{quote} I don't know what to say MathTrader but..
..
..
Trading is the hardest way to make easy money...
- #161
- Apr 28, 2015 6:46am Apr 28, 2015 6:46am
- Joined Aug 2014 | Status: Trading | 2,163 Posts
- #162
- Apr 28, 2015 6:55am Apr 28, 2015 6:55am
Alright, some stats.
TF: 1 min
SL: 20 points
TP: 20 points
28 different pairs tested.
A total of about 1000 trades taken during the past 24 hours.
On pairs with higher spreads this was a disaster, more or less. 90%-100% losers. Some trades where closed immediately due to the fact that the spread was above 20 points at the time when the trades was taken. (GPBCHF was one of them, those pairs where removed from the test when discovered)
This was not a surprise, but it proves the point that a spread filter would be a good idea. There will be times (news events for example) when the spreads widen even on high liquidity pairs such as EURUSD.
EURUSD was the best performer and gave me 53% winning trades.
Another thing i noticed is that many trades went straight into profit, just to turn around and close at a loss. I beleive that a really tight trailing stop loss could improve the winning rate.
I have set up a few new tests with different settings and one where i have attached a trade manager that immediately trails the stop loss. Lets see what this gives us
Oh, and keep up the good work abokwaik, this is really exciting stuff!
TF: 1 min
SL: 20 points
TP: 20 points
28 different pairs tested.
A total of about 1000 trades taken during the past 24 hours.
On pairs with higher spreads this was a disaster, more or less. 90%-100% losers. Some trades where closed immediately due to the fact that the spread was above 20 points at the time when the trades was taken. (GPBCHF was one of them, those pairs where removed from the test when discovered)
This was not a surprise, but it proves the point that a spread filter would be a good idea. There will be times (news events for example) when the spreads widen even on high liquidity pairs such as EURUSD.
EURUSD was the best performer and gave me 53% winning trades.
Another thing i noticed is that many trades went straight into profit, just to turn around and close at a loss. I beleive that a really tight trailing stop loss could improve the winning rate.
I have set up a few new tests with different settings and one where i have attached a trade manager that immediately trails the stop loss. Lets see what this gives us
Oh, and keep up the good work abokwaik, this is really exciting stuff!
{quote} Thanks Traffex, What you need to do Traffix is to run both N4S_03_Auto and N4S_03_Auto_Contra with same settings, same pairs on two separate account and report the results at end of day. God Bless
Intermediate result after 4 h: The "contras" are in front (63 % winrate, 50 % the normal version)
- #164
- Apr 28, 2015 7:11am Apr 28, 2015 7:11am
- Joined Aug 2014 | Status: Trading | 2,163 Posts
{quote} Intermediate result after 4 h: The "contras" are in front (63 % winrate, 50 % the normal version)
Could you report how many trades generated your results?
Trading is the hardest way to make easy money...
- #166
- Apr 28, 2015 7:27am Apr 28, 2015 7:27am
- Joined Sep 2012 | Status: AlgoTrader | 3,610 Posts
No guts, no glory
- #167
- Apr 28, 2015 7:40am Apr 28, 2015 7:40am
It seems that there is a probability distribution over two possibilities that can happen after an impulse: continuation and reverse. I am running the EA on 5 low spread pairs (SL/TP=100 points), so far breakeven. I'm recording the trades, signals, a moving average value, and a few other quantities to analyze later when I have 300-400 trades. This way, we may be able to improve the strategy. {image}
- #172
- Apr 28, 2015 8:32am Apr 28, 2015 8:32am
- Joined Sep 2012 | Status: AlgoTrader | 3,610 Posts
Good day. I am new here. I would simply like to know abokwaik why hasnt your system worked out during trading demo
I can summarize few reasons here ...
1. Robot was running in aggressive mode (taking the first signal - no filtration), High leverage , for a worst case scenario.
2. Money Management is not built in.
3. Risk Management is not built in.
4. System still in development phase.
Or maybe it was a bad idea after all.
I am out of here.
No guts, no glory
- #173
- Apr 28, 2015 9:22am Apr 28, 2015 9:22am
{quote} Yes, you are correct. If i take the EURUSD and look at the first hour from London i get 67% wins, the first hour from NY open gives me 80% win. If I look at the first 3 hours from open that gives me 42% win on london and 62% win on NY.
- #174
- Apr 28, 2015 9:23am Apr 28, 2015 9:23am
{quote} Good Day English, I can summarize few reasons here ... 1. Robot was running in aggressive mode (taking the first signal - no filtration), High leverage , for a worst case scenario. 2. Money Management is not built in. 3. Risk Management is not built in. 4. System still in development phase. Or maybe it was a bad idea after all. I am out of here.
Leading by "Being an example" is hard to find then actual examples.
- #175
- Apr 28, 2015 9:25am Apr 28, 2015 9:25am
{quote} Good Day English, I can summarize few reasons here ... 1. Robot was running in aggressive mode (taking the first signal - no filtration), High leverage , for a worst case scenario. 2. Money Management is not built in. 3. Risk Management is not built in. 4. System still in development phase. Or maybe it was a bad idea after all. I am out of here.
Could you set the mode ...aggressive or no in the next version ?
- #177
- Apr 28, 2015 12:04pm Apr 28, 2015 12:04pm
- Joined Sep 2012 | Status: AlgoTrader | 3,610 Posts
{quote} Many thx abokwaik....you're a great coder Could you set the mode ...aggressive or no in the next version ?
Trading Hours and Entry Options (Mode) , along with other features will be included in next version as mentioned in Post # 138
I am waiting for more input before next version, lets give the bot a chance to allow for further analysis ...
I am not in a rush ... are you ... ?
Oh I forgot we are in Neeed4Speeed not in the lab ..
No guts, no glory
- #178
- Apr 28, 2015 12:05pm Apr 28, 2015 12:05pm
- Joined Sep 2012 | Status: AlgoTrader | 3,610 Posts
{quote} Thank you very much for your reply. I for one love to see new systems, only because it allows one to learn. Keep up the good work my friend.
No guts, no glory
- #179
- Apr 28, 2015 12:08pm Apr 28, 2015 12:08pm
- Joined Sep 2012 | Status: AlgoTrader | 3,610 Posts
Hi everyone, first need to thank Abukwaik for his great job, all other guys for the improvements and the notes they added to make it better. I will start to use the EA from now, but I will use the Camarilla daily/hourly lines as exit levels, I usually use them when I am scalping in my regular trading days. I will try also to add some filters on each pair to give it more efficient. Keep you updated
Glad to see you here ...
Using pivots for targets would be a nice touch ... Share with us your findings
Thanks,
Khalil
No guts, no glory
- #180
- Edited 12:27pm Apr 28, 2015 12:13pm | Edited 12:27pm
- Joined Sep 2012 | Status: AlgoTrader | 3,610 Posts
Just for Information (As some people might check the TE).
Latest closed trades were entered by the robot and closed by me manually (partial profit, trailing, BE ... etc).
Current trades and coming trades tonight will be completely managed by the robot. I raised the Threshold for 3 pairs in order to only take SERIOUS moves. I also increased TP to 200 and SL to 150. I am at home now and the robot is by itself (himself) ... whatever
Latest closed trades were entered by the robot and closed by me manually (partial profit, trailing, BE ... etc).
Current trades and coming trades tonight will be completely managed by the robot. I raised the Threshold for 3 pairs in order to only take SERIOUS moves. I also increased TP to 200 and SL to 150. I am at home now and the robot is by itself (himself) ... whatever
No guts, no glory