How to optimize a Trading system So That win % or RR increase | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Help to optimize this excellent system 3 replies

How do I optimize my system based on my manual backtest? 1 reply

EA idea - to optimize manual trading 3 replies

Optimizer won't optimize anything... 2 replies

How to optimize an EA? 0 replies

Attachments: How to optimize a Trading system So That win % or RR increase
Exit Attachments

How to optimize a Trading system So That win % or RR increase

  • #41
  • Oct 8, 2020 11:15am Oct 8, 2020 11:15am
Quoting hasib
1. Macd cross above 0 lines. 2. RSI crosses above 50 lines. 3. Price crosses above 20 sam. 4. 3 ema crosses above 20 SMA.
Here I see a little problem. You have trend indicator and oscillator together. Oscillators are useless on trend. Trend indicators are useless on flat. So mixing them will give just more mess. It is better to use for example trend as entry, but oscillator as exit. It is just and example.
Observer effect
  • #42
  • Oct 8, 2020 11:54am Oct 8, 2020 11:54am
Quoting ryuryu
{quote} Here I see a little problem. You have trend indicator and oscillator together. Oscillators are useless on trend. Trend indicators are useless on flat. So mixing them will give just more mess. It is better to use for example trend as entry, but oscillator as exit. It is just and example.
If you develop algo with this system we may see good or bad results. if we find removing one or two indicators to make the system work we can do it. Let's not focus on opinion lets see the backtest result.
MA provides the market's current direction and strength.
  • #43
  • Oct 8, 2020 12:29pm Oct 8, 2020 12:29pm
The system I have share is profitable in manual trading. with 1 to 1.5 R trading you will find 56% win rate in H4 chart.
I have tested this Trading system in GBPUSD H4 chart. I took 9 consecutive loss, In manual trading- I will not take 9 consecutive lose since I will draw a simple s&R level or I will see BB is very narrow. But in Testing I have considered all trade.
I strongly believe this system will generate more than 56% win rate in manual trading.
MA provides the market's current direction and strength.
  • #44
  • Oct 8, 2020 2:00pm Oct 8, 2020 2:00pm
Quoting JamesSmithFX
This is my first post in ForexFactory but I will try to help. According to my understanding and experience, additional system improvement is only possible within and up to the limits of the available edge. In other words, there is no way to improve the final expectancy of a system beyond the hardcoded theoretical maximum determined by price history which by itself is impossible to reach due to unknown friction cost and the high potential for edge degradation in the future. Generally speaking, the market tends to modify the expectancy of most strategies...
I suggest you make that your first and last. Never heard so much drivel. The market does what it does and anyones strategy isn't going to change it. It may /will change over time but not because of someone's strategy. Automating something doesn't remove any of its inherent rules although it does expose it to your broker as they can tell what trades are automated.
  • #45
  • Oct 8, 2020 2:44pm Oct 8, 2020 2:44pm
Quote
{quote} If you develop algo with this system we may see good or bad results. if we find removing one or two indicators to make the system work we can do it. Let's not focus on opinion lets see the backtest result.

Please don't take it personally, but if you want to create a system you must understand what instruments are you using. In your strategy there are MACD and RSI. And this is like you have electronic speedometer and analog one in your car. They show you SAME results since they are using average. If MACD will cross zero then RSI will cross 50. Okay you can smooth one to be slower for example, but why not to use 2 RSI then? Try to read how the indicators works and what they calculate and showing.

Quote
{quote} The system I have share is profitable in manual trading.
Try this:
3 EMA (50, 100, 200) & MACD for entry. Stoch for exit.
--
price below 50 ma, and MA order is correct for short (50, 100, 200 up).
MACD cross 0 down
--
price above 50 ma, and MA order is correct for short (200, 100, 50 up).
MACD cross 0 up
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot2.png Size: 339 KB
Observer effect
  • #46
  • Edited 3:35pm Oct 8, 2020 3:23pm | Edited 3:35pm
Quoting hasib
Now Strategy and strategy entry rules. Buy
1. Macd cross above 0 lines.
2. RSI crosses above 50 lines.
3. Price crosses above 20 sam.
4. 3 ema crosses above 20 SMA.

You ' re entry rules are not precise. Perhaps I missed a post, if so, sorry. Do you wait for step 1 to happen, then 2, etc. ? Or no order matter ? The second all conditions are met you enter ? conditions met on candle close or on tick or how precisely ? Also, it would be helpful to know how much(number of trades and dates at least) you visually backtested this and your complete results of course. Cheers !
If I don't mention SL and/or TP = longer-term setup possibly in the works!
  • #47
  • Oct 8, 2020 5:09pm Oct 8, 2020 5:09pm
Hello everyone,
After reading all the suggestions I think we can simplify the system further.
Here is it. I will explain tomorrow.
It is a trend trading system so we have to avoid a choppy market we have to develop simple rules for the choppy market. if you have any suggestions let us know. I will share my idea tomorrow.

