on any one trading component.  My trading system component model is a useful model, but it's not the only way to think about how trading methods fit together.  Examining the effects of changing individual components of a trading method often yields useful (and counter-intuitive) results. 

 

Not much time left over for blogging

Paul King, March 23rd 2007 

As you may have noticed I haven't written a blog entry in a couple of weeks.  I'd take it as a good sign that I've been pretty busy recently.  A few of the things keeping me occupied:

• Having surgery on my shoulder and recovering with physical therapy.
• Taking a trip to Boston.
• Having 2 articles published in Futures magazine.
• The release of my first book.
• Helping my financial advice clients.
• Working with a UK firm to develop a US equity trading long/short incubator hedge fund trading program.
• Starting a new trading discussion forum on this site

If things quieten down a little I'll be able to get back to the at-least weekly blog, but for now I'm just happy to be busy on other things.

 

Keep checking back, you never know when I'll be able to start blogging again.

 

_______________________________________________________________________________________

 

 

"Long Option" or "Short Option" trading program profile?

Paul King, March 6th 2007 

It's been estimated that 80% of option contract expire worthless.  Some people use this as evidence that this kind of trading is a loser's game.  But what if the other 20% of the option contracts expire so far in the money that it more than makes up for the 80% losers?

 

This "long option" profile of many small well-defined losers, and a few outsized winners, is typical for a trend-following trading program.  Discipline to stick to initial stops keeps losers small.  Trailing stops that protect profits, but also let them grow big creates outsized winners when markets are trending.

 

A typical winning percentage is between 20% and 40% which means you have to endure the long slow torture of "bleeding to death" when you go through a prolonged losing steak - even though each loser is relatively small.  On the flip side, you are not taking a significant risk of "blowing up" by having a losing trade that represents multiple year's worth of winning ones.

 

The opposite side to a tend-following trading profile would be a "short options" profile to your trading.  This is characterized by a high winning percent due to lots and lots of small winners, but with a (perceived) small probability of a catastrophic loss every time you trade.

 

Personally I believe that a well-defined profile of "bleeding to death" is far more psychologically manageable than having to deal every single day with the possibility of giving up your last 5 years of profit (and maybe your initial capital, and more besides).

 

The only difficult thing about trading with a "long option" profile is that the consistent, positive returns with lots of winners of a "short option" strategy looks superficially way better than the "long option" strategy - but only until you hit the inevitable "blow up" point.

 

Do you want the huge future price swings to be with you or against you?

Do you want to make money or feel good about your trading because you win a lot of the time?

 

Choose the trading profile you are designing into your trading program carefully.

 

_______________________________________________________________________________________

 

 

Equity markets are not a "zero-sum" game

Paul King, February 26th 2007 

Commodity futures are a zero-sum game.  Each contract outstanding is between 2 counterparties, one long and one short.  Every dollar one makes is a dollar taken from the counterparty on the other side of the contract.  The transfers won or lost are applied daily in "real cash" rather than some intangible notion of value.  When you run out of cash and can't cover margin, your position is closed.  Wealth is neither created nor destroyed in the environment, simply transferred.  When trading costs are included it becomes a negative-sum game overall.

 

Equities, on the other hand are not a zero-sum game.  If there is no short interest in an equity, and it makes a new all-time high, every single holder has a profitable position.  This is "equity" that did not exist before, and has not been taken out of someone else's trading profits.  It has simply been "created" by the price going up due to demand.

 

Because there is no physical constraint on how much prices can move up or down (for individual equities anyway) this leads to an environment where trends can, and do, build up significant "momentum" that is far removed from any "real" concept of value.  We call this a mania and it's what profitable trend-following is all about capturing.  The reverse is true in a panic which is simply a mania in reverse.

 

What practical applications does this have?  Consider these:

• Trends can and will travel farther, and persist longer, than any financial model anticipates.
• Always position-size so that when prices go into a mania or panic you don't blow up your account.
• Always assume current volatility is low compared to what it could be tomorrow.
• Never believe a price has "gone too far" and will "revert" to some kind of "mean".

Where is it specified that equity markets have to be "mean reverting" anyway?

_______________________________________________________________________________________

 

 

The PMKing Trading Forum is launched

Paul King, February 21st 2007 

Due to the release of my first book, The Complete Guide to Building a Successful Trading Business, I've been getting more questions and comments from people than usual (all positive so far I'm pleased to report :-).  I anticipate that many of your questions are going to be repeated, have lots in common, and be interesting to all my customers, clients, and trading friends.

Therefore, I have decided to launch a trading forum where everyone can ask any questions they like about me, PMKing Trading, The Complete Guide, our other publications and services, or anything else about trading you think I may be able to help you with.

Obviously I have limited time and resources like everyone else, so I can't promise any kind of "turnaround time" or "service level agreement" for responses, but I will try my best to answer any questions people have that I feel will be of benefit to everybody.  The needs of my paying clients will always come first, but I am hoping to be able to respond on the forum in a timely manner (I can type pretty fast :-).

Happy trading and I hope to see you online soon.

