New EA - "Predator" - page 10
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Anyone know where I can find the explanation of the envelope 2.11. What triggers the trades etc... Thanks
envelope 2.11 thread in this elite section.
Besides all the links are in files thread including link to this Predator EA thread.
tf
Hi m8. There's really not much room for improvement with Predator. I think everyone has gone to the Goblin thread. Goblin pretty much represents the best of what Predator had to offer. With Goblin, the approach is to get away from this idea of multiple trade signal options and go with the one that seemed to work best in Predator.....JRSX & JVEL. From there, the focus is on improving the order management aspect via hedging. The other thing is that with the exception of a dedicated few, there just didn't seem to be a whole lot of interest in demo testing here with Predator to provide the kind of feedback needed to really improve this EA.
In response to this comment, i don't really think that JSRX and JVEL were the best choices. If they were then why don't they get you in the trade in the correct direction and keep you there. You are relying on the fact that the trade will go against you to make bigger amts of money. I'm not saying that I know of anything better but now that you have a system whereby you are protected against most losses, you should be looking for a way to get you in the trade in the correct direction and then build or at the very least protect what you gain. Like adding to a position as it increases in value, or exiting on a fast moving up or down bar and then re-evaluating the entry again, adding trailing stops to protect what you have, because once the trade moves into the postive, there is no reason to take a loss. I do realize you may not be doing any more posting of EA's but maybe you could incorporate these ideas in whatever you work on. The work done is still excellent but it's not complete till you finish the other half of the EA. Just my 2 cents worth. BTW Bipolar is a nice idea but it won't work in the long run. the trends will kill it and we both know that is the key to catch those trends. tf
Hi m8. There's really not much room for improvement with Predator. I think everyone has gone to the Goblin thread. Goblin pretty much represents the best of what Predator had to offer. With Goblin, the approach is to get away from this idea of multiple trade signal options and go with the one that seemed to work best in Predator.....JRSX & JVEL. From there, the focus is on improving the order management aspect via hedging. The other thing is that with the exception of a dedicated few, there just didn't seem to be a whole lot of interest in demo testing here with Predator to provide the kind of feedback needed to really improve this EA. In response to this comment, i don't really think that JSRX and JVEL were the best choices. If they were then why don't they get you in the trade in the correct direction and keep you there. You are relying on the fact that the trade will go against you to make bigger amts of money. I'm not saying that I know of anything better but now that you have a system whereby you are protected against most losses, you should be looking for a way to get you in the trade in the correct direction and then build or at the very least protect what you gain. Like adding to a position as it increases in value, or exiting on a fast moving up or down bar and then re-evaluating the entry again, adding trailing stops to protect what you have, because once the trade moves into the postive, there is no reason to take a loss. I do realize you may not be doing any more posting of EA's but maybe you could incorporate these ideas in whatever you work on. The work done is still excellent but it's not complete till you finish the other half of the EA. Just my 2 cents worth. BTW Bipolar is a nice idea but it won't work in the long run. the trends will kill it and we both know that is the key to catch those trends. tf
Your observations are pretty much valid. The development of Predator and Goblin were no more than an effort to improve on 10Point3 with regards to the trade signals. I'd disagree with you about JVEL & JRSX though...they've proven superior to the original standard MACD.
The answer to this quest is Goblin II which completely re-engineers the 10Point3 engine and provides risk/equity management, order optimization and hedging.
Your observations are pretty much valid. The development of Predator and Goblin were no more than an effort to improve on 10Point3 with regards to the trade signals. I'd disagree with you about JVEL & JRSX though...they've proven superior to the original standard MACD.
The answer to this quest is Goblin II which completely re-engineers the 10Point3 engine and provides risk/equity management, order optimization and hedging.
hi, bluto
are you intended to share Goblin II on public forum, elite section or private email ?
best regards, giapel
hi, bluto
are you intended to share Goblin II on public forum, elite section or private email ?
best regards, giapelUnfortunately, no. Not in the near future after some of the experiences I've had recently.
Dynamic pipstep/TP/SL
I was working with Bluto's Goblin and had started using ATR to assign values for pipstep based on the recent volatility of the pair.
I've had mixed results. I did also use similiar code to set the takeprofit and initialstop values.
I am posting my most recent Options.
This started as Goblin. The external's are fairly well explained if you open the .mq4 with the editor.
Anyway, if someone can get some ideas from this, great.
Also I had this version optimized for USDCAD 4h.
Used Phoenix indicators, not claiming any of it as mine.
I only wrote the code for the Dynamic stuff.