Here I use 50 ema, 20 sam, and 3 ema.
Thanks
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_1.png Size: 52 KB
MA provides the market's current direction and strength.
  • #48
  • Oct 8, 2020 7:01pm Oct 8, 2020 7:01pm
Quoting ryuryu
{quote} ~~ Oscillators are useless on trend. Trend indicators are useless on flat. So mixing them will give just more mess. ~~
In an uptrend, would you have a problem using an oscillator to show over sold areas to enter on pullbacks?
Good Trading
  • #49
  • Oct 9, 2020 12:09am Oct 9, 2020 12:09am
Quoting ryuryu
{quote} Try this: 3 EMA (50, 100, 200) & MACD for entry. Stoch for exit. -- price below 50 ma, and MA order is correct for short (50, 100, 200 up). MACD cross 0 down -- price above 50 ma, and MA order is correct for short (200, 100, 50 up). MACD cross 0 up {image}
Hello Ryukyu
Did you test this strategy in EA? What is the win rate, how much RR you Set up for this strategy?
MA provides the market's current direction and strength.
  • #50
  • Oct 9, 2020 12:19am Oct 9, 2020 12:19am
Quoting Spikehunter
{quote} I suggest you make that your first and last. Never heard so much drivel. The market does what it does and anyones strategy isn't going to change it. It may /will change over time but not because of someone's strategy. Automating something doesn't remove any of its inherent rules although it does expose it to your broker as they can tell what trades are automated.
Hello spikehunter,
I was thinking of adding a few more indicators, After all the suggestions I am afraid adding indicators will make the system inefficient.
I have to go for a simple strategy. I Need your valuable suggestions.
I am thinking about this Entry rules see in Picture.
Entry Rules Buy
1. Price above 200 ema
2. If 3 EMA Cross Standard divisions 1 we will buy
I don't test this strategy before. I think this will work as a trend trading strategy. what do you think? If you think this entry rule is not as good as 20 sma and 3ema cross with the directions fo 50 ema, we will strict with the previous one. Thanks
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_2.png Size: 66 KB
MA provides the market's current direction and strength.
  • #51
  • Oct 9, 2020 12:48am Oct 9, 2020 12:48am
Example 1- Price is above 200 ema and 3 ema cross upper Standard deviations 1. after candle close we will buy.
Example 2 and 6 are a little different - Price is above 200 ema and 3 ema cross above Lower standard deviations we will buy.
- Entry rules are fixed so now we can discuss exit rules.
- We can determine when Not to trade
- with the candle count we can exit your trade with a break-even or Little loss before large-stop is hit.

Can anyone code this strategy for me? It would be a great help.
Thanks
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_3.png Size: 67 KB
MA provides the market's current direction and strength.
  • #52
  • Oct 9, 2020 12:57am Oct 9, 2020 12:57am
Quoting Spikehunter
{quote} I suggest you make that your first and last. Never heard so much drivel. The market does what it does and anyones strategy isn't going to change it. It may /will change over time but not because of someone's strategy. Automating something doesn't remove any of its inherent rules although it does expose it to your broker as they can tell what trades are automated.
We all have different experience and knowledge levels. You know something, I know something, everyone knows something. Generally we all know exactly what we know.
But maybe it's even more important to know what we don't know!

I don't claim to be an expert in terms of Forex and speculative trading in general. I'm still researching it.
But I have a lot of experience with software development, industrial automation and the solving of some really complicated problems. Back in the day I was one of the best at my job, and some of my inventions and solutions are still in use today. Now I'm semi retired and I no longer work in the industry, but I'm always looking for new challenges to keep my mind sharp.
Maybe, just maybe I can figure out this whole trading thing!? Or at least I can try.
  • #53
  • Oct 9, 2020 1:17am Oct 9, 2020 1:17am
Quoting JamesSmithFX
{quote} Maybe, just maybe I can figure out this whole trading thing!? Or at least I can try.
Hi, Do you know MQL4 programming?
MA provides the market's current direction and strength.
  • #54
  • Edited 1:50am Oct 9, 2020 1:32am | Edited 1:50am
Quoting hasib
{quote} Hi, Do you know MQL4 programming?
I'm a bit old school. My primary focus is on C and C++ languages. This is what we used in the industry for the most important things. Today they have all kinds of "specialized" languages mainly because most programmers today are bit too lazy to learn how to code with a proper language like C++.

