Please Help with coding a Trading System

I agree my friend...there are much better things to do in life than "Chart Gazing." Enjoy those other things to the fullest. Looking forward to any updates.

Yoges, Here is a small update. Looking at the EUR/USD from 12-29-2019 up until today & waiting on todays close.

Time frame again @ 15M using the MT4 provided Fibonacci. Total of Trades = 22. Winners - 17. Losers - 5. This is using the close ABOVE, or BELOW the 50% retracement level requirement. Nice results I would say. I really have no experience with EA's, but can see the need for one in order to shorten the workload of checking past history. I realize 22 days is not enough to determine if this method is viable to use. From what I have observed this method has much potential.

I might add some of the 22 winners some either start out a losers, or after making a strong move in the expected direction producing some gains then in the end became what would have been a loser using the 50% retracement level as the trigger to enter the trade. Most usually continue in the direction they would be expected to go, but not always. With this get the pips when they are available & get out is the approach needed at times.

One observation I noticed.....the further away PRICE CLOSES from the 50% level (either UP or DOWN) the more likely the trade will close a winner. NOT ALWAYS.

I will continue to monitor this & update you @ some point.

If this method wasn't so simple you would have people lining up to code an EA & others wanting to code an indicator. It always seems the more confusing & the more many indicators are needed the more that will latch on to the cause. Anyway keep looking & just maybe some kind soul will code up that EA for you.
____________________________________________________

Hey acetrader, glad to hear that you are still onto it. I am not sure if you are keeping records. I think you should and keep a proper backtest history. I keep mine in the format in the picture below.

I am not focusing on number of winners and losers because I have seen traders blow up accounts even with 75-80% accuracy/winloss rate. It is because they have not clearly defined all the rules or bad Money/Risk Managemant or too deluded to admit they are wrong when they are.

I am only focusing on total amount of cumulative gains or losses over a long period of time. But all pertaining to rules set followed religiously.

Meantime, i was checking out the results for monthly range on weekly candles of Gold from 2006 to 2019. Cumulative profits are good but number of losing years are more.

I will be going live with this system as soon as my yearly review period for all year go positive. Till then more work needed. I haven't checked out forex pairs because all i want is to know what i will be expecting in my results for the asset which i will trade and they are commodity futures.

Here is the result of monthly range on gold.

I didn't quite understand what you meant.

"I might add some of the 22 winners some either start out a losers, or after making a strong move in the expected direction producing some gains then in the end became what would have been a loser using the 50% retracement level as the trigger to enter the trade. "

I guess you are pointing out that there will be drawdown period of sometime. Yes there will be drawdown. Entering trades on low or the high of the day or the chart period almost impossible. Same thing like i said before, i am focusing only on cumulative gain or loss not on highs or lows. Also the price range between 38.2% and 62.8% is kept neutral biased to offer enough cushion to keep us in trades that will end up being winner.

Initially, my backtest was based on entry price of 50% but i noticed sometimes price doesn't retrace enough to the 50%. Introducing 38.2 and 62.8 levels gave more winning entries, results were drastically good. One drawback was bigger losses on individual trades. Also there are instances where even those entry levels are not hit during strong trends. So, still tweaking the system.

You are correct about the necessity for coding. It is all about reducing the workload.

I know the system is so simple, someone who hasn't even traded for a few days will be able to execute it after a short brefing session. Trading will be boring routined task which is opposite to what the industry potrays.

Hi Yoges,
I hope your still testing this method. It has got to be time consuming, but I feel it would be well worth your time. At this time I have this going on 5 pairs using the MT 4 provided Fibonacci tool. Seeing more wins than losses, but you have to get those wins quick because sometime the gains go into a loss before the end of the day. There are times when you enter a trade & it continues in the right direction during the whole trading time without ever turning into a loss.

From the forward test I'm conducting, your own to something with this idea. One thing I have observed is the cycles need to be considered. Some pairs are going down or up for 2-4 days then they reverse in the opposite direction. Take note of those cycles & help yourself avoid a losing trade.

Hi acetrader,

Of course on it as usual. Yeah it is very time consuming. To realize the gains TP levels should be entered for the positions as soon as positions are entered. Only stops monitored.

Yes i have felt about the cycles as well, sometimes gives an opposite entry to the trend makes loss and gives entry towards the trend and the trend continues.

Now, i am testing out variations on stops and TP to see if there is improvements to be made. Still no indicator used to filter trades... Maybe i will at some point in future.

Hi Yoges,acetrader

Let me know if you still need to code this up. Happy to help. Feel free to pm me .

Hi Mtbb,
Don't know if Yoges ever contacted you as this is his idea & thread. I found his thread & idea & thought it was a good method that is simple by design & a thinking outside of the box method. He was looking for an indicator which I think is another good idea. His method is sound, but as with anything in trading it is not 100% accurate & I know Yoges never said it was & never expected it to be. During my research into this method I noticed this tended to work for a number of days (CYLCES) in a row & then it would fail for a period (CYCLES), then the process would restart. On some pairs the winning days could be a many as 4 in a row & then it would fail. I have this setup on the EUR/USD on the 15M time frame & it is interesting to watch as each day unfolds.

This is Yoges idea & thread & all I am doing is offering what I see. A custom indicator would be good, but what could the indicator do in conjunction with the Fibonacci? I believe Yoges wanted an indicator for back testing the method, but I'm not sure. Back testing for me is less important than forward testing. Put the method on a chart & see what happen. All in demo of course. Put this method on any chart on the 15M & go back 3 months & you can clearly see pips to be made, but losses to be had too. The key for me is to watch for the trend cycle to change. It want continue in one direction but a certain amount of time & then change comes.

Yoges idea is solid. If an indicator can be coded to back test this method that would be great. Then, how to safely trade this needs to be explored. Put the Fibonacci on the EUR/USD on the 15M & go back to January 1, 2020 & see how many times it worked & failed. Or try any pair.

These are just my views. Not @ all trying to take over or derail Yoges thread. Even though I have my own method of trading Yoges method caught my attention because it is an outside of the box idea requiring no standard indicators such as RSI & MACD. The indicator you would code is a custom made indicator & much more helpful.

If Yoges has responded to you in a private message please disregard what I have to say. This is Yoges thread & I'm only trying to help with his great idea. That being said I am interested in your thoughs & view on this. Thank you & best of trades to you!

Hi acetrader,

I had a help and backtested the system. Works with Gold on long term. But is a loser with forex in longer term. Although everything did not happen as i planned. Not giving up on the system but need more research. Thank you anyway.

Hi Yoges,
That is good to hear. You idea is sound & I have seen some interesting results on the Dow & NASDAQ. I traded this method on the EUR/USD for a month with decent results. Still following this on the EUR/USD. Glad you got some help with this. These types of methods always get my attention. Thank you for sharing your idea & best of trades to you!

Traded as in forward testing right?

Well, any system that will give entries and exit will have some kind of performance. If wins are better than losers in long timeframe than thats good. Best trades to you as well.

Hi,

Here is a version of the EA created by Sangui. Test if you would like to.

Please do ask if anything is not clear.