ok, we will go through in this example, did you get driving license to be a F1 driver?
I don’t have a driving license.
and still you want to be a competition for other F1 drivers
I was just comparing it to what you said about not recommending M1 to beginners
So
I was looking for bad trades using the strategy, and it is actually not as easy as I thought.
I tried looking for them manually, but then tried using the previous losing trades from my backtests.
Most of them aren’t trades anymore
Interesting
I trade for 10 hours a day but I don’t want to watch the chart for 10 hours waiting for a setup. So I made an EA that sends alerts when a possible setup is about to form.
I say possible since the EA only check is price is above or below the 200 EMA and not if a set has actually formed. I would make it only alert when a setup is formed but working with Fractals in MQL5 is suffering because the fractals don’t output the price level which they formed, they just output some random big number.
Over the course of 2023, there were 3087 alerts for buy setups and 1607 alerts for sell setups.
Maybe I’ll be able to make the whole strategy into and EA
someday
In this screenshot of USD/JPY on the 26th of April, 2024, you can see that there were two setups and the candles, where the 2 vertical lines are, are the places I would have taken the trades. The reason I wouldn’t have taken the trades is because the main line of the Stochastic Oscillator is below the signal line on M6. This is the final confirmation I need before I enter a trade.
This is actually a distinctive characteristic of this strategy because it tends to miss really good trades because of the somewhat slow settings of the Stochastic Oscillator which I have made that way to act as a filter between possible trend reversals and flash movements in the market that don’t lead to trend reversals.
The good thing is that losing trades that were taken on backtests using the old settings won’t be taken with the new settings, but winning trades that had massive profits of over $100 also won’t be taken with the new settings.
If the Stochastic Oscillator on M6 had confirmed the trade and I took it according to the strategy with a lot size of 1, I would have made a potential $659.75
On the bright side, at least taking no is better than taking losing trades 
After 4 weeks of backtesting, optimization and guess work, I have developed a set of rules for my strategy that seems to produce the best results (for now)
INDICATORS
- 9 SMA
- 28 LWMA (HLC/3 “Typical Price”)
- 200 EMA
- Fractals
- Stochastic Oscillator
- 16 %K
- 3 %D
- 8 Slowing
TIMEFRAMES
- M1 (Entry)
- M2 (Confiramtion)
- M6 (Confiramtion)
- Daily (Analysis)
RULES
- Trade from 7:00 - 17:00 UTC +1
- Trade from Tuesday - Thursday
- No levels are allowed to be set outside the levels designated by the strategy (even if the SL is is greatly larger that the TP)
- Levels that were broken as a result of candles that formed during news aren’t counted
PAIRS (BACKTESTED)
- EURUSD
- GBPUSD
- USDJPY
- GBPJPY
ANALYSIS
- If Daily candle opens below all 3 moving averages, take only SELL trades
- If Daily candle opens above all 3 moving averages, take only BUY trades
BUY SIGNAL
- Go to the M1 timeframe
- Wait for price to drop below the 200 EMA
- Wait for an upward facing fractal to form
- Wait for a downward facing fractal to form on a candle whose bottom part of its body is below the top part of the body of the candle which has the upward facing fractal
- Mark the top of the body of the candle with the upward facing fractal (PEAK) and mark the bottom of the body of the candle with the downward facing fractal (BASE)
- If an upward facing forms on a candle and the top part of the body of that candle is below the PEAK but above the BASE, remove the old PEAK level and mark the top part of the body of this candle as the new PEAK
- If a downward facing fractal forms on a candle and the bottom of the body of that candle is below the BASE
- If the Stocahstic Osciallator is oversold, remove the old BASE level and mark the bottom of the body of that candle as the new BASE
- Wait for a candle to pass through the PEAK level. If a candle opened a above the peak level, don’t count it
- Wait for a candle whose opening time coincides with the opening time of a candle on M6, E.g. 10:00, 11:36, 9:54, not 11:11
- Go to the M2 timeframe
- If the main line of the Stochastic Oscillator is above the signal line, go to the M6 timeframe
- If the main line of the Stochastic Oscialltor is above the signal line, go back to the M1 timeframe
- Wait for the candle to close
- Open two BUY trades and set their SL to the BASE level
- Set the TP of one of the as the difference between the entry price and the BASE level
- Set the TP of the other at any recent crossover of the 9 SMA and 28 LWMA where the 9 SMA crossed the 28 LWMA from above it
SELL SIGNAL
- Go to the M1 timeframe
- Wait for price to rise above the 200 EMA
- Wait for an downward facing fractal to form
