MACD Simple Cross System

Hi DrDave, thank you for sharing your results with us. I have done some testing and let's discuss it here.

First regarding this post, I have not used the FXCorrelator indicator before, so this is still confusing to me. I will try to get a grasp of it and let you know, because so far it just confused me.


I did some checking, and I don't think the issue is with the EA or the buy/sell condition, I think it's with the system itself, and here's why:
The system does not follow the trend, but it buys when the MACD and Signal Line are both below zero and MACD is crossing up, this means that the price has been going down for a while, hence the MACD and Signal Line are below zero, and since there's a cross up, we're expecting (or hoping) that the trend is about to reverse, so we buy. If it goes down further and crosses up again, we buy more because we're expecting it to go up now, and we keep buying even if the price is going down because we're saying when it reverses and goes up, we will get everything back. Sometimes, trends last longer than we hope for, so we end up losing because it goes back up, but not high enough. When does that happen? Nobody knows, it just happens, because nobody knows where the market is going. So, without a strategy for us to throw the towel and exit with a loss earlier, we can see some really huge losses. Now, here are some photos with explanations to prove my point.




Above, is the EURUSD H1 chart, as you can see here, on Feb 04, 2019, a sell order closed and a new buy order opened (from left to right, first blue line). Now, since a sell order just closed, it means the MACD and Signal Line went below zero and the MACD was below the Signal line and now they have crossed, which means the price was moving down, and now we bought because we received a signal that told us the trend is about to reverse its course, but it was just a retracement. The price went down further and now MACD is below Signal Line. When the next retracement happened, the lines crossed again and we received another buy signal, but the price went down further, and so on and so forth until we had 9 open buy orders. Now, the price did go up eventually, but it wasn't high enough, so only the 9th order won, but the first 8 were losses. Now, why did the EA kept buying? Because it kept getting the buy signal from the MACD which is: MACD and Signal Line below zero and histogram has turned green (or MACD crossed up). See the MACD photo below with the lines corresponding to the orders.


Now, those are the instances where the system collapses and loses everything. And I think if you test it again and attach the MACD indicator to the chart, you will see that it buys on buy signals, and sells on sell signals, and you'll see that the EA is doing its job, but market is not working with us here.

Now, I'm doing more tests and will share my results with you shortly.

Same thing happened here in both directions, it started on Mar 01, 2019:




That's why I suggest we put a limit to the number of open orders in a specific direction, for example if it hasn't turned around in four buy signals, it's probably going to go lower and lower, so it's better to have four losses in that direction than to have eight losses and one win.

Some back-testing results:

If you had used this EA to trade the EURUSD H1 chart from January 1, 2019 to July 26, 2019, you would've ended up with a total loss of 25.15% on a $1,000 account. See image below:


// ---------------------------------------------------------------------------------------------------------------------

If you had used this EA to trade the EURUSD H4 chart from January 1, 2019 to July 26, 2019, you would've ended up with a win of 15.97% on a $1,000 account. See image below:


// ---------------------------------------------------------------------------------------------------------------------

If you had used this EA to trade the EURUSD H1 chart from January 1, 2017 to December 31, 2017, you would've ended up with a win of 42.41% on a $1,000 account. See image below:


// ---------------------------------------------------------------------------------------------------------------------

If you had used this EA to trade the EURUSD H4 chart from January 1, 2017 to December 31, 2017, you would've still ended up with a win of 8.05% on a $1,000 account, but it would've been a real roller coaster. See image below:


// ---------------------------------------------------------------------------------------------------------------------

So, just like all other trading systems, it works when it works, but when it fails, we have no money management plan to protect us, that's why it takes everything back. Now, I think I have an idea for a risk management plan that I'm going to test during this week and will share the results with you.

First, edit the inputs so that all of the currencies are set to false except the two for the pair of your chart.

If you are looking at EURUSD, the lines show the RELATIVE strength of the two currencies. So when EUR line is above the USD line, you see price going up. Because it is RELATIVE, you can see price going up under several configurations of the lines as long as the EUR is above the USD: 1.) EUR going up, USD going down; 2.) EUR going up, USD flat; 3.) EUR going down, USD going down; 4.) EUR flat, USD going down.

