Australian (ASX) Stock Market Forum

Trading System by Andrew Gibbs - Member of Larry Williams Hall of Fame

Joined
28 July 2010
Posts
42
Reactions
0
Hi Guys

This post and trading system is EXCLUSIVE to Aussie Stock Forums. I am going to show you a complete share trading system, back tested performance and something that any trader can take away and use.

The following post outlines a share trading system that is designed for trading stocks that meet a specific scanning criteria that reduces the stocks you are trading down to a basket of say 20 shares. It is to active to trade across an entire index, say the ASX200. As a part of this article I am going to also include some exceptional (and backtested) screening approaches, along with some shortcuts on how to choose stocks to trade.

Have you ever run your favorite scan based upon either fundamental or technical criteria, come up with a group of stocks and then struggled for entry and exit criteria to get you into and out of a stock. This system addresses that very problem. It is designed to give you an efficient way to get you into and out of trades that is profitable across the majority of stocks.

A warning first. This system, like the stockmarket and most systems suffers from large drawdowns. In other words leverage at your own risk. It does however offer a highly profitable method for entering and exiting the market.

A second warning. The scanning criteria is the way to tilt the market well and truly in your favour. If your stockmarket scans give you a group (or basket) of stocks that are most likely to outperform the market you can really tilt the playing field in your favour before even applying a technical trading system to the stocks. Let me ask you this.. would you rather trade stocks that belong to an index and average say 8-10% per annum or a basket of stocks, if you re-weight the basket every so often returns 20-40% per annum. The right answer is trade the basket that average 20% plus..not just the stocks that make up an index.

Following on:

This trading system is designed to work across just about any stock and is designed to be in the market a lot.

In building a trading system that works across almost anything we need to analyse the charactertics of the stockmarket itself.

The stockmarket has two characteristics, it regresses to the mean after a downturn, in other words turns up and doesn't trend very often to the downside and it shows momentum after breaking to new highs or in other words shows trending characteristics on the upside. The result is that two types of signals should work, an oversold signal and /or a breakout signal. This system uses both.

Before I disclose the entry rules, I emntioned earlier that it is the scanning criteria for your portfolio that really stocks the odds in your favour. If you get your scanning criteria right and the criteria has a history of market outperformance the best way to enter the market is during a dip.

There are lots of ways to do this for example buy when price goes BELOW a moving average. You could use an oscillator like an RSI and buy when it shows an oversold reading etc.

Our technique simply uses the parabolic indicator. If the price of the stock is BELOW (Note I said BELOW!, not above) and the 4 period RSI is below 25 we will buy at market.

Rule 1: If Close< Parabolic Indicator (AF 0.01, maxAF, 0.2) and RSI(close,4)<25 then buy the next bar at market.

The second rule kicks in if the first one does fill. If close is below the Parabolic buy at the 15 period high price on stop.

Note that traditional technical analysis say to buy above the Parabolic indicator, I disagree, it's not as effective as anticipating a trend change, although does leave you bruised at times (like anything).

This chart demonstrates the entry rule 1 in action.

Entry Rule 1.png

This chart demonstrates entry rule 2 in action

Entry Rule 2.png

Note that the shaded green boxes illustrate when price is trading below the Parabolic SAR....

If you think I'm cherry picking the good trades, you'd be RIGHT! I'll show some of the uglier trades later.

Let's look at the exit rules:

The first one is as follows:

Once we are in uptrend, which in this case is established when the Parabolic indicator flips so it is below the closing price.. as shown above when the close is ABOVE the red dots we will use the Parabolic indicator to trail the stop.

This exit works great in strong up trending markets as you can see in the previous trades shown, but what about when the market is trending lower... the second exit shunts us out of the market more quickly. Basically if the index (we are using the S&P500 Cash Index (SPX) in this example is below where it was 100 bars ago we will exit as soon as the parabolic flips to uptrend and the close is ABOVE the Parabolic indicator.

Exit Rule 2.png

YES.. still cherry picking


Here's a BAD trade:

Bad Trade 1.png

And here's an UGLY one... followed by two mediocre trades

The Ugly.png

In the next post we'll look at the backtesting history and the use of stops.

For the FUNDAMENTAL guys out there, you've probably already guessed but our screening techniques we use to pick stocks are based almost exclusively on fundamental analysis and yes, unlike almost everybody else we backtest our screening techniques as well.. until tomorrow, enjoy your evening. I've written a book on how to trade BHP as well, click the link in my signature to go the the landing page.

Enjoy

ZZ...
 
