Stable & safe TREND strategy - 30% to 40% yearly profits | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Need Help: Monthly, Quarterly and Yearly Profits Breakdown 6 replies

Trade suppressed FX pairs - collect safe profits 8 replies

ECN && STP && Scalping && Hedging - Broker Review - Oct 2009 24 replies

Stable profits weekly - Monitoring transactions 18 replies

8- Hour's & Yearly Timeframe? 7 replies

Attachments: Stable & safe TREND strategy - 30% to 40% yearly profits
Exit Attachments

Stable & safe TREND strategy - 30% to 40% yearly profits

  • #21
  • Dec 12, 2020 2:03pm Dec 12, 2020 2:03pm
Here some another version using ADR for GOLD trading 11 years backtests results:

Attached Image (click to enlarge)
Click to Enlarge Name: gold_adr.png Size: 32 KB

Attached File(s)
File Type: xlsx gold.xlsx   595 KB | 323 downloads

This was asked via private message - just post results here as PM cant be sent files.
  • #22
  • Dec 12, 2020 3:12pm Dec 12, 2020 3:12pm
Hi Gepards, your system looks really nice, but experience tells me to be careful with this type of MM. I follow the post, as I follow everyone else. I keep telling you, keep your feet on the ground, 'not all that glitters is gold'.
NB: at least tell me that the backtest was performed with real ticks ?!
Best regards
  • #23
  • Dec 13, 2020 1:11am Dec 13, 2020 1:11am
1. How this EA handle flash crashes?
2. Consider adding trailing to let winners run for more profits
3. Consider adding break even turn on at level N so get out of too risky situations asap
4. Consider adding some more specific conditions. Now "run X pips" seems random. From where run? At market opening? Should be repeatable with clear rules. For example "run X pips from moving average" or "run X pips from London opening"

Keep improoving. Idea is good. Maybe hunt for ranges breakouts with Boilinger bands - let range live several day's - not jump in too early. Just ideas...
  • #24
  • Dec 13, 2020 9:19am Dec 13, 2020 9:19am
Tried implement various additional parameters and conditions but seems thats overcomplicating and no better results in backtests.

But...

Tried 1 suggestion - close losing deal and then open oposit deal. So new version of EA. Idea the same - just not keep losing trades opened.

Here some results:


GBPUSD:
Attached File(s)
File Type: xlsx GBPUSD_distance_60.xlsx   723 KB | 391 downloads
Attached File(s)
File Type: xlsx GBPUSD_distance_160.xlsx   128 KB | 329 downloads


EURUSD:
Attached File(s)
File Type: xlsx EURUSD_distance_20.xlsx   1.0 MB | 374 downloads

Seems this is much better because no huge volumes. So it is posible to scale. Start on cent account, then go on ECN account.

Added EA in post Nr1.
  • #25
  • Dec 13, 2020 9:40am Dec 13, 2020 9:40am
Thinking... backtests can't guarantee future... therefore better to go with multiple setups on 1 account.

For example, if max losing serie in 11 years was let's say 7 then must have backup money to survive at least 10. Life is life.

So with max drawdown 144 and profits 1200 usd I need account with minimum 1000 usd. When run 5 different setups with similar profitability/drawdown then should be about 50-60% yearly profits.

Need to calculate/combine varios setups on 1 account scenarious...
  • #26
  • Edited 10:38am Dec 13, 2020 9:57am | Edited 10:38am
Quoting joyny
1. How this EA handle flash crashes? 2. Consider adding trailing to let winners run for more profits 3. Consider adding break even turn on at level N so get out of too risky situations asap 4. Consider adding some more specific conditions. Now "run X pips" seems random. From where run? At market opening? Should be repeatable with clear rules. For example "run X pips from moving average" or "run X pips from London opening" Keep improoving. Idea is good. Maybe hunt for ranges breakouts with Boilinger bands - let range live several day's - not jump...
tnx for advices.

1. EA works with open prices (if parameter is_open_price_mode set) - with open prices on M5 timeframe it will skip 5 minutes... if the price goes significantly wrong way EA will open opposite order and will have a bigger losing deal which must be then covered... 2016 7 October GBPUSD flashcrash was 670 pips... a huge one... it partly recovered in several minutes...