Tross
Replaced 2.11 with 2.12, it bugged me
Hi bluto,
The problem with these martingales is that they go fine until they blow up your account
. Do you know if somebody tried already to prevent this from happening by programming the martingale so that he does not always mutiplies the trades in the same direction. In this way an unsuccesfull buy order should be closed and followed by a (double) sell order. If that was also unsuccesfull, than closed and followed by a quadruple buy order etc.
Just an idea, but I can't program it.
JP
tf
ok then bluto, back to my original question, If they were then why don't they get you in the trade in the correct direction and keep you there. The macd got you in the trade too late but these do also. Even on the 4 hr charts the trend can reverse and you are still stuck with it moving against you. When that trend reverses is when you make money. As long as you are with the trend it will make money but not like it does when it moves against you. The logic is backwards. It's nice to reduce the risk when you are trading but too often, you are only increasing the risk by compounding the lots. If you got in the trade in the right direction to begin with, then a higher percentage of your trades would be profitable from the get-go. Yes I'm interested in the hedging EA. I have read the whole Kokas thread and I believe there is potential, but even then your risk is reduced if you get in with the correct direction. I will PM you when you let us know it is ready to be tested. And thanks, I'm just trying to stimulate more thought on this matter. tf
ok then bluto, back to my original question, If they were then why don't they get you in the trade in the correct direction and keep you there. The macd got you in the trade too late but these do also. Even on the 4 hr charts the trend can reverse and you are still stuck with it moving against you. When that trend reverses is when you make money. As long as you are with the trend it will make money but not like it does when it moves against you. The logic is backwards. It's nice to reduce the risk when you are trading but too often, you are only increasing the risk by compounding the lots. If you got in the trade in the right direction to begin with, then a higher percentage of your trades would be profitable from the get-go. Yes I'm interested in the hedging EA. I have read the whole Kokas thread and I believe there is potential, but even then your risk is reduced if you get in with the correct direction. I will PM you when you let us know it is ready to be tested. And thanks, I'm just trying to stimulate more thought on this matter. tf
Thought stimulation is a good thing!
Here's my take on a few of your observations:
1. The behavior of the original 10Point3 EA was such that whenever you dropped it on a chart and activated trading, you'd instantly get an order placed depending on the status of the MACD histogram regardless of where you were in the swing. In Goblin, regardless of when you activate the EA, a trade doesn't get placed until the next buy or sell signal is fired based on the Jurik RSX and Velocity indicator. That's a big difference right there. At least now we've made a more intelligent entry point using indicators that are much more responsive than the std. MACD. Of all things, this was the single biggest advantage that gave a lot of folks better results than with 10Point3.
2. Considering the above, that's where the excitement ends. What you've observed about Predator & Goblin still having issues with getting into the trade at the right time is half correct. The initial entry will be correct but from there, it's a crapshoot for each successive swing because everybody has their own recipe for Maxtrades & PipSteps. The EA looses synch because a fixed Maxtrades value prohibits the EA from responding to a new swing when Maxtrades hasn't been exhausted. This is why you'll see that 5th or 6th buy order being placed when the trend has clearly swung to the downside and visa-versa.
3. I never liked the faux-Martingale progression methodology of the 10Point3engine but that was an another problem for another day, meriting a complete overhaul to do it right. It's more suited to countertrend EA's imo. If you're coming out of the starting block on a fresh signal, you should be wagering your larger chunk of capital on the prospects of a quick short move at best, scaling in or out and taking profits as the trend ensues and the signal holds. 10Point3 starts out with small probes and builds as the luck holds out. Kinda dumb.
4. So what's the fix for this quandry? An overhaul. An entirely new EA, but built on some parallel concepts.
- Jurik indicators for signals. I love these things and swear by them.
- Dynamic expandable/collapsable pipstep interval settings calculated in realtime based on near-term price volatility. These are what keep you out of trouble when towering blastoff bars appear and likewise allow you to scoop small profits during ranging periods. Forget about trying to avoid the news. News is what makes you money.
- Separate maxtrades counters for buy and sell orders that keep you from getting stuck waiting for a cycle to complete while the swings change. Intelligent order placement in conjunction with the signal direction.
- Intelligent predictive money management that factors in the entire lot progression cost, not just the original order.
- Automatic hedging of order cycles that get off track based on a pre-determined threshold to minimize losses at a static value and avoid continued drawdown until swing reversal. This is not the same thing as what kokas has with correlated pairs in his Hedge EA 6.0.
This is the essence of Goblin II which I've actually renamed as "JitterBug" to get away from the entire Goblin thing which has now become a bit of a polluted and convoluted mess over on the regular forum. Not to blow any smoke, but I started a live $8K pilot with JitterBug on 12/19 and it is on the verge of having tripled the starting capital.