WHat does this mean pls

"The second rule kicks in if the first one does fill. If close is below the Parabolic buy at the 15 period high price on stop".
 
WHat does this mean pls

"The second rule kicks in if the first one does fill. If close is below the Parabolic buy at the 15 period high price on stop".

Sorry, it's a typo (does should be doesn't)...

Here's an attempt at a clearer explanation

The price needs to be trading below the Parablic SAR for both entry rules (shown by green shaded areas in the example):

You buy if the 4 period RSI Indicator is below 25 (the next bar at market on open)

or

You buy on stop at the highest high in the last 15 bars using a stop entry order.

Whichever one occurs first.
 
*** Warning to members ***

Please be VERY wary of posts that provide you with a trading system. Ensure you backtest/forwardtest them YOURSELF and can demonstrate profitability.

Quite often, people post these systems up on sites for free, because they want to take the other side of your trade. I.e. they give you a system that sends you a buy, but in reality it is a 'sell' on their side and they milk you.

Now, zipzap may or may not be dodgy, I don't know. If their intentions are pure, then I apologise - no insult intended.

This is just a caution to beginners who are still learning and haven't come across something like this before.

Please please please validate the system personally (as you should no matter who posts).
 
Thanks for that... sadly, you're right, there are many unscrupulous operators out there and therefore it's important for you to do your own investigations.. these are just trading ideas some I trade, some I don't but this one has value.

The image below shows the back tested results on the S&P100 stocks going to back to 1980.

Assumptions: $10,000 invested per trade, no slippage or commission included.

Backtest to 1980.png

The Equity Curve..

Equity Curve.png

ZZ...
 
Performance Reports Per Stocks

Per Stock performance Report.jpg

And the next part of the table

Per Stock Performance Report2.jpg

Hard to read as I had to re-size, the challenge is to see if you can find a stock in the S&P100 index the system loses on....

ZZ...

p.s. this systems wins on every single stock that makes up the current S&P100 index. There are issues such as survivorship bias etc that we'll touch upon later. As I mentioned earlier, this is a very active system, it is designed to be traded on baskets of stock that are much smaller than the S&P 100 stocks and we'll get to how these are selected another day as well.
 
Highest number of consecutive losses 52 - ouch!! That takes some strong mindset there to keep going.

Some good results, but requires you to be very active in your trading. 18000+ trades across 32 years = ~12 trades per week. Good expectancy although payout ratio is a bit low for my tastes (just personal trading style) but fine given the high win rate. Problem is that you need to remain "right" most of the time for this to be effective.

Some interesting ideas for better entries/exits.

Thanks for sharing.
 
Highest number of consecutive losses 52 - ouch!! That takes some strong mindset there to keep going.

In reality it is not practical to trade a basket of 100 shares, this system is designed as a way to enter and exit stock baskets of say 20 stocks that are selected based upon some fundamental screening techniques which really tip the playing field in your favor way further than any technical system can.

The purpose of testing across such a wide number of stocks (and it shows similar results on the S&P500, Russell3000 etc) is to show how robust the trading method is before refining the basket of stock we're looking at.

More on this tomorrow.

ZZ...
 
Had dinner with Andy
Very very talented guy.
Has a great sense of humor too boot.

Indeed fortunate to have Andrew post let alone
Posting a fully disclosed system plus his feed back.

He's he real deal.
Didn't know about the Hall of Fame
How'd you get that?
 
It doesn't appear your backtesting accounts for changes in the S&P100 index. You are backtesting the current constituents of the S&P100 going back 1980. Your backtest is invalid.

What is the purpose of statements like this

the challenge is to see if you can find a stock in the S&P100 index the system loses on.

except to sensationalise the idea.
 
I can't get over just how intellectually dishonest this backtest is.

How many of the stocks in your backtest haven't even been in existence since 1980? CISCO listed 1990, Google 2004, Intel in 1989, Amazon in 1997, so on and so forth.

So exactly how is the equity curve calculated? How are the backtest stats adjusted to account for this? In which year did the backtest actually start to cover 100 stocks? 2004?

You say you use a fundamental screen, Fama and French data is freely available and you could have easily constructed a fundamental universe each year going back much further than 1980 for the backtest but instead you chose to present an impossible scenario as profitable, and say we will touch on the issue which completely invalidates it, later.

If this isn't a marketing strategy to drive customers to SCM/Halifax I don't know what is. As someone who has previously mentioned SCM in a good light on this forum I am pretty disturbed.

I think OP should come clean about who he is (is he actually Andrew Gibbs? I doubt it), what is his affiliation with SCM/Halifax, and wherever possible not post BS backtests.
 