As we see in backtests - EA survived that flash-crash somehow... But maybe luck.. maybe level 1... if would be level 7 with huge volume maybe would be hit hard.

What could be the solution for flash-crash?

2. Added trailing - optimizations didn't find better setups with that

3. Added break-even - optimizations didn't find better setups with that

4. This was a really good suggestion - implemented in EA version Nr2. Now exact hours used to open deals. No more "randomness" you mentioned. So EA no more depending on when it is turned on...
  • #27
  • Dec 13, 2020 10:25am Dec 13, 2020 10:25am
Quoting TRexTrader
Hi Gepards, your system looks really nice, but experience tells me to be careful with this type of MM. I follow the post, as I follow everyone else. I keep telling you, keep your feet on the ground, 'not all that glitters is gold'. NB: at least tell me that the backtest was performed with real ticks ?! Best regards
Here real ticks result for EA version Nr2 GBPUSD - have real ticks from 2016:

Attached Image (click to enlarge)
Click to Enlarge Name: gbpusd_distancer_real_ticks.png Size: 41 KB

Here report file:
Attached File(s)
File Type: xlsx GBPUSD_distance_160_real_ticks_no_open_prices.xlsx   71 KB | 336 downloads

Here .set file:
Attached File(s)
File Type: txt gbpusd_distance_160_no_open_prices.txt   2 KB | 290 downloads

The thing is... my EA has the parameter "is_open_prices_used". When true - then EA works on open prices only - this is for fast optimizations. When go on real account this parameter I will turn off. Otherwise, 5 minutes of skipping can give huge disadvantages...
  • #28
  • Edited 2:07pm Dec 13, 2020 10:57am | Edited 2:07pm
An important concept of optimizations and data quality.

EAs have parameter "is_open_prices_only" (as 1st parameter). It is used for optimizations to get them fast and to see 10 years best setups very fast (minutes). EA is specially designed so that if the parameter is true then all logic happens only on new bar opening moment (according to timeframe - if M5 then every 5 minutes - everything between is ignored...)

When working with real ticks (or go with live trading) that parameter must be set to false... Otherwise EA will work with 2x more drawdowns.

See here examples for EA version Nr2 GBPUSD:

Optimizatin with open prices:
Attached Image (click to enlarge)
Click to Enlarge Name: gbpusd_distancer_open_prices_only.png Size: 42 KB

Real ticks with is_open_proces_only set to true:
Attached Image (click to enlarge)
Click to Enlarge Name: gbpusd_distancer_real_ticks_open_prices_set.png Size: 41 KB

Real ticks - correct setup is_open_prices_only = false (real life mode):
Attached Image (click to enlarge)
Click to Enlarge Name: gbpusd_distancer_real_ticks_open_prices_off.png Size: 41 KB

Notice the difference... and be carrefull.
  • #29
  • Dec 14, 2020 11:24am Dec 14, 2020 11:24am
Quoting gepards
{quote} Here real ticks result for EA version Nr2 GBPUSD - have real ticks from 2016: {image} Here report file: {file} Here .set file: {file} The thing is... my EA has the parameter "is_open_prices_used". When true - then EA works on open prices only - this is for fast optimizations. When go on real account this parameter I will turn off. Otherwise, 5 minutes of skipping can give huge disadvantages...
Very nice. I continue to follow you willingly
  • #30
  • Dec 14, 2020 11:29am Dec 14, 2020 11:29am
https://www.myfxbook.com/members/AJ_...system/6985760

Maybe you can be this guy too, who knows.

Just throw a 200$ into trash 1:500 offshore account and go on.


AND PLEASE, only rely on %99+% modeling quality backtests, anything below can be misleading.

Also, if you are serious about trying this thing, try to skip fridays & weekends.

