GBP breakout system

Hi,

I have borrowed a few ideas of some people here (mostly JohnnyT) and might have come up with a very simple system for trading the swiss franc future on globex. The rules are as follows:

1) Plot 21 sma on daily chp chart and see previous day close - if above, only look for long; if below, only look for short
1) Breakout of 7-10am bar (all uk time) in direction daily chart has signalled
2) If get a signal, close trade at 12 59 59
3) If not got signal by 12 30, wait until 15.00 and re-enter trade of 7-10 breakout level
4) close at 20 59 59
5) no stops

Over the last 6 months, this has produced a profit (excluding slippage but including commission of $6.25/RT) of $7,800 trading one contract. With good money management, this would be nearer $10k. Max drawdown was an ridiculously low $600!!

However, as i use esignal, it wont let me import data to further backtest this so was hoping someone might want to have go doing it on tradestation or one of the other programmes with a large intra-day database going back 2 - 5 years.

The logic behind this is that the 21sma filter keeps you on the side of the MT trend; closing at 12 59 59 keeps you from gambling on the economic data; re-entering at 15.00 gives the chance of capturing the afternoon reversal.

Any help/improvements would be greatly appreciated.

David

interesting ideas there. hadnt thought about re-entering on the similar break in the arvo.

will have a look at it with my slightly wonky data to see if it corroborates.

in the meantime, i'll see you guys on the other side of a 96-hour bender.

toodle-oo.

FC

be careful FC !!

it starts off with seemingly inocuous SMAs, and before you know it, you'll be on the harder WMAs and EMAs, and turn into a trend-follower.

its a very slippery road you may be embarking upon!

PS; ddune82: is this specific to GBPCHP, or have you tested it more generically for other pairs ?

ddunne82 said:

3) If not got signal by 12 30, wait until 15.00 and re-enter trade of 7-10 breakout level

David

Click to expand...

Hi David,

I'm not sure what you mean by the above. (It's the 're-enter' that I find confusing. )

Are you saying that you only enter in the afternoon if you have not entered in the morning. Or do you mean that if you have taken a position in the morning, you then exit it at 1:00pm and go back in at 3:00pm. Or am I even more confused than I thought and you mean something else.

Also, I think I've found the symbol on IB (PSF ?) but I can't see how much the contract is worth. Could you let me know, please.

(Very interesting post, by the way. Thank you for sharing it with us.)

Regards,

Just realised i made an error in the post - this isnt for GBP - its for the swiss franc. The title of the thread is wrong but i think i mentioned it was for the swiss in the text!

This set-up also works for the EUR but there is a diifferent set-up for the GBP - 11-15 breakout with the same sma filter. Over the last 6 months, all 3 have returned similar positive results which gives me pretty good confidence that this could be the basis for a system.

The draw downs have all been small EUR was around $1300 and GBP and $500 with profit for EUr of $10000 and GBP $5000. As this takes commission into account and cos they are futures, slippage should be minimal so the results should pretty accurate. The settings have not been optimised so if some nice results over 2 years get given, it could be game on.

Ivor,

The contracts are for the foeign exchange futures traded on globex. I can give you the esignal symbols as that is what i use - EUR is 6E; GBP is 6B; Swis franc is 6C. The tick value is $12.50 for EUR; $6.50 for GBP; $12.50 for CHP. Hope this makes it clearer - if not, just go to the cme website and they are on there!

On re-entering for CHP/EUR:

If a trade was triggered in before 12:30, then that is you done for the day (closed out a 12.55); if not triggered, re-enter the trade at 15.00 and close at 21.00.

For the GBP, there is no re-entry point.

Sorry for the muddled explanation, english was never my strong point!

David

ddunne82 said:

Hi,

1) Breakout of 7-10am bar (all uk time) in direction daily chart has signalled

David

Click to expand...

Hi,

is this a breakout of the high/low range or of the open/close range of the 7-10am bar?

Cheers,
Jan

Jan,

Its the breakout of the 7-10 range - i.e. the high and low made between 7 and 10 with a buy/sell stop (depending on sma) placed one tick outside these points. This is only for EUr and CHP - GBP is different and is in one of my ealie posts

David

ddunne82 said:

I can give you the esignal symbols as that is what i use - EUR is 6E; GBP is 6B; Swis franc is 6C.

Click to expand...

Thanks for posting this, David, it's certainly worth further study.

I don't use esignal, but the IB / CME symbol for the Swiss Franc / USD cross (CHF) is 6S (curently 6SM5). 6C is the Canadian Dollar (CAD)

Hi,

Because the US dont put their clock forwards till April 3, the times will be slighlty out.