I can't get over just how intellectually dishonest this backtest is.

How many of the stocks in your backtest haven't even been in existence since 1980? CISCO listed 1990, Google 2004, Intel in 1989, Amazon in 1997, so on and so forth.

So exactly how is the equity curve calculated? How are the backtest stats adjusted to account for this? In which year did the backtest actually start to cover 100 stocks? 2004?

You say you use a fundamental screen, Fama and French data is freely available and you could have easily constructed a fundamental universe each year going back much further than 1980 for the backtest but instead you chose to present an impossible scenario as profitable, and say we will touch on the issue which completely invalidates it, later.

If this isn't a marketing strategy to drive customers to SCM/Halifax I don't know what is. As someone who has previously mentioned SCM in a good light on this forum I am pretty disturbed.

I think OP should come clean about who he is (is he actually Andrew Gibbs? I doubt it), what is his affiliation with SCM/Halifax, and wherever possible not post BS backtests.

He did mention survivorship bias was an issue. Give him a chance to explain on that front.

Halifax in Australia is a very dodgy set up. Not sure about NZ.
 
He did mention survivorship bias was an issue. Give him a chance to explain on that front.

Sorry, no I cant accept that.

OP told us

The purpose of testing across such a wide number of stocks (and it shows similar results on the S&P500, Russell3000 etc) is to show how robust the trading method is

If OP was for real, and not just paying lip service by mentioning survivorship bias (then dismissing it for later, after you're reeled in), he would absolutely not post the equity curve and stats he has to "show how robust the trading method is".
 
Hmmm, 18,000 trades, 4 simple rules across 100 stocks, same rules tested across all of them, no optimisation at all.. tested back 30 years and the system makes money on every single one of them... yeah..crap system, throw it in the bin.
 
Hmmm, 18,000 trades, 4 simple rules across 100 stocks, same rules tested across all of them, no optimisation at all.. tested back 30 years and the system makes money on every single one of them... yeah..crap system, throw it in the bin.

Same old question. Why share it with the general public hmm?
 
Same old question. Why share it with the general public hmm?

Creates discussion
Has a desire to implant ideas in others
He can
Thank god it's not ANOTHER general un related to trading thread.
I created TechTrader it's been published---my motive was/is?

Andrews created 100s of systems.
 
Same old question. Why share it with the general public hmm?

You guys are so clever, you've seen through my cunning plan...

I actually think what I am posting has value and yes, better than some id..iot posting a stock tip and yes... I have a business interest, I own a broking firm and manage millions of dollars in client funds. The difference being I really care about making my clients money whether they are self directed or via a managed account, it's good for everyone..win:win, I get long term clients and make money from fees, client makes profits from trading or investing and taking on risk. But.. if you want to listen to Charlie next door and follow his system then be my guest.

The other thing is if you're in business you have to market, it's that simple... I'll be straight up, I've just expanded the business and am now in a position to expand my client base. We have the best research products in the market but if you're alrady making money from stock or CFD trading you don;t need to read this, if you're not I'm showing you how you can.

All you guys need to do is sit back, read and enjoy, be thankful that I have a commercial interest or you wouldn't be getting anything. I also trade my own account using my trading systems. Take it or leave it.

So... moving on..

As I've mentioned several times throughout this thread already this system simply outlines a set of rules for entering and exiting the market, it's nothing special, there's no secret sauce and it can have large losses so risk needs to be managed through diversification, but this system has the elements it was designed for. It captures virtually every move and generally is hsown to be profitable across a wide range of stocks and time periods.

But the entry and exit rules are only half of the story, the system only makes money because of the upward bias in stocks... Apply the same thing to commodity futures or forex (which do not always have an upward long term bias due to cost of carry) and it won't win.. but that's fine because this is about trading stocks. The point is without good stocks the system won't make money.

The most important consideration is which stocks to trade and how do we select them. Given the system is in the market a lot and very active, say one or two steps down from buy and hold as I just said the key to the entire thing is to find stocks that are most likely to outperform the market, we only need about 20 and that is the real edge and is what I'll cover next.

ZZ..
 
I remember ZZ from years ago. He's no scamster and I'm interested in hearing more on this.

Hope to hear from you today my man.


Yep ZZ has been around for ever and unless he has joined the Mongrel Mob should be straight up would encourage everyone to ask questions with respect and allow ZZ to explain until proven other wise.

The points that ZZ makes are fundamental in trading the main one is moving the bias in your direction I would'nt burn him until at least he explains his method........unless you are a genus making millions.
 
Top