You don't wanna be dealing with bullshit gaps, it'll kill you.
  • #31
  • Dec 14, 2020 12:31pm Dec 14, 2020 12:31pm
Quoting PixelSniper
https://www.myfxbook.com/members/AJ_...system/6985760 Maybe you can be this guy too, who knows. Just throw a 200$ into trash 1:500 offshore account and go on. AND PLEASE, only rely on %99+% modeling quality backtests, anything below can be misleading. Also, if you are serious about trying this thing, try to skip fridays & weekends. You don't wanna be dealing with bullshit gaps, it'll kill you.
That same guy has a Holy Grail account with a 92 profit factor. He’s definitely doing something right.
  • #32
  • Dec 14, 2020 12:34pm Dec 14, 2020 12:34pm
Quoting ChicagoRob
{quote} That same guy has a Holy Grail account with a 92 profit factor. He’s definitely doing something right.
He is proving all the biased people wrong.
Yeah, his accounts are pretty nice.
Fun fact, once on myfxbook there was a guy who traded 11m$ account, and he hedged too. Then he lost %55 of his capital, because he was martingaling. ^^
Hedging is an unholy sword, it'll hurt your enemies by a lot but maybe one day it'll hurt you too, who knows.
  • #33
  • Edited 3:05pm Dec 14, 2020 2:51pm | Edited 3:05pm
I'm now stuck a little bit... comparing results with is_open_prices_only = true/false results..

EA behaves differently with this option. When open prices used deals are closed sometimes with 30 or more pips in more profits... (the same for losing/new level opening).

And logic is - as the deal is closed EA waits for X pips in any direction to open the next deal in the same direction.

Therefore on M5 bars opening mode results will be completely different than in all ticks mode. Different deals are picked up...

Definitely with all ticks mode drawdowns will be smaller because losing deals will be closing precisely on the preset level.

But with open prices profits (and drawdown) more when comparing results.

GBPJPY.. with open prices max losing serie is 9. But when mode = false then the losing serie is 16. That was a long ranged market from 12 October 2016 till 25 October 2016.. Why "precise mode" performed so badly.. need to think... too big difference 9 vs 16. Open prices skipping 5 minutes somehow give better results... skips bad deals too...
  • #34
  • Dec 14, 2020 9:03pm Dec 14, 2020 9:03pm
Quoting gepards
I'm now stuck a little bit... comparing results with is_open_prices_only = true/false results.. EA behaves differently with this option. When open prices used deals are closed sometimes with 30 or more pips in more profits... (the same for losing/new level opening). And logic is - as the deal is closed EA waits for X pips in any direction to open the next deal in the same direction. Therefore on M5 bars opening mode results will be completely different than in all ticks mode. Different deals are picked up... Definitely with all ticks mode drawdowns...

Don't even worry about it, even if the max series are n9-n20 on backtesting, it is decent and a good account can survive that.
Now let's add our own approach and tricks to the market.
I have a question for you.
What are the chances of a currency pair staying at the same level when it is at the top or bottom?
Rather than taking random entries we can simply setup an ambush, that way we'll skip ranges right.
A good traders approach will definetly increase the win rate, the more you win the more you skip the bullshit scenarios + the more risk appetizer you can take(e.g. instead of starting with 00.01 you can take off with 00.05 everytime and when the accounts great enough keep scalping with 00.10's)
In my opinion it is better to manual trade this strategy. Have your own approach, and hedge.. simply. It can make tons, the market does not have rules,
so why should you have so many rules.. play dirty as the markets are.
There is also another trick, enter the trade when there is a huge spike up or down, the price won't stay there for long. Your hedges will be banked quickly.
Simple market psychology.. i hope it helps
  • #35
  • Dec 15, 2020 1:55pm Dec 15, 2020 1:55pm
Quoting gepards
GBPJPY.. with open prices max losing serie is 9. But when mode = false then the losing serie is 16. That was a long ranged market from 12 October 2016 till 25 October 2016.. Why "precise mode" performed so badly.. need to think... too big difference 9 vs 16. Open prices skipping 5 minutes somehow give better results... skips bad deals too...
Now clear why difference... somehow on MT5 for example at 16:55:00 there are different data for "Open prices only" compared with "Every tick based on real ticks"...

Attached Image (click to enlarge)
Click to Enlarge Name: gbpjpy_open_prices.png Size: 22 KB
Attached Image (click to enlarge)
Click to Enlarge Name: gbpjpy_real_ticks.png Size: 23 KB


I was expecting that on round minutes like 16:05:00, 16:10:00, 16:15:00 etc there should be the same price (ask/bid) no matter what data modeling in strategy tester chosen... but that's not so...

was doing according to this info regarding Open prices... "Some traders do not wish to depend on particularities of intrabar modeling, so they create Expert Advisors trading on the bars already completed. The fact that the current price bar is fully completed can only be known when the next one appears. These are the Expert Advisors, for which the mode of "Open Price" modeling is intended."

