You can download this podcast in audio format at the iTunes Music Store by clicking here.

DONATE

Hello and welcome to Financial Audio, an information series providing listeners with detailed and tactical guidance on today’s complicated financial world.  My name is Patrick and I’m your host.  You can find written versions of these podcasts at FinancialAudio.com and I encourage your candid feedback at the same location.  Today, we’ll be looking at automated trading platforms so let’s get started.

The Market Timing Systems we discussed in the last chapter are all built on automated trading platforms.  These platforms take daily price-volume action and calculate a variety of sophisticated technical indicators, yielding automated buy and sell signals.  With computer technology becoming more and more accessible to the average person, experts in technical analysis and mathematics are transforming their expertise into these automated systems – and you and I can too.

Obviously, it takes a particular type of online brokerage platform to execute such scripts.  E*trade or Charles Schwab, for example, probably don’t have much support for automated platforms but there ARE others that specialize in exactly that approach.  For frequent traders, the most popular platform is called TradeStation.com.  It incorporates a sophisticated charting platform with virtually endless technical indicators and a brokerage capacity so it can facilitate your trades directly.  Well, as it turns out, TradeStation also specializes in automated trading.

As you can imagine, many automated trading systems generate frequent trades so it was a natural fit for TradeStation and they pioneered one of the most popular computer programming languages designed specifically for automated trading scripts.  It’s called Easy Language but I guarantee it’s far from easy.  I’ve struggled with it but am slowly making progress.  The point is it’s designed for exactly this purpose and can be used to build your own custom script with your own buy signals and your own sell signals.  You can also incorporate money management modules so the system will invest different amounts depending on market conditions.

This chapter is obviously geared for those interested in automated trading platforms but I encourage all of you to listen even if you’re not directly interested.  There’s a fascinating discussion here and it certainly serves to explain the foundations of Market Timing Systems and that understanding may increase your confidence in such programs.

I’d also like to say, quickly, that TradeStation.com provides two extensive manuals on Easy Language.  They’re in PDF format and available for free download on their website.  They’re quite good and will give you an excellent introduction to the language and how to use it for your own platform.

For the most part, automated trading platforms are built on oscillators of one kind or another.  But there’s an inherent problem with oscillators and they often generate false signals as a result.  Let’s use the example we used right at the beginning of this series; an example involving a stock trading between $20 and $25.  So the price trends up for a while, hits $25 and starts coming back down again.  Eventually, it hits $20 and starts working its way back up.  For the most part, the price is somewhere in between.  Only from time to time does the price get all the way up to $25 or down to $20.

If you took the daily closing price of the stock over a couple months and plotted them on a graph between $20 and $25, the highest concentration of dots would probably be around $22.50 and there would be less and less on either side and tail off entirely at the edges; at $20 and $25.  So it would look like a bell curve with the peak at $22.50 and a declining number of dots stretching out towards the edges.  Well, it’s precisely that bell curve that represents a problem for oscillators.

Go to BigCharts.com and pull up a chart along with the MACD oscillator.  Now, the underlying concept of the MACD is excellent but it always generates a lot of false signals at the top and at the bottom.  The reason is that the turning point data is stretched out at the edges.  The stock’s price is changing directions at $20 and at $25 but that’s exactly where there’s the least amount of data.  So you frequently have these sweeping sideways periods where the indicator doesn’t know which way to go.  The indicator is really sharp through the center, when the stock is traveling through the center of the channel but its poor an indecisive on the edges, exactly where the price will likely change direction.

Well, there are a number of formulas you can impose on the data to stretch out the CENTER and pull more of the data towards the sides.  In fact, there are mathematical ways to literally invert the bell curve so most of the data is pushed up on the sides and the center part is bare and stretched out.  Think about it like you’re just adjusting the scale of the numbers along the bottom, between $20 and $25.  If the data between $22 and $23 was stretched into a much wider area but the data between $20 and $21 as well as the data between $24 and $25 was squeezed into a much narrower area on either side, the curve would look very different.