- Wait for an upward facing fractal to form on a candle whose top part of its body is above the bottom part of the body of the candle which has the downward facing fractal
- Mark the bottom of the body of the candle with the downward facing fractal (BASE) and mark the top of the body of the candle with the upward facing fractal (PEAK)
- If a downward facing forms on a candle and the bottom part of the body of that candle is above the BASE but BELOW the PEAK, remove the old BASE level and mark the bottom part of the body of this candle as the new BASE
- If an upward facing fractal forms on a candle and the top of the body of that candle is above the PEAK
- If the Stocahstic Osciallator is overbought, remove the old PEAK level and mark the top of the body of that candle as the new PEAK
- Wait for a candle to pass through the BASE level. If a candle opened a above the peak level, don’t count it
- Wait for a candle whose opening time coincides with the opening time of a candle on M6, E.g. 7:00, 15:36, 12:54, not 17:13
- Go to the M2 timeframe
- If the main line of the Stochastic Oscillator is below the signal line, go to the M6 timeframe
- If the main line of the Stochastic Oscialltor is below the signal line, go back to the M1 timeframe
- Wait for the candle to close
- Open two SELL trades and set their SL to the PEAK level
- Set the TP of one of the as the difference between the PEAK level and the entry price
- Set the TP of the other at any recent crossover of the 9 SMA and 28 LWMA where the 9 SMA crossed the 28 LWMA from below it
MDG.txt (4.1 KB)
If anybody has any questions or suggestions, let me know and if you end up testing this strategy, also let me know
I don’t know if you might have mentioned this somewhere in between the thread but i would like to ask what your trading style is, scalping, intradaytrading, daytrading or swingtrading?
Asking this in proposition to your M1 timeframe.
A mix of intraday trading to scalping.
Sometimes I can have 1 trade a day but the most I’ve ever had is 16 trades in a day on my backtest and 14 of them were losses.
For the first time, I have decided to venture out of the forex market with my backtesting and I went into
CRYPTO.
BTC to be specific, and this is the most shocking backtest yet.
For this backtest, I ran it from March 14th, 2024, to April 11th, 2024, trading for a total of 14 days within that time frame.
Surprisingly, this is the first backtest to have a win rate below 50% because, out of 103 trades, 53 were losses.
I also used 10 lots instead of 1 for this test because it gave a more reasonable profit numbers on a $250,000 account.
At the end of the first day, the account was down $15,531, bringing the balance down to $238,468.90.
At the end of the third day and after 12 trades, the account was up $6,121.30, bring the balance up to $256,121.30.
In the above picture, you can see the balance after every trade. To me, this kind of looks like an uptrend with somewhat hefty pullbacks, but the main thing is that it is going up, which shows that even though I’m taking more losing trades than winning ones, the winning ones are somehow managing to cover the losses.
If we compare this chart to the last GBP/USD backtest:
We can observe some similarities between the two and notice that since the market was somehow moving sideways like it did during the GBP/USD backtest, there are also hefty pullbacks with some of them taking the balance from $1,100 to below the starting balance in 8 trades.
This shows that it may not mean that trading crypto with this strategy won’t work, but it may mean that the strategy just gets-by during times of choppiness in the market, which cold place more importance on the day you start trading.
This is the balance of the USD/JPY backtest after every trade. In the far right, we can see that before the big move upwards, there is a big move downwards.
During this big move, the account lost a total of $743.60 ($43 over the starting balance) within a day. 11 trades (the longest losing streak I’ve ever seen), but immediately after, I saw the longest winstreak I’ve ever seen.
So far, the strategy is taking everything thrown at it, even if it is just barely.
1 Like
This is the longest trade of any backtest, which lasted exactly 7 hours from 09:18 - 16:18.
I’ve also noticed that after more than 500 trades with this strategy, not a single one has crossed over to the next day, which is good because I can avoid swaps and inflated spreads, which we can all agree we hate .
1 Like
I finished making an EA that tells me price crosses above or below the 200 EMA so I can look at the chart only when a potential setup is about to form and not 10 hours
The alerts get sent by notification.
The only downside is I have to leave my PC on or get a VPS since unlike alerts, MT5 has to be connected to the internet for them to work.
So, does anyone know how a VPS works.
Does MT5 just leave your account open on a server somewhere so the EA can do its thing?