But the problem is those "bars completed" are different... and when run with real tick EA behaves different. But should be the same... need to clarify this.
  • #36
  • Dec 15, 2020 3:59pm Dec 15, 2020 3:59pm
Quoting gepards
{quote} Here real ticks result for EA version Nr2 GBPUSD - have real ticks from 2016: {image} Here report file: {file} Here .set file: {file} The thing is... my EA has the parameter "is_open_prices_used". When true - then EA works on open prices only - this is for fast optimizations. When go on real account this parameter I will turn off. Otherwise, 5 minutes of skipping can give huge disadvantages...
Not enough sample size. You need atleast 1000 trades to be meaningful. If not thousands... cheers
When you see it, BET big.
  • #37
  • Dec 16, 2020 5:01am Dec 16, 2020 5:01am
When a backtest looks like this you know that it's a crazy martingale that is going to crash.
Did you take into consideration slippage, spread, commission...
You can use martingale with a 100$ gambling account but will you do it with à 10m$?

I agree with you that the market is always going somewhere.

Instead of using martingale that will for sure blow your account you can use an other approch:
Choose a 10 pips range on any pair at the begining of your trading session for exemple 1.3000 to 1.3010
Buy at 1.3010 with a 10 stop and no limit
Sell at 1.3000 with a 10 stop and no limit
Their can and their will of course be loosing trades but you can be sure that you will win at some point and that you won't blow your account. Risk only 1% for each position or less.

Simple strategie: You loose 10 your next tp will be 20, you lose 20 your next tp will be 30. You can have a lot of loosing trade but you won't blow your account in 10 trades like a martingale. You can even trade on every 10 pips range but it's hard to follow after.
  • #38
  • Dec 16, 2020 5:56am Dec 16, 2020 5:56am
Quoting Hyperion75
Simple strategie: You loose 10 your next tp will be 20, you lose 20 your next tp will be 30. You can have a lot of loosing trade but you won't blow your account in 10 trades like a martingale. You can even trade on every 10 pips range but it's hard to follow after.
Interesting strategy and quite simple. Will try to implement and share backtest results. tnx
  • #39
  • Edited 6:22am Dec 16, 2020 6:04am | Edited 6:22am
Attached Image (click to enlarge)
Click to Enlarge Name: 0112.png Size: 202 KB


Quoting gepards
{quote} Interesting strategy and quite simple. Will try to implement and share backtest results. tnx
I use it on gbpusd
I start trading at 8gmt
I trade every 10 pips range of the day
I don't have a rule for when to stop but usualy around 4-5pm

Here is an exemple of 01/12/2020
red- loosing trades
blue- winning ones

I enter trade manualy to lower spread and slippage but you can enter on orders
  • #40
  • Dec 16, 2020 1:43pm Dec 16, 2020 1:43pm
Quoting Hyperion75
{image} {quote} I use it on gbpusd I start trading at 8gmt I trade every 10 pips range of the day I don't have a rule for when to stop but usualy around 4-5pm Here is an exemple of 01/12/2020 red- loosing trades blue- winning ones I enter trade manualy to lower spread and slippage but you can enter on orders
backtested. with 10 pips range no good consistent results. But here 4 years real ticks results with 20 pips range and only trade on Mondays, Tuesdays GBPUSD from 6AM London time till 12 PM (included) and close all deals no mater what at 2 PM.

Attached Image (click to enlarge)
Click to Enlarge Name: gbpusd_try_mon_tue.png Size: 114 KB

Attached Image (click to enlarge)
Click to Enlarge Name: gbpusd_try_mon_tue_stat.png Size: 244 KB


But maybe EA works little different than you do manually..It counts losing series for BUY and SELL seperate.. you do the same?

Here report:

Attached File(s)
File Type: xlsx gbpusd_2016_real_ticks.xlsx   232 KB | 320 downloads


Here no martingale used..

I didn made optimizations. Jsut picked up 10 pips and 20 pips and saw bad week days which excluded. Optimization with real ticks runs more than 44 hours... welll... thats too long..

Will look is it posible to find optimizations with open prices... then re-check with real ticks.
2 traders viewing now