Anyway, you don’t necessarily need to understand the mathematics behind it but the process significantly sharpens the signals at the top and bottom, yielding much more precise buy and sell signals.  In fact, with most oscillators, you end up with incredibly jagged corners at the top and bottom of the indicator’s channel and you could simply program your platform to either buy or sell (or both) when the line changes direction.

Now, even with these new precise indicators, you’ll still end up with false signals.  In fact, the way different platforms are compared is insightful to the discussion.  Automated trading platforms are essentially graded according to three metrics.  First, you have the percentage of correct trades and any score above 45% is pretty darn good.  Let me say that again.  45% correct trades is a GOOD score.  So don’t think you need to get 75% or 80% correct trades to make money.  Nothing could be further from the truth.  If you can be correct 45% or more, you’re off to a good start.

The second variable is generally called the Profit Factor.  The Profit Factor is a ratio of your average profit to your average loss.  So if your average profit is 2% and your average loss is 1%, your Profit Factor is 2.  Likewise, if your average profit is 1.5% and your average loss is 1%, your Profit Factor is 1.5 and as it turns out, anything above 1.5 is pretty good.

The third variable is the number of trades generated each week.  Now here’s the interesting part.  You could easily get a higher percentage of correct trades but you’ll end up with fewer trades per week.  You could also get a higher Profit Factor but that will also lessen the number of trades each week.  In other words, you can simply program your platform to require a higher level of confirmation before generating a buy or sell signal.  By making that change, you’ll increase the percentage of correct trades and you’ll probably increase the Profit Factor as well but the number of trades per week will go down as a result.

Believe it or not, you’re sometimes better off allowing a higher number of trades go through, even though your percentage of correct trades and your Profit Factor will suffer.  It’s all about testing, testing and more testing.  You can also build an Excel model pretty easily that will translate these variables into possible outcomes but it’s important to remember that some combination of variables will not generate a predictable and specific result.  The percentage of correct trades will happen at random and even the Profit Factor is based on averages.  With any one combination of variables, the platform could yield a wide variety of different outcomes.

For those who study this process, it all boils down to an average annualized return with a standard deviation for each combination of variables.  And there’s no magic combination that works best.  Some platforms have impressive profit factors but low everything else while other platforms have a high percentage of correct trades at the expense of the other two.  I recently looked at a Market Timing System online that was boasting a 65% correct trades record and a 1.2 profit factor.  It didn’t say how many trades it was generating but it was generating some impressive returns so it was the percentage correct trades that was DRIVING that platform.  For others, it might be the profit factor that’s DRIVING results.  It will be different for every script.

And there’s a fourth factor as well.  You can incorporate a money management component in the platform so it will invest more or less money depending on market conditions.  When the indicator is really strong with good confirmation, the system might invest more money than when the indicator is weaker.  Perhaps it could be based on the underlying trend.  So when the market 13-day moving average is going up and the platform is advocating a buy, the system might commit more money to the trade than when the market 13-day moving average is going DOWN when the platform generates a buy.  Going a step further, you could program the platform to ONLY take the trade if the market 13-day moving average is congruent with the signal advocated by the platform.  A strategy like this would definitely reduce the number of trades but it would likely improve the other two variables.

The Market Timing Systems we discussed in the last chapter were all automated trading platforms.  All of those systems are built on sophisticated technical indicators and algorithms designed to generate buy and sell signals with the highest possible success ratio but there are a few critical differences that are worth pointing out here.

For starters, the percentage of correct trades is MUCH more important for a Market Timing System than an automated trading platform.  It’s a perception issue.  Subscribers need to feel like they’re making progress, even if it’s slow.  If a Market Timing System has a low percentage of correct trades, it’s going to run into problems.  Even if the profit factor is impressive and the overall returns are positive, subscribers will have a hard time continuing the service if they see too many losing trades on the system’s recommendation.

Second, an automated trading platform can initiate trades every day or even multiple times each day without a problem.  There’s no human involvement so the number of trades isn’t really an issue.  On the other hand, you can’t provide a Market Timing System to the public that could potentially generate multiple buy and sell signals in a single week.  Subscribers would quickly get frustrated and would not be able to keep up with the activity advocated by the system.

