what can we do with 90% winrate? | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Made a 100% Winrate EA 26 replies

Naustingfx End-Of-Day Trading System Journal @ 68.26% WinRate 34 replies

Winrate >90% - Unbelievable 4 me: omg 8 replies

MACD +60% Winrate 1:1.5 RR 63 replies

Statistical Gambling System (High Winrate, but NO EDGE) 372 replies

Attachments: what can we do with 90% winrate?
Exit Attachments

what can we do with 90% winrate?

  • #21
  • Oct 21, 2015 6:31am Oct 21, 2015 6:31am
Quoting PipMeUp
How are the gaps handled?
I think the presence of gaps will enhance the profitability ! as it is counter trend.
No guts, no glory
  • #22
  • Oct 21, 2015 6:31am Oct 21, 2015 6:31am
No HG here. From April 2007 to today on E/U daily the winrate I get is 72.1%.
No greed. No fear. Just maths.
  • #23
  • Oct 21, 2015 6:35am Oct 21, 2015 6:35am
Why don't we try the opposite (follow the previous candle direction) ?
And maybe also with a reverse RR (profit at 1.618 and SL at 0.8 (or 0.38 for a 1:1 RR)).

Edited
No guts, no glory
  • #24
  • Oct 21, 2015 6:47am Oct 21, 2015 6:47am
i make a quick back test on gbp/usd start from august till now
it's about 57 candles with 6 stops and 51 wins
Attached Image (click to enlarge)
Click to Enlarge Name: GBPUSDDaily.png Size: 38 KB
  • #25
  • Oct 21, 2015 7:19am Oct 21, 2015 7:19am
Here is a draft EA to test different ideas.

Inputs :

// Magic Number (Unique for each chart) ---> default 40851
// Risk Percentage ---> default 0.01
// Stop Percentage ---> default 0.62
// Target Percentage ---> default 0.2
//Follow Previous Candle --- > default false
Attached File(s)
File Type: ex4 Retracer_Extender.ex4   9 KB | 581 downloads
No guts, no glory
  • #26
  • Oct 21, 2015 7:24am Oct 21, 2015 7:24am
Quoting abokwaik
Here is a draft EA to test different ideas. Inputs : // Magic Number (Unique for each chart) ---> default 40851 // Risk Percentage ---> default 0.01 // Stop Percentage ---> default 0.62 // Target Percentage ---> default 0.2 //Follow Previous Candle --- > default false {file}
Where? Has anyone built an EA based on this idea yet?
  • #27
  • Oct 21, 2015 7:38am Oct 21, 2015 7:38am
EUR/USD 2015 - EA sample tests with different inputs for Stop and Target, risk % is fixed at 3%.

Attached Image (click to enlarge)
Click to Enlarge Name: sample_test_results.PNG Size: 79 KB
No guts, no glory
  • #28
  • Oct 21, 2015 7:49am Oct 21, 2015 7:49am
EUR/USD 2015 test with inputs suggested by roughtrader.
Stop = 0.62
Target = 0.2
Follow Candle Direction = False

The EA uses the following assumptions:
- Close previous running trade that didn't hit SL or TP at next daily open

Note that my broker's daily candle starts at GMT+3.

Attached Image (click to enlarge)
Click to Enlarge Name: re_EURUSD_2015_thread_settings.PNG Size: 49 KB
No guts, no glory
  • #29
  • Edited 8:23am Oct 21, 2015 7:59am | Edited 8:23am
Here is the first version of the EA I developed for this strategy.
Attached Image (click to enlarge)
Click to Enlarge Name: opt.png Size: 38 KB
Attached File(s)
File Type: ex4 MathTrader7_RoughEA.ex4   53 KB | 510 downloads

Some notes:
- I backtested from 2000 to 2015.
- The EA doesn't close a trade if it doesn't hit SL/TP at the end of the day.

@Khalil: I don't know why we get different result
Trading is the hardest way to make easy money...
  • #30
  • Oct 21, 2015 8:02am Oct 21, 2015 8:02am
Quoting MathTrader7
Here is the first version of the EA I developed for this strategy. {image} {file} @Khalil: I don't know why we get different result
Looking at the number of trades in your test, it seems that you run it for several years( 10 years maybe), mine was for 2015 only.
No guts, no glory
  • #31
  • Oct 21, 2015 8:02am Oct 21, 2015 8:02am
Quoting abokwaik
The EA uses the following assumptions: - Close previous running trade that didn't hit SL or TP at next daily open
I let them hit the SL or TP.

