Interesting GBP System

I found this system on another forum, and it looked very interesting. It is a very simple "set it and forget" type system, and it has been performing fairly well.

I was hoping that Merlin or someone else with the resources and capability could do some backtesting on it.

Here are the rules:

At 5PM EST Sunday-Thursday:

1- Close out any open positions
2- Cancel any unexecuted orders
3- Set an Entry Buy order 1 pip above previous day's high (We are now taking the spread into consideration on BUY orders)
4- Set an Entry Sell order 1 pip below previous day's low
5- Based on our discussion set stops and limits using the following guidelines:

50 pip limit/25 pip stop if ADR ABOVE 200
40 pip limit/20 pip stop if ADR BETWEEN 175 and 199
30 pip limit/15 pip stop if ADR BELOW 175
ADR BELOW 100 -> Gone Fish'n

Account balance $300-$599 1 mini lot
Account balance $600-$1199 2 mini lots
Account balance $1200-$2399 4 mini lots
Account balance $2400-$4799 8 mini lots
Account balance $4800-$9599 1.6 lots
Account balance $9600-$19199 3.2 lots
Account balance $19200-$38399 6.4 lots
Account balance $38400-$76799 12.8 lots
Account balance $76800-$153599 25.6 lots
Account balance $153600-$307199 51.2 lots
Account balance $307200-$614399 102.4 lots
Account balance $614400-$1228800 204.8 lots

This system was devised for the GBP/USD, but you test it on other pairs to see if it will work. Also, you might try different stops and limits such as just 40/20 or even a totally different value like 32/16.

Anyway, the group from the other forum have been trading it live with some success, but I would like to see some backwards testing before I give it a try, just so I can trade with confidence.

Thanks.

What do u use for the time frame to calculate ADR?

To get the ADR, take the daily high - the daily low and multiply that number by 10000. This gives you the daily range. Then take the daily range for the past 14 days, add them together and divide by 14. That will give you the 14 day ADR.

Hello All,

There is revision to this system.

"If we have a large move on news and the GBP closes at on near the Hi or Lo of the days range, we will place ENTRY LIMIT orders instead of ENTRY STOP orders, treating the Hi and Lo as support and resistance. In the event we do get a breakout, we will have ENTRY STOP orders placed 15 pips above and below the previous days Hi and Lo, allowing for a stop and reverse trade."

Here is an example:

Suppose on the day of NFP, Days High is 1.8000. You will place Entry/Sell at 1.8000 and Entry/Buy at 1.8015

Some one had backtested this system with since Jan 04 with no losing month and average +300 pips profit per month. Although I wont recomend to use money managment plan mentioned in the system.

This is one of the best free trading system I found on the Internet.

Mkcky

such that there is another thread that states there are too many rules and thus to go back to the basic system...

I am not sure what forum it is on.

LOU

Yes this system is refined many times but the system rules posted above by Rosscub are almost recent. Using this statergy you get returns of 5224 % in just six months.

So if you had
300 Dollors Invested, Profits after 6 months 15,972 Dollors.
1000 Dollors Invested, Profits after 6 months 52,440 Dollors.

And you can achive this just by trading 5 minutes a day.

Too good to be true? Have a look at excel sheet.

You can use ATR Indicator which is included in most of charting software, Set ATR to 14 and switch to Daily chart.

ATR value multply by 1000= ADR

Mkcky

Look at post number 4 above. It said:

"If we have a large move on news and the GBP closes at on near the Hi or Lo of the days range, we will place ENTRY LIMIT orders instead of ENTRY STOP orders, treating the Hi and Lo as support and resistance. In the event we do get a breakout, we will have ENTRY STOP orders placed 15 pips above and below the previous days Hi and Lo, allowing for a stop and reverse trade."

Here is an example:

Suppose on the day of NFP, Days High is 1.8000. You will place Entry/Sell at 1.8000 and Entry/Buy at 1.8015 "

Some things here either dont make sense or I am looking at them wrong. In the first paragraph are you talking about the same day or the next day? Are you then over writing rule # 1 ?

In the example, are you talking about the same day or the next day? If the same day then you are over writing rule # 1. If the same day or the next day the market is at the high of 1.8000 what good is placing a Sell Entry order at the same price as the market?

Clarification please.

Thanks.

LOU

Hi Lou,

Its always a next day. For more reading about this system see following URL


http://www.visualtradingcharts.com/f...631205bf8ee63f

http://www.visualtradingcharts.com/f...ce35323c701bc5

Mkcky

August 26 hi-lo :
- High 1.7990, LONG at 1.7991
- Low 1.7916, SHORT at 1.7915

ADR reading 144, so SL 15 pips and TP 30 pips

Today enter the market LONG at 1.7991, SL 1.7976, TP 1.8021
TP hit at 1.8021

August 27 result +30 pips
August result +30 pips

ps. using NetDania, indicator ATR(14) on daily chart

I use different timings. I place trades at 0:00 EST just before Europen session. If the trades don’t get filled by 12:00 EST, I cancel them.I prefer this timmings because I can trade this statergy along with 5 minutes Alternate Strategy.

I think EUR/JPY is secound best pair to be traded with this statergy. Can any one backtest it?

August 27 Results
GBP/USD +30 Pips
EUR/JPY +40 Pips

Mkcky

Hi Rosscub,
I did the backtest on Amibroker; unfortunately I got not very good results. Taking in account a 4-pip spread for GBP/USD (and not taking into account slippage from entering at gaps or spikes), the results are:

From 2004/1 to now: 452 pips
From 2003/1 to 2004/1: -386 pips

Also, bear in mind that until July, we were barely above breakeven:
From 2004/1 to 2004/6: 72 pips


Using a fixed 50/25 for profit/loss, we get better results, although it makes less sense "conceptually":
From 2004/1 to now: 1019 pips
From 2003/1 to 2004/1: 100 pips.