The very best runs up are when the EUR and USD lines are very far apart and the EUR line has steep up slope and USD line has steep down slope. Where the two lines cross, they have equal strength so the price tends to run flat. When the two lines are very close together, they have strength very nearly the same, so price will tend to run in a narrow range.

Yes, we all agreed on that earlier. Money management is the main issue because of "anticipating" the trend will change from down to up and buying at every up crossing while MACD < 0 rather than waiting for confirmation of a trend change. posts #26, #36, and #38, and #43 address this.

According to what I see in the capture below, I am not convinced the EA is working according to the original rules. I left all of the default settings of the MACD when I ran the Strategy Tester. The yellow vertical is where the MACD crossed down, so signals SELL on the open of the next bar. The next bar open is 2019.05.01 13:00 1.12411. The EA report shows first BUY 2019.05.01 00:00 1.12207; that is the blue vertical; the red horizontal is the price. As you can see, a BUY is not valid here; we are in selling conditions. Even more troublesome is that there is not a MACD crossing any time near that time; crossings are about 6 bars away on either side of that time!
You can see in the report the BUYs 1,3,5 are when the MACD is ABOVE 0, not below. No SELLs were made during the time the MACD was ABOVE 0. Clearly, there is a problem.

I will play with that and let you know how it goes.

I am really confused here, as I went back and did more testing and it seems that the EA is working exactly like it should be. I will show you below.


As you can see, those are the default settings that I'm using:



As you can see the name of the EA here, the pair I'm using to test, the chart time frame, and the period. I also have Visual Mode turned on. This forces the EA to work in visual mode so I can monitor tick by tick what is happening and I can make it go fast or slow as I wish, so this is how I know it's working fine. As you can see, the white line indicates the start of the test.



As you can see below, so far, there's no order because no cross has happened. Here, I pointed out an instance where it looks like it's a cross, but actually it didn't cross down.



Below, as you can see, the first cross happened exactly at 2019.05.01 21:00, no order is triggered so far because the candle has not closed yet.



Once the candle has closed, an order was triggered at 2019.05.01 22:00 as you can see in the image below.



When the reverse signal came (the buy signal) as you can see the EA closed the sell order and opened a buy order.



This is another screen shot during the Visual Mode of the test. If you don't have that turned on, I suggest you do, it's very helpful when you're testing EAs.



I checked the code, it seems there's nothing wrong with it. I checked the results order by order, it seems there's nothing wrong with it. But, just in case, I'm uploading the EA again here just in case it had been corrupted when I uploaded it. Here it is:


Also, from the image you shared, I have noticed that you have closed buy orders, but not sell orders were opened, plus the lot size started with 0.2 and moved to 0.1, which is weird. So, try out this one again. I really don't know what's wrong because it works fine here. I hope that other members share their input on this so hopefully we can find a solution.

Best.

Bfxs, I tried running the EA in the strategy tester today using the settings you posted and it seems to be running OK.

I was watching EURUSD H1 running in the strategy tester, seeing many buys taken during sideways moves that ended as losers because price continued down on another down leg, so I thought about changing the direction criterion. When signal x 0 down, it is in SELL window until signal X 0 up. Make the first sell at this crossing. Make add-on sells every time MACD X signal down.

Initially, I considered keeping the trades open until signal X 0 up, closing the SELL window, so all the sells closed out and a BUY window now activated. Although that can work, it did give back quite a few pips. Therefore, I tried running a trailing stop a few pips below the most recent entry bar. I also included the high of the entry bar before this one and used the larger of those two values. Then closed the buys on whichever of the following conditions occurred first: the stop hit; the signal X 0 up. This trailing stop will also be beneficial for a trade window that has only the initial trade opened then price reverses; the trade will get closed with only a small loss based on the trailing stop.

Same idea for selling.

I did not evaluate periods where price ran nearly flat for a lot of bars, because I think you might be able to generate another EA with these new rules, so the EA can be used to see what happens in those flat periods.

Attached is the H1 and H4. White verticals are where signal X 0 to demarcate the buy/sell windows. Red verticals are MACD X signal down when MACD and signal below 0. Green verticals are MACD X signal up when MACD and signal above 0. Red horizontal was last stop for sells. Green horizontal last stop for buys .