Quoting zelenko
i make a quick back test on gbp/usd start from august till now it's about 57 candles with 6 stops and 51 wins {image}
My candles open at 00:00 GMT. I skip week-end candles. I also skip big candles. I get these losers (MM=fixed 1 uLot):

LONG GBP/USD 1 Close: Wed, 12/08/2015 01:28:52.708 at 1.55594 PnL:-2.51 USD
LONG GBP/USD 1 Close: Wed, 26/08/2015 11:02:26.231 at 1.56161 PnL:-8.19 USD
LONG GBP/USD 1 Close: Thu, 03/09/2015 08:30:01.354 at 1.52327 PnL:-7.08 USD
LONG GBP/USD 1 Close: Fri, 04/09/2015 07:07:05.160 at 1.52199 PnL:-3.96 USD
SHORT GBP/USD 1 Close: Tue, 08/09/2015 05:38:32.135 at 1.53425 PnL:-6.92 USD
SHORT GBP/USD 1 Close: Thu, 17/09/2015 18:46:24.362 at 1.56285 PnL:-11.77 USD
SHORT GBP/USD 1 Close: Fri, 18/09/2015 08:46:41.835 at 1.56203 PnL:-5.55 USD
LONG GBP/USD 1 Close: Wed, 23/09/2015 13:01:52.168 at 1.52374 PnL:-11.63 USD <-- took a 4.2 pip gaps beyond the SL (NFP Friday maybe)
SHORT GBP/USD 1 Close: Fri, 02/10/2015 08:30:00.165 at 1.51641 PnL:-2.78 USD
SHORT GBP/USD 1 Close: Wed, 07/10/2015 07:55:39.803 at 1.52832 PnL:-5.83 USD
SHORT GBP/USD 1 Close: Tue, 20/10/2015 07:13:49.842 at 1.55003 PnL:-3.14 USD

This is the graphical view of the trades:
Attached Image (click to enlarge)
Click to Enlarge Name: GBPUSD-D1.png Size: 22 KB
No greed. No fear. Just maths.
  • #32
  • Oct 21, 2015 8:08am Oct 21, 2015 8:08am
This is a visual from EU 2015 test.
@roughtrader : kindly confirm the logic.

Attached Image (click to enlarge)
Click to Enlarge Name: EURUSDDaily.png Size: 59 KB
No guts, no glory
  • #33
  • Oct 21, 2015 8:31am Oct 21, 2015 8:31am
Quoting abokwaik
This is a visual from EU 2015 test. @roughtrader : kindly confirm the logic. {image}

I think it would be better to backtest using the hourly data for daily trading. It would give much more error free simulation, and more accurate I think.
  • #34
  • Oct 21, 2015 8:48am Oct 21, 2015 8:48am
Quoting MathTrader7
Tack s mycket fr att dela det! Statistically speaking, the net expected profit factor for this system (assuming 90% win ratio and 0.33:1 Reward to Risk ratio) in long run would be (0.33*0.9)/(1.0*0.1) = 2.97 which seems to be a promising approach. I may create an EA to back test the idea.
sounds great, be sure to share the EA in here if you do​
Bulls are stupid Animals!especially when Im short!
  • #35
  • Oct 21, 2015 8:52am Oct 21, 2015 8:52am
Quoting GoldTheHun
{quote} I think it would be better to backtest using the hourly data for daily trading. It would give much more error free simulation, and more accurate I think.
Hi! the problem I see with using hourly data is that, since the target is so small, there will be very small pips earned on hourly,
maybe only sometimes 5 pips? therefore I think daily is the minimum for any decent pips, but you are of course welcome to test​
Bulls are stupid Animals!especially when Im short!
  • #36
  • Oct 21, 2015 8:53am Oct 21, 2015 8:53am
Quoting roughtrader
{quote} sounds great, be sure to share the EA in here if you do​
You can find the EA in Post 29. :-)
Trading is the hardest way to make easy money...
  • #37
  • Oct 21, 2015 8:57am Oct 21, 2015 8:57am
Quoting Anusragger
Maybe use a daily candle of your own choosing. For example Frankfurt or London open.
that is a good idea, I am sure that will work, since we only play the wicks, I am sure you still would get the wick no matter what time you decide to start your candles, then we don't have to wait until midnight
Bulls are stupid Animals!especially when Im short!
  • #38
  • Oct 21, 2015 8:57am Oct 21, 2015 8:57am
Quoting roughtrader
{quote} Hi! the problem I see with using hourly data is that, since the target is so small, there will be very small pips earned on hourly, maybe only sometimes 5 pips? therefore I think daily is the minimum for any decent pips, but you are of course welcome to test​
What I meant was using the daily High, Low, Close for retracement calculations but backtesting it on 1 hourly data. That might give finer results.
  • #39
  • Oct 21, 2015 9:06am Oct 21, 2015 9:06am
Quoting mkuefne
Subscribed. This sounds very interesting. @roughtrader: many thanks for sharing your approach. Did you figure out if the method works better on certain pairs than on others? And on which kind of circumstances the losers mainly occurred?
Thanks, by sharing I hope this will turn out to something very great in the end, about the losers, what I have noticed while testing is that, very large momentum bars, especially when they also have broken support, could be the ones that is a possible candidate for a loss, because sometimes ( not always) the momentum is so strong that no wick on the next candle occurs, but at the same time, those long candles also gives the highest return if they work. maybe there should be a separate rule for Those particular candles, or we just simply will not trade them at all,​
Bulls are stupid Animals!especially when Im short!
  • #40
  • Oct 21, 2015 9:10am Oct 21, 2015 9:10am
Quoting PipMeUp
How are the gaps handled?
with gaps there is going to sometimes be a loss, nothing we can do about it,
but the good thing is that it might as well gap in our direction,
the chances of that I think is 50/50 I think that's just something we have to live with
Bulls are stupid Animals!especially when Im short!
1 trader viewing now