Visit the forum here.

 

_______________________________________________________________________________________

 

 

Book Review: The (Mis)behavior of Markets, by Benoit Mandelbrot

Paul King, February 12th 2007 
 

 

Now I do understand, and it was a confirmation of some of the conclusions I had already come to by trial, error, losing money, and a lot of research and deep thinking.  Reading this book has helped me formulate my own "Stochastic Price Change Model" which will be the subject of a future eBook or article and has helped me reach a different perspective on how to more effectively test the robustness of trading methods.

 

You may have heard of Mandelbrot from his work on fractals, and in this book he applies that mathematical model to explain exactly why all the existing financial models (the Black Scholes Option Pricing and the Capital Asset Pricing Model to name but two) are not as useful as they are touted to be, or just plain dangerous.

 

"Prices are normally distributed", "fat tails are simply anomalies", "prices are independent of each other".  These are all fundamental assumptions of the mainstream models of the market and of risk taking.  Mandelbrot challenges all these assumptions and presents a more useful (or at least accurate) fractal view of markets which may change your perception on how much risk you are really taking and how likely "unprecedented" events may actually be.

 

I immediately put this book on my trading "must read" list.  The implications of what it discusses will continue to be influential in my own trading and my trading beliefs in the future.  Read it with an open mind and see what happens.

 

Purchase the book from Amazon. 

 

See our other recommended trading books here.

 

_______________________________________________________________________________________

 

 

A simple percent-based stop is not adaptable enough

Paul King, February 9th 2007 
 

 

Consider the situation where you use a, say, 20% change in price as an exit point.  For a position that moves about by 1% per day, a 20% stop loss represents about 20 days "worth" of movement.  If you have another position that moves by 0.5% per day then a 20% stop loss represents 40 days "worth" of movement.  You are actually taking different amounts of risk in both positions because they have different daily average daily movement, or volatility.

 

Maintaining constant risk across all positions helps to minimize the variability of your results, so having a volatility-based exit point rather than a fixed percent is more effective.

 

A reasonable way to estimate volatility is to use the average true range (ATR) over a recent period as a proxy for current volatility.  You could simply use the high minus the low as a guide, but True Range takes into account gap openings too.  You can then use a fixed multiple of the ATR to use as your exit point.

 

In the example above it would mean that the stop for the position that moved 1% per day, would be twice as far away as the stop for the position that moved 0.5% per day.  This would mean that you should have a position half the size in the higher volatility instrument.  In this way each position in your portfolio would be taking the same amount of risk, with the same probability of being stopped out.  This helps makes your results more consistent and less volatile than a fixed percent stop would.

 

 

_______________________________________________________________________________________

 

 

The Complete Guide is now available

Paul King, February 6th 2007 
 

 

Click here to purchase the paperback edition.  Use the contact page on pmkingtrading.com if you have any questions, or want a hardcover version.

 

I have articles in the February and March issues of Futures magazine that are adapted from the book just in case you want to check out my writing style before buying.

 

The book should be appearing on major online booksellers over the next few weeks, but right now, Lulu is the only place you can get it.  Thier innovative and creative publishing model minimizes time-to-market and maximizes the author's revenue, so that's why I'm able to sell it for less than twenty bucks, and also publish in paperback immediately.

 

I sincerely hope this book can help people improve their trading.  Please let me know your thoughts, impressions, and feedback (good or bad) about the book if you purchase it from Lulu.

 

_______________________________________________________________________________________

 

 

Buy and hold is an optimal strategy, but not for you

Paul King, February 2nd 2007 
 

 

• Market timing is impossible
• Buy and hold is the only sensible investment strategy

 

These two beliefs basically get people to be 100% invested at all times and hold on for dear life whatever happens while their "investments" go down the toilet.  The interesting thing is that this is an optimal strategy for investments; not for the clients, but for the businesses that manage your money.

 

If you are making money by charging, say, a 1% per year asset management fee, then the best way to make the maximum fee is to have all your clients (as a group) take the maximum possible risk with their cash.  The way to take maximum risk with your cash is to be fully invested and have no "uncle point" when you will take your losses.

 

This means that some clients may make huge returns, some may lose all their investment, and some may make or lose a little.  Overall you will maximize the asset management fee and have to find new clients to replace the ones who do lose all entire investment.  This is a very similar model to the life insurance industry that can set premiums to make sure they collect in more than they have to pay out based on an actuarial analysis.  They don't care or know when each client is going to die,  but they do know how many are likely to die over any given period of time.

 

The same goes for a large collection of clients with an asset management fee model.  The company has no idea which clients will lose it all, just that for the lost asset management fees on the ones that "blow up" should be compensated for by the ones that have lucky high returns.  Overall they are simply maximizing their fee by maximizing the total risk taken across all clients collectively.

 

So, if you have "investments" with a compmay that uses a simple asset management fee compensation model, you'll now know exactly why they are telling you to be fully invested and hold on for the long term rather than actually helping you limit your risk (by having an exit strategy); it's in their best interests.