I've seen this system published repeatedly at Moneytec and here, I think it's become popular due to its good performance throughout 2004; however, I feel it may stop working any time soon. Years 2001 and 2002 also had poor results.

Best regards,
Bonete

Firedave,

Are you still testing this GBP sytem?

This is modified (simple) statergy with Stop/Reverse and is lot more profitble then the orignal statergy.

- At 00:00 EST, Cancle all the traders from yesterday.
- mark yesterdays hi, low and mid price.
- Set up buy at 5 Pips above yesterdays high and sell at 5 Pips below yesterdays low, If yesterdays was a very large range day use the hi or low and middle.
- Once trade triggers cancel the other and replace it with a 50 pip stop and reverse.
- If no trades triggered by 12:00 EST, cancel all outstanding orders for the day
- Maximum 2 trades per day , 1 breakout and 1 stop and reverse if 1st trade fails.
- If the market is very close to the hi or low of yesterday add 20 pips for the trigger.

This statergy works best on GBP/USD. I have tried EUR/USD and EUR/JPY but they are not as profitable and GBP.


Mkcky

Mic, can you post the result day by day ? I'll post the result using the original ADR system. how about that ?

Hope it will help other

Hi Firedave,

Done deal I will start posting results from today.

Mkcky

Hi guys,

I've been keeping a close eye on the modified system described by Mkcky although only in detail from mid- May. I've seen some figures posted that this made 4000 pts between January and July but not verified them myself.

Last 2 weeks in May +454 pts
June +360 pts
July +645 pts
August 0 pts (inc today)

Until this week it went 150 pts down a couple of times but this was the worst week by far, i got it as -264 pts - stopped out twice for full 100 pts and a few more. There have been another 3 negative weeks since May: -50, -48, -65, so this is definitely a step down. It's still a good overall average, but I was just about to try this one live, but got delayed in opening a separate mini account to test it. Now I'm not sure whether to just dive in next week or monitor it some more.

My figures are from eyeballing only so give them a bit of leeway.

Hi Knossos,

Did you have 0 Pips in Aug? I think you are refering to Sept.

If you have noticed 14 days ATR is between 160 to 140 in Aug. I would suggest to use 30 pips stop loss till ATR rises above 175. This will cut your losses and also add some pips to your secound (Stop/Reverse) trade.

Mkcky

Hi Mkcky,

no, definitely August (though forgot to mention I didn't use a 5 pip filter, just a straight break, though looking back it wouldn't have made much difference - also no trade on Monday).

For August i got:

2-6th: -50
9-13th: +222
16-20th: -48
23-27th: +140
31-3 sept: -264

I suppose you could call this last week september, then august wudda made +264 , and september is down 264.

Yes, been looking at ATR also, but July was a good month with ATR below 170. I had checked over reducing the stop and it seemed that 40 pips would have seen an improvement since May at least. 30 pips got a bit too much whipsaw, but again, these were just rough checks.

With flexi contracts (as with Proedgefx, fxsol, gft etc.) or continuous contracts (as with Oanda), one gets the flexibility to chose the amount of risk better. If back test shows that the best performance is obtained when the stop is set at say 0.5 times ATR(14), then we should stick with that. In that case, only trade a smaller number of contracts, so that if the stop is hit, the loss to the account is the same as say if 0.3 times ATR(14) was used, but a larger number of contracts/lots were traded. It is my belief that one should not just mess around with systems to reduce risk arbitarily. It may just have the opposite effect, increase risk. In this case, if the lower multiple of ATR is used, the number of times the stop loss is hit will increase, and that will degrade performance.

I have been following this thread and the one on Moneytec with interest, and I have a couple of comments. As usual, I maybe totally off base, and so Merlin, or any one of the experts or any one for that matter, please correct me.

1. The system by itself appears to be robust. Why mess with it and introduce more filters? It may appear to be more profitable in the short run, but it may lose robustness and degrade performance.

2. All mechanical systems that I know of, are subject to periods of drawdown. That is a fact of life for these trading methods. I have been there done that, when I screwed with system parameters just because had a few losing trades. If you have stress tested your system, and it passed, then believe in its abilities. Prepare to undergo two to two and a half times the maximum historical drawdown before you give up or tinker around. I wanted to trade a well known futures system, and called a few reputable system brokers. I had the developer's minimum recommended capital and was ready to rock and roll. One of the best brokers (in my esteem, of course) told me outright that he doesn't want my account, unless I can fund it with a much higher amount. He said that is the amount which equals to twice the historical drawdown plus a certain amount for margin for new positions. I have since spoken with many times, and have hopefully learnt a few pointers about system trading.

3. Don't be afraid to trade around news. If historical testing has shown that your system should be traded tuesday through friday, then place your trades religiously. Don't move stops just because a news is due and you are in profit, but your system does not call for moving stops. You are being discretionary, and that is a sure way to mess a mechanical system. Again, I have been there done that. A robust backtested system surely faced those kind of news events in its back test period, and if it withstood the rigors of testing, then believe it! However, for once in a lifetime kind of news maybe an exception, but I hate to break rules even for that... unless of course the markets are closed.

Sorry for rambling... I see that others maybe on the verge of committing the same mistakes that I have made in the past, and couldn't help myself. Again, sorry to have bored you.

Have a great labor day weekend.