Here's the new "Goblin" ea...little brother to Predator - page 64
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
I've been playing around with Goblin code to see if putting a limit on TrendVal variable that could prevent trades during volatile market. I was able to backtest this last week of 2006.12.03 - 2006.12.08 with the NFP news and results were very similar to Demo results. With TrendVal limit, Goblin was able to avoid the Big Hit from NFP news on Friday. Trade volume was down ~40%, but the ending balance was good and MaxDrawDown% decreased from ~27% to ~1.4%
Goblin-H1-GU-Exp1 is backtest as is.
Goblin-H1-GU-Exp2 is with TrendVal limit.
I don't know how beneficial this TrendVal limit will be for future volatile markets, but just found it interesting for last week.
Wackena
Wackena,
What TrendVal settings did you test ?
While certainly not a replacement for the hedging strategy I'm building into Goblin, I've been tinkering with another algorithm to anticipate these huge account wrecking moves. It goes something like this....the EA constantly monitors the Bid/Ask spread and when it detects unusual widening which typically begins to occur well ahead of big news announcements, it dynamically reduces MaxTrades to some proportional value. In other words, Maxtrades is always scaled to the normalized spread....this keeps you from being trapped in a relentless countermove buy/sell cycle and minimizes the number of pending orders until you can jump on the new move. Just an idea.
Of course, this is not an alternative to the outstanding advice Janus has provided here on leverage management and other techniques. I'm glad he has provided his valuable insight and participates in the Goblin trials.
Wackena, What TrendVal settings did you test ?
I used 0.03 for Buys and -0.03 for Sells.
Wackena
I used 0.03 for Buys and -0.03 for Sells. Wackena
hi Wackena,
i am testing goblin with fibonacci progressione, could you post me goblin-5 related to your backtesting ... , in my EA there is not Gap-setting
thank you
giapel
hi Wackena,
i am testing goblin with fibonacci progressione, could you post me goblin-5 related to your backtesting ... , in my EA there is not Gap-setting
thank you
giapelpluto is the originator of this thread and I do not want to post non-authorized versions of Goblin. I will post ideas and concepts for consideration, but it must be pluto's decision whether to use or not.
Wackena
pluto is the originator of this thread and I do not want to post non-authorized versions of Goblin. I will post ideas and concepts for consideration, but it must be pluto's decision whether to use or not. Wackena
It's....ehrrr...."bluto".
LMAO.
It's....ehrrr...."bluto". LMAO.
ROFLMAO...
you planet you,
Ok.
I wrote some hedging and reversal into GoblinFibo1.2
I tried it two different ways.
First Way
2. Place first order normally
3. If second order is required place it opposite
4. Reverse again for every subsequent order.
This showed decent results with the right settings.
Takeprofit was lower, Stoploss was higher, and I used MaxLossPerOrder.
Profitable trades on long/short and both were over 90%
Drawdowns were 6-15% depending on settings.
Testing with 99% modelling.
Second Way
1. Place first order normally
2. If second order is required, check trend, place order appropriately
3. Check trend before placing subsequent orders as required.
Results here were not as good.
Tried a variety of settings, still seemed to take some very large account hits.
I wanted to see what direction you were going with your next revision.
Also wanted to get this info out, as I'm too busy to post many results or versions. Moving, xmas, etc. etc.
Tross
Please share with us !
you planet you,
Ok.
I wrote some hedging and reversal into GoblinFibo1.2
I tried it two different ways.
First Way
2. Place first order normally
3. If second order is required place it opposite
4. Reverse again for every subsequent order.
This showed decent results with the right settings.
Takeprofit was lower, Stoploss was higher, and I used MaxLossPerOrder.
Profitable trades on long/short and both were over 90%
Drawdowns were 6-15% depending on settings.
Testing with 99% modelling.
Second Way
1. Place first order normally
2. If second order is required, check trend, place order appropriately
3. Check trend before placing subsequent orders as required.
Results here were not as good.
Tried a variety of settings, still seemed to take some very large account hits.
I wanted to see what direction you were going with your next revision.
Also wanted to get this info out, as I'm too busy to post many results or versions. Moving, xmas, etc. etc.
TrossHello Tross,
i really like your goblin Fibo1.2 or 1.1 and i would like to test your EA with the hedge protection.
But anyway as i can't program well i would understand if u will refuse .
Merry christmas Tross and all !