As it turns out, one of the largest determinants for a potential subscriber when considering a Market Timing System is the number of trades generated in a given week or month.  Myself, I don’t want a platform that consistently generates more than 15 or 20 trades per year.  I’m not interested in switching sides every 2 or 3 days.  I’d rather try to catch the larger sweeping trends, at least if I have to make the trades myself.

But for those who don’t need to be personally involved in the individual trades, an automated trading platform might be for you.  I have already mentioned the free manuals provided by TradeStation.com and I really encourage those of you who are interested to download those files and take a look.  And they’re not small files, by the way.  One’s 144 pages and the other is 292 pages so we’re not talking about small resources here.  These are entire books that are available for free to the public.  Take advantage of it and start learning the possibilities.

I also recommend a book that deals with the transformation of oscillators in detail and it even provides the specific Easy Language code so you don’t have to figure it out on your own.  It’s called the “Cybernetic Analysis for Stocks and Futures; Cutting Edge DSP Technology to Improve Your Trading” and it’s by John F. Ehlers.  It’s not cheap but an incredibly valuable book in my opinion.

It’s worth noting that the FOREX markets are ideal for automated trading platforms.  The FOREX markets deal exclusively with foreign currency markets and the beauty of those markets is that they’re open 24 hours each day and only close for the weekends.  One of the biggest risks in the stock market is the fact that information often becomes public when the market is closed and the security will often GAP up or GAP down, meaning the opening price is materially different the closing price of the previous day.  That’s not a risk in the currency markets, at least during the week.  And because of that, the automated trading platforms can continue working all day and all night without incurring the risks of a temporarily closing market.  We’ll be discussing the FOREX markets in a later chapter.

I recently had a conversation with someone in the help desk of a FOREX brokerage and they suggested the vast majority of traders with FOREX accounts lose money and that the ones making money almost all had automated trading platforms working.  The beauty of these platforms, of course, is that you can back-test them to see how they would’ve faired in past markets.  That way, you can isolate problems and optimize results before giving access to your hard-earned money.

There’s no question automated trading platforms are valuable investment tools and they’ll only become more popular as technology increases accessibility.  In fact, I believe the Exchange Traded Funds industry will be incorporating more and more technical trading features into their offers, making automated trading accessible to more investors.  I mean; let’s face it.  The reversal and magnification of market movements is already a significant manipulation and I believe the incorporation of other technical indicators isn’t far behind.

Whether it’s through an Exchange Traded Fund, a Market Timing System or building your own automated trading platform, I hope this chapter sheds some light on the science and mathematics driving the process.  It’s fascinating stuff and will hopefully broaden your outlook on the various investment options in today’s financial world.

If you’d like to see what I’M doing in the market, just create a FREE account and login at FinancialAudio.com and then click on Market Position.  Now, it’s important I tell you that I’m NOT a licensed Financial Advisor.  This is FREE information and I do NOT accept money to give investment advice.  I’m just giving you a little glimpse into my own trading activity, that’s all.

You can also read market commentary on the blog page.  If you’d like to see a list of upcoming topics, just click Podcasts on the homepage.  If you have a suggestion for a future topic, please use the Contact form to let us know.  And finally, all the websites referenced on this podcast have been included on the Links page.

Okay, thank you very much for listening.  If you like what you hear on these podcasts, please tell a friend about them.  Modern technology like podcasting can help elevate new and innovative thinkers but we all have to play our part to help spread the word for those who deserve our endorsements.  If I am deserving of yours, my thanks.

Again, you can find written versions of these podcasts at FinancialAudio.com and I do offer workshops, seminars and keynote speeches as well as a variety of more advanced information products so please email me at Patrick [at] FinancialAudio.com for more information.  I’m also doing a series on innovative marketing and strategic business positioning.  That series is called Tactical Execution and you can find it on iTunes.

Stay tuned.  There’s a lot more to come.  In the meantime, think big, take action and invest strategically.  Bye for now.