Looking at H4, it appears that the white verticals placed on H1 correspond approximately to the H4 MACD X signal. So clearly, this approach is making entries on H1 in the same direction as a price leg on H4.

DrDave, I'm happy to hear that the EA has done its job.

Now, I've read your post, and I have to say that you have some really interesting ideas here, and I think they can be translated into an EA than can be profitable. But in order to do that, I have to make sure that I do clearly understand what you're saying. So, I went ahead and created some detailed photos to see if I'm understanding you clearly, and I also tweaked the stop loss thing (I will explain it with the explanation of the photos). I will post the photos here and explain on them. If they are along the same line as your post, I can go ahead and use the existing EA to copy some code and create a new one according to our new rules. I have only created a sell scenario, but it will be the same for a buy scenario except the signals are reversed.

1. On 2019.04.23 at 15:00, the Signal Line crossed down below zero (marked by the thick vertical white line) which opened the sell window for us, so we opened a sell order at the close of that bar which was at 1.12174 (marked by the highest dotted white horizontal line). Our stop loss is two pips above the highest high of the crossing bar and the six bars before it, so in this case it's 1.12629 (dotted yellow horizontal line).


2. On the next day at 09:00, the MACD crossed down below the Signal Line (first dotted red line from the left), we're still in the sell window since they're both below zero, so we opened an add-on sell at the close of the crossing bar which was at 1.12075. Stop loss now has become two pips above the highest high of the new crossing bar and the six bars before it which is now 1.12258 for both open orders (dotted yellow horizontal line).


3. Then at 16:00, another cross happened, another sell order was opened at 1.11923. Stop loss, using the same formula has become now 1.12250 for all three open sell orders (dotted yellow horizontal line).


4. Another cross happened at 12:00 on the next day so another sell order was triggered, and stop loss became 1.11640 for all open orders. Now, we exit the trades if the Signal Line crosses up above zero, or if our stop loss was hit before reaching that point. In this case, the cross happened at 22:00 on the 26th, but, the stop loss was hit at 17:00, so, we were stopped out before reaching the buy window (stopped out at the dotted vertical yellow line, buy window opened at the vertical green line). When we reached the buy window at 22:00, we opened a new buy order, following the same criteria we did here, and so on.


This means, on 2019.04.26 at 17:00 we closed four sell orders with the following results:
Order 1: 1.12174 ---> 1.11640 (+53.4 pips).
Order 2: 1.12075 ---> 1.11640 (+43.5 pips).
Order 3: 1.11923 ---> 1.11640 (+28.3 pips).
Order 4: 1.11322 ---> 1.11640 (-31.8 pips).
Total: +93.4 pips.

Am I getting you right? Should I create the new EA to do just like this?

Yes.

But also, if you have time, the following options
---make an extern to choose either MACD or Signal cross 0 signal to make an entry on the open of next bar that marks a direction window. In a good trend, I expect that the earlier entry via MACD would harvest more pips. However, it could also result in greater losses during the low volatility periods.
---make an option to enable the Signal (or MACD) x 0 signal to allow or deny an initial order on the open of next candle (You can see why in the m15 and H1 below)

This is EURUSD m15 and H1.
The white verticals mark Signal X 0.
Notice after the first BUY vertical on the left, there are no add-on buys before the SELL vertical gives reversal signal. This is almost guaranteed to be a losing trade under these conditions.
Likewise for the SELL window between the 2nd and 3rd verticals; no add-on sells, and the initial sell a loser.
And once again between the 4th and 5th verticals.
Also note that after the BUY window that begins with the 3rd white vertical there is one add-on buy marked by the green vertical. It is clear that even though it initially went to a small unrealized profit, the down turn hits the stop for a loss (or if none used, definitely is a loser at the white vertical exit).

So as I expected, this strategy will lose during low volatility periods and win during good trends. So this raises four questions: 1. Over the long term, are the losses sufficiently small compared to the gains from the wins that it is not important to filter out the low volatility?; 2. Over the short term, are the losses small enough that we do not need to be concerned about the amount of drawdown incurred while the trades are open?; 3. Over the short term, are the losses small enough that we do not need to be concerned about the account balance?; 4. Is there any significant difference in the net profits if we do not enter an initial trade at the signal of a new window, and only use add-on trades?