I think the ammended times should be:

1) EUR/CHP - still 7-10 b/o but re-enter at 16.00 (all uk time)
2) GBP - 11-16 b/o

These ammended times still take into account the time of the announcements but with the US being an extra hour behind, the re-entry needs to be slighlty changed. Come April 3, revert back to original posted times.

David

A quick update on how the system did this week:

1)Combined - $63 profit (after commission)
2)EUR - $182 loss
3)CHP - $113.5 loss
4)GBP - $357.5 profit

Bit of a quiet week, both the EUr and CHP getting hit early on and not moving with any follow through. Friday's GBP trade added the trade that made the system profitable this week but without using the sma as a filter, you're account would be a hell of a lot smaller than it was on monday morning!

System results for this week:

1) Combined - $130
2) EUR - $287 loss
3) CHP - $225 loss
4) GBP - $383 profit

Once again the pound took some of the heat away from the other two currencies. Got me thinking about re-entering the trade in the afternoon even if a morning trade was triggered. Though it would have turned around the results for this week, over the testperiod, it actually reduced returns. Thus i would be sceptical about using it as it could be a case of "curve fitting" and the resulting test doesnt support the results.

I also took out the 21 sma filter with single trade as well as afternoon re-entry and again the results were worse than the original set up.

I also slightly miscalculated the initial results for the EUR which are now correct as follows (Oct 04 - present):

1) Net profit (incl commission excl slippage) $8400
2) Win% - 51
3) Loss % - 46
4) winner size:loser size - 2.43

Not quite as good but still pretty tasty. Max drawdown was $1200. With realistic but semi-aggresive money management, it would have taken $5000 into $29500 in the six months!

Anyone had a chance to backtest this further?

Last week, made a nice little profit of $467 incl slippage and commission. This week, it was the swiss franc and the euro that were the star perfomers with a real nice short trade on thurs morning.

I thought i would mention that on the re-entry in the afternoon, this is not done blindly. What i mean is that you have to use a degree of common sense as to whether to re-enter the trade. As you are using a trend following system it should be fairly obvious as to what it is (i am not that smart a guy so its nothing very complicated) but i am not going to set out what it is here as half the fun is finding out these things yourselves and helps you develop a better understanding of how the system works. This knowledge of the mechanics should allow you to make necessary changes/adjustments in the future.

David

hi ddune82.

is there any way to convert the dollar-values into pips?
pips made/lost may be an easier way to compare different systems aganst each other.

thanks

Trendie,

If you wanna convert the results into pips (cant see why you would want to as your bank account doesnt show how many pips you made but how many $) for the EUR/CHP just divide the amount by 12.5 and for the GBP divide by 6.25. The results i posted INCLUDE commission (half a pip for EUR/CHP and 1 pip for GBP) so if you want to exclude commission from the results (cant see why you would want to though), work on the basis that there were roughly 80 trades in the period tested (i dont have the exact figures on me). Thens its just a case of simple addition!

Hope this helps.

David

Results for this week: (NET results so commission AND slippage included)

1) Combined - $304.38 profit
2) EUR - $300.68 loss
3) CHP - $11.82 profit
4) GBP - $593.24 profit

In a nutshell, at the end of play on tuesday, we were up a tidy sum but unfortunately when Mr G and his little helper spoke on Thursday, we were already in a long position on the EUR and CHP and their comments went against us. One could argue the case for using stops and though they would have reduced the loss last thurday, the back tested results do not correspond to this. Just gotta suck up the losses when they happen. However, the last two weeks have produced a NET profit of over $750 so its not all bad news!

David

This week the system produced a net profit of $193 - mainly due to a nice trade on friday on the EUR. For the month of april, all instruments were prpfitable with GBP being the star. Performance closely mirrored that of the back-tested results which is always an encouraging sign.

David

Last two weeks results:

1) w/e 06/05/05 - total loss of $920.25
2) w/e 13/05/05 - total proft of $799.00.

As we speak, system is still in the positive with the 21 sma filter working really nicely.

David

Slippage is likely to be around 5 ticks each side given how iliquid this contract is - especially at that time of day. Have you factored this in?

2 trades a day = 4 rt a day. Thats 20 ticks you are losing straight away (worst case). So take off at least $250 each day you traded from your results.

Does it still seem attractive now? Trading with forex will not be much better as you will still be paying a wider spread of 3 pips or so I would imagine, not to mention having entry stops run on many occasions resulting in a higher number of losing trades from false entries that may not have occurred in the more honest futures markets.

Back testing will never beat hands on knowledge & experience of trading markets. Your time could well be better spent trading rather than building magic systems. This isn't a numbers game.

Hope that helps.