Now I'm looking at all these different trading platforms and they all seem pretty basic. From what I see MQL4 and MQL5 seem to be based on C++ so this is where I plan to concentrate my efforts. I'm confident that I can pick any "modern" language and I can learn it in a few days or a week max.
  • #55
  • Oct 9, 2020 1:56am Oct 9, 2020 1:56am
Quoting JamesSmithFX
{quote} I'm a bit old school. My primary focus is on C and C++ languages. This is what we used in the industry for the most important things. Today they have all kinds of "specialized" languages mainly because most programmers today are bit too lazy to learn how to code with a proper language like C++. Now I'm looking at all these different trading platforms and they all seem pretty basic. From what I see MQL4 and MQL5 seem to be based on C++ so this is where I plan to concentrate my efforts. I'm confident that I can pick any "modern" language and...
Could you code the above bb and 200ema strategy?
MA provides the market's current direction and strength.
  • #56
  • Oct 9, 2020 2:38am Oct 9, 2020 2:38am
Quoting hasib
{quote} Could you code the above bb and 200ema strategy?
Of course I can. So far this looks like a very simple trend following strategy.
But first we have to determine all of the rules. This is the thing about automation. You have to start with well defined rules and parameters.
The alternative is to automate only one side of the strategy depending on the available definitions or external factors.
For example manual open + automated close or automated open + manual close.

In your case you only have clear definition for the open. This makes it impossible to back test properly and to determine the actual trade expectancy of the system.
  • #57
  • Oct 9, 2020 4:05am Oct 9, 2020 4:05am
Quoting JamesSmithFX
{quote} Of course I can. So far this looks like a very simple trend following strategy. But first we have to determine all of the rules. This is the thing about automation. You have to start with well defined rules and parameters. The alternative is to automate only one side of the strategy depending on the available definitions or external factors. For example manual open + automated close or automated open + manual close. In your case you only have clear definition for the open. This makes it impossible to back test properly and to determine...
Buy
Entry rules- 3 ema cross Upper Standard deviation 1.
Stop rules- Stop Below Lower standard deviations2.
Take Profit rules- 1.5 times SL. ( Keep an option in Ea so that we can reduce or increase RR Multiplier)
Sell is just the opposite.

Once we see little light on this strategy we will add filters like 200 ema or ATR base TP and SL. What you guys think. Can anyone code it?
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_1.png Size: 58 KB
MA provides the market's current direction and strength.
  • #58
  • Oct 9, 2020 4:27am Oct 9, 2020 4:27am
we can use the first one that will be easier

entry rules- 20 sma cross above 50 ema.
whenever 3 ema cross above 20 Sma
we will buy as long as 20 sma above 50 ema.

Stop rules 2.5 ATR
Take profit 2R.

Please keep options to change EMA and SMA value
Stop Loss ATR Multiplier increase decrease options.
and Take profit RR change options.

So that we can adapt to the market by backtesting the ea certain pair.
Coding this system will be great.

I think this ea will be much simple and better. BB can give us a big stop
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_2.png Size: 44 KB
MA provides the market's current direction and strength.
  • #59
  • Oct 9, 2020 4:33am Oct 9, 2020 4:33am
Quoting aud
{quote} In an uptrend, would you have a problem using an oscillator to show over sold areas to enter on pullbacks?
You can have 6month uptrend and you will have 100500 oversold signals every day higher. Or it will be only one and it will not be a pullback. It will reverse forever. And it will be for next year. Oversold -> reverse. Oscillators useless on trends. Trend indis useless on flat. But you can open position using trend indicator and exit on oscillator. I just gave example how better use provided instruments. Because hasib used 2 same indicators on the chartm that is useless and just add more mess.

Quoting hasib
{quote} Hello Ryukyu Did you test this strategy in EA? What is the win rate, how much RR you Set up for this strategy?
I think I have tested them all. They all work good as entry. As I said before RR is magic. I don't use magic. And as I said before strategy must have entry, exit and sl calculated. This one (mine example) has Entry and TP but doesn't have sl. So this can't be tested in strategy tester, because it will be great on past week and bad on next. I just wanted to show you that this is useless to have 2 oscillators on the chart. You can use 1 trend and 1 osc. That's enough. You will have same Entries with same quality.

Your last strategy will not work it has magic TP and magic SL. In fact it has only good entry. It will gain loss. No need to code it. Don't waste your time. You have too big SL and it will wipe all the profit and gain big loss. Imagine you have 100p tp and 2000p sl. If sl will be hit 2-3 times in a row there is no chance to get balance back.

We need to think where to set TP and how calculate it. And where to set SL and how to calculate it. There must be hardcoded rules. Like the rules for entries. Then we can code it.

I have found my old BB, Stoch, CCI, ATR, candle color, size, shadow size and volume EA. I'll compile it soon and attach here. The strategy seems to be very close to yours.
Observer effect
  • #60
  • Oct 9, 2020 4:50am Oct 9, 2020 4:50am
I am not a programmer but I can see Just 3 to 4 ema cross is working very well in the H4 chart.
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_3.png Size: 57 KB
MA provides the market's current direction and strength.
0 traders viewing now