three (de natchos2)
Lutilisateur a supprimé ce système.
Discussion three
Mar 18, 2010 at 17:46
Vues 994
10 Replies
Membre depuis Mar 13, 2010
posts 23
Apr 27, 2010 at 12:51
Very nice, Joe. What other pairs would you consider running with this one? Do you have a feel for its safety?
Apr 27, 2010 at 14:30
Look this thing is not 'safe' at all. In fact it will wipe eventually (like all EA's), the thing is the probability of at least doubling your money before it wipes is pretty high; in current market conditions this should quadruple before a 'black swan' event.
If you can find any other pair that's 'stable' like EURCHF, go ahead and backtest for it. I'd like to know how it turns out
One idea i had that would be interesting it to go SHORT only on EURAUD and run it kind of like a swing trader
If you can find any other pair that's 'stable' like EURCHF, go ahead and backtest for it. I'd like to know how it turns out
One idea i had that would be interesting it to go SHORT only on EURAUD and run it kind of like a swing trader
Membre depuis Mar 13, 2010
posts 23
Apr 28, 2010 at 14:47
I am fascinated by this approach, since the five min. window might mitigate the drawdown. I started ten pairs: GU/EU/UJ/EJ/UC/EC/CJ/EG/UCAD/AudUDS
Close profit at 0.5% and close loss at 10% within each pair.
Also added Steve Hopwood's trade manager EA with the same 0.5/10.
Obejctive is 2%/day, compounding.
Off to a nice start. Cleaning the table after a global equity rise wipes out all the troublesome DD, and locks in the profit.
Close profit at 0.5% and close loss at 10% within each pair.
Also added Steve Hopwood's trade manager EA with the same 0.5/10.
Obejctive is 2%/day, compounding.
Off to a nice start. Cleaning the table after a global equity rise wipes out all the troublesome DD, and locks in the profit.
Apr 28, 2010 at 15:07
There was a post i made on the SSS thread that talks a lot about the theory behind this, its a long post but worth the read
I've come to the conclusion that a 1:3 ratio is ideal. So for example a lot size of 1% and a global TP of 3% etc.... I've stretched this to 2.5 and 7.5 on EURCHF without wiping in from Jan to Dec 2009.
I can tell you right off the bat the GU, EU, UJ, UCAD and AudUSD pairs will wipe pretty quickly. The loosing batches also have a cumulative effect on lot size and TP
I'm not super familiar with Steve's trade manager, I've played around with it however... Let me know how it goes.....
This backtests nicely so I'd recommend running it on the backtester before any demo accounts. I backtested this for over a month to get it just about right.
This is really based on chaos theory and all that goodness so starting conditions do a play a role, just a heads up.
If we go through May without any big problems I'm going live with this set.
I've come to the conclusion that a 1:3 ratio is ideal. So for example a lot size of 1% and a global TP of 3% etc.... I've stretched this to 2.5 and 7.5 on EURCHF without wiping in from Jan to Dec 2009.
I can tell you right off the bat the GU, EU, UJ, UCAD and AudUSD pairs will wipe pretty quickly. The loosing batches also have a cumulative effect on lot size and TP
I'm not super familiar with Steve's trade manager, I've played around with it however... Let me know how it goes.....
This backtests nicely so I'd recommend running it on the backtester before any demo accounts. I backtested this for over a month to get it just about right.
This is really based on chaos theory and all that goodness so starting conditions do a play a role, just a heads up.
If we go through May without any big problems I'm going live with this set.
Membre depuis Mar 13, 2010
posts 23
Apr 28, 2010 at 22:36
Thanks. Already , DD is growing pretty fast. I thought the hedge and the trade manager would control the beast. Oh, well.
Membre depuis Mar 13, 2010
posts 23
Apr 29, 2010 at 16:23
Joe: If one focuses on the single pair, Euro/CHF, and then really hits the Lot% hard, such as 4, then the limiting profit factor will be the frequency of the black swan event. If it presumably doubles the account each month, and a black swan event occurs twice each year, the the focus can be directed to controlling the % loss of the black swan event. If one sets a Management loss limit of 15 or 20%, then this will be a small price to pay for running a monthly or bi-monthly doubling machine.
How would you assess this logic?
How would you assess this logic?
Membre depuis Mar 13, 2010
posts 23
Apr 29, 2010 at 22:54
Joe: I was comparing points of entry with SSS on the five min. window Euro/CHF. The price channel stop might actually be a fairly useful filter, in preventing the occasional bad trade. I also activated the SSS close on signal change feature, as an experiment. Glenn
Apr 30, 2010 at 13:44
Winst,
1st point about lot%
At lot 4% you have 4x less wiggle room than at 1%. So our tolerance for pain will be minimal here, ie any big move will wipe us out. Our 'black swan' events are random, in 2008 for example we see more volatility in EURCHF, and get about 4 or 5 of these events throughout the year. 2009 is great and 2010 conditions are similar if not more flat than 2009.
The only way to really find out is to test test test test.
Backtest wise price channel stop doesn't work too well, at least with the settings I've tried. Even if you filter out 2 bad trades to 1 good trade ratio (2:1) or better., the fact that u filter out a good trade could increase our holding time which does two things, increases our risk and causes opportunity loss.
The idea here is to keep everything simple and elegant. I recommend watching a documentary called 'The Secret Life of Chaos', its on youtube. I came up with this after watching it....
my best,
-joe
1st point about lot%
At lot 4% you have 4x less wiggle room than at 1%. So our tolerance for pain will be minimal here, ie any big move will wipe us out. Our 'black swan' events are random, in 2008 for example we see more volatility in EURCHF, and get about 4 or 5 of these events throughout the year. 2009 is great and 2010 conditions are similar if not more flat than 2009.
The only way to really find out is to test test test test.
Backtest wise price channel stop doesn't work too well, at least with the settings I've tried. Even if you filter out 2 bad trades to 1 good trade ratio (2:1) or better., the fact that u filter out a good trade could increase our holding time which does two things, increases our risk and causes opportunity loss.
The idea here is to keep everything simple and elegant. I recommend watching a documentary called 'The Secret Life of Chaos', its on youtube. I came up with this after watching it....
my best,
-joe
Membre depuis Mar 13, 2010
posts 23
May 03, 2010 at 17:04
Joe: Do you know how often the DD exceeds ten percent, or something similar. My thinking is that I can kill off trades at a given level, presuming that each of those events might progress to a wipeout. I note with interest that XUX has encoded a 10% Global equity closure for Euro/CHF, in his setup for SSS 30 min. For example, perhaps the DD remains less than ten percent for 95% of all trades, then killing off those that exceed that amount would shoot the black swans before they can land. Very interested in you out of the box idea. Regards, Glenn
Membre depuis Mar 13, 2010
posts 23
May 03, 2010 at 21:06
Also, Joe, have you considered using the ATR for TP and SL? Presumably that would prevent any extended-time holding of positions, if it fails to hit the fib levels for exits. Sometimes it might be OK to have a loss, to clear the table.
*Lutilisation commerciale et le spam ne seront pas tolérés et peuvent entraîner la fermeture du compte.
Conseil : Poster une image/une url YouTube sera automatiquement intégrée dans votre message!
Conseil : Tapez le signe @ pour compléter automatiquement un nom dutilisateur participant à cette discussion.