 

If you want a model that aligns the company with the investor then you need some kind of shared profit model which is not allowed by SEC regulations unless you are an accredited investor with a high net worth investing in a hedge fund.

 

My advice:  make sure you have an exit strategy that limits your risk for all your investments so you don't become part of the collective herd maximizing the asset management fee for your financial company.

 

_______________________________________________________________________________________

 

 

Trading in the now

Paul King, January 25th 2007 
 

 

_______________________________________________________________________________________

 

 

Procrastination can be very destructive

Paul King, January 17th 2007 
 

 

Procrastination takes many forms, some more subtle than others, and the key thing is to find out the reason behind the procrastination.  If you haven't written down objectives for your trading why not?  What's stopping you?  If something else is always more important then maybe you're not that committed to trading.

 

If you seem to be in an endless loop of learning and "evolving" your trading towards that "perfect" system (that doesn't exist by the way), then what's holding you back?  Are you afraid of failing if you actually do some trading and it turns out to be bad?  Are you comfortable risking the cash in your trading account?

 

If you've found time to research, develop, and test a suitable trading method that fits your personality and meets your objectives, what's stopping you implementing it?  Are you trading with money you can't afford to lose?  Do you fear failure?  Do you fear success?  Do you prefer to stay in your comfort zone where you have a beautiful tested trading system that you don't want to subject to the stress of trading real money in case you find out the system (or worse still, the trader) is a total loser?

 

The objective of trading is to make money within your tolerances for risk.  The only way to achieve your objectives is to actually trade and take risk.  If procrastination is stopping you from doing any of these things then it's time to sit back and think about why you are really sabotaging your trading by failing to make decisions that get you closer to your goals.  Only you can stop you from procrastinating but that also means that you are the only thing between no trading method and a successful one.

 

Stop reading this blog and nodding now and go do something that's useful to meeting your trading objectives.

 

 

_______________________________________________________________________________________

 

 

Brief update and other excuses

Paul King, January 14th 2007 
 

 

As well as the normal Christmas and New Year stuff, I've had a Birthday that's rapidly approaching 40, and also had surgery on my right shoulder to stabilize it due to a recurring disclocation problem that started with a nasty fall ice skating in Central Park a few years ago.  Life with a sling on your right arm is interesting - who knew putting one your own socks was such a tricky maneuver?

 

As well as all that, I had a total computer meltdown of the hard drive (and backup drive) of my primary trading computer.  That can be a good stress test of your contingency plan.  All is back to normal now but not without a stress test of my patience too.  I hope you all have a good contingency plan in place for your trading for that kind of problem that absolutely can't happen.

 

Business related items are listed below:

• PMKing Trading has moved office (I know, when did I find time to do that too?)  Our new address is 5 Park Street, Suite #2, Middlebury, Vermont.
  
• I recently signed a contract with a UK firm to start and manage a US-based incubator fund.  The "place holder" web site is at http://www.emblumoven.com.  More details to come when I have them.
  
• I have finished my first book "The Complete Guide to Building a Successful Trading Business".  If you can't wait until global distribution is approved and want a pre-release copy right now please contact me.
  
• I have another article in the February 2006 issue of Futures Magazine and this one is about exits which are the most important aspect of trading (after position-sizing) so try and check it out.

 

Normal blog entries should resume as soon as things are back to normal round here (whatever that means).  I hope you are all having a successful and prosperous start to 2007.

 

 

_______________________________________________________________________________________

 

 

Product Review: Why Traditional Technical Analysis Doesn't Work from InvestorFLIX

Paul King, November 15th 2006 
 
In this 1 hour DVD by Grant Noble, there are some useful ideas about why traditional technical analysis doesn't work.  The presentation is 10 years old, and the main body of what is presented isn't that useful, but there are two main ideas that are definitely worth thinking about.  These are:

• All charts should be adjusted for inflation and base-currency rates
• Technical analysis applied to fundamental data can be effective

The whole presentation lasts for about an hour and doesn’t once mention position sizing or exits in much detail although he does say that knowing when to take profits is a key part of trading, but doesn’t go on to say how you should do it.

 

Two beliefs that are prevalent through the whole talk are that a) off-floor traders have huge trading implementation expenses to contend with and so must trade longer-term than floor traders and b) successful trading is all about making accurate predictions and having effective entry signals.

 

Another quirky idea is that it’s OK to have 12 or so indicators in your trading so you aren’t dependant on any one indicator being right.  Which 12 to use is not covered and nothing is stated about what to do when they all contradict each other.

 

The discussion of historical cycles in the S&P 500 makes the normal mistake of “spotting” a pattern that has only occurred 3 times over the last 150 years and extrapolating from that what will happen in the future.

 

I haven’t read the book that Grant touts in the presentation (saying that this presentation is not a commercial) and I don’t feel inclined to after watching this talk.

 

My advice is to think about adjusting your charts for inflation, and possibly using some technical analysis techniques on non-price data. These may both be areas for useful research, but you don’t necessarily need to watch this presentation to find that out.

Visit InvestorFLIX

Read our other InvestorFLIX reviews here