So you can see that at least in this example, it would have been better during the low volatility period to not make an initial trade at the white verticals. However, looking at the SELL window with the red verticals, how much profit would be lost if the initial sell were not made? This is where the EA would help to more easily determine which option would be better to implement.

The H1 chart shows very clearly how low the volatility went by the tiny histogram bars. Perhaps a sufficient filter to use when an add-on trade signal forms would be to require some minimum price change from the last entry to be able to allow another add-on. You can see that in a good trend, there tends to be a significant change in price between trades, whereas during the low volatility periods, the change is quite small. This, in effect, filters out whipsaws of a specific range.

Hi DrDave, I will start working on it these days and hopefully I can have it ready by this weekend.


Both options seem doable to me and I will put them in my planning to be included in the EA, so, no problem here.


Regarding your four questions, let me address something, and provide a suggestion that can help me answer some of them. If we already decided that the entry price is known (close of crossing bar or open of next bar), and we already know that our stop loss is going to be two pips above the highest high or two pips below the lowest low of the last seven bars (cross bar included, and the number could change but I found that seven bars is not too low like five or too high like 10, but I'm open to suggestions. We can also use an extern for that), now, since we know what the entry price is and the stop loss, I can insert a code that I found online that can calculate the lot size based on risk % that we can set by an extern. This means, we are going to risk the same percentage on each trade. So, to answer your questions in order:

This is what the EA is going to tell us, we can create it and let it run a test on data from 2010 to date and have an answer.

While the trades are open, let's say you risk 2% per trade, I think there are very rare cases when you have three open orders and you're still risking 2% on each one of them, mostly, if it's low volatility, then there will be a small number of open orders like you mentioned it could be only one or two at a time, so, a small percentage is at risk. If it's high volatility and the price moves against you, you're out of the trades earlier. If it's high volatility and the price moves in your directions, you have your stop loss modified every time a new order is opened so you don't need to worry about DD while the trades are open because most of the trades will turn into winners.

Again, if we use a lot size calculator based on risk, the more you lose, the smaller your lot size will be in the next order, so it can help manage your capital during looong periods of time of bad luck. Again, the EA will tell us once it's finished.

The extern that I'm going to put will help us decide. And, if there's a native indicator that we can use (because I'm still new when it comes to coding so I can only work with native indicators so far), I can incorporate that to let the EA decide whether or not to take an initial buy order when the buy window opens and the MFI is above 50, for example, or the RSI, or whatever, as long as it's native. For now, we're going to leave it as is, and we can tweak it later on.

Now, let me know if the lot size calculator is something we should include so I know what to do. I will start working on it these days and again, hopefully I can have it ready by the weekend. I expect it will be ready by then because I have current EA to steal lines from haha and will only write new lines for the new suggestions.

Great!
As far as the decision about entering a trade or not on the Signal (or MACD) X 0, you can make an InitialTrade a boolean extern. Then you will still look for that cross 0 signal to determine if it signals a BUY or SELL window, and if InitialTrade is true, the function to enter a trade is called but if false, the function is ignored.

By the way, I used m15 and H1 as examples just so I could more easily find a low volatility streak. I am sure Lukass' still prefers H4 so he won't get many alerts during a day.

Go for an average of 10 pips per day and you will have positive results every month!

(OOPS! Accidently hit the CLOSE button when trying to modify a trade; just took a tiny loss, no worries. LOL)

(Just hit profit of 10 pips on GBPJPY. Oh, another 10 pips. OOPS! Already exceeded my 10 pips for today! What will I do if I hit my next target with 2 orders at 20 pips profit each?? No problem, because I did not trade this week, so that will make 60 pips, 10 pips over my weekly goal.) LOL

The spread is running about 1 pip. Price came down within 0.1 pip (plus spread) of my target then reversed. LOL
OK,no problem; expect to see another rally then drop again.

Heh-heh... up, then DOWN she went--and another 40 pips in the bag!

C'mon folks! While waiting for the next round of updates, JUMP ON THE GRAVY TRAIN! Just today on GBPJPY I did all sells,
lots_____pips
0.01____11.2
0.02_____1.1
0.01____10.2
0.01____00.0 accidentally cancelled it LOL
0.02____21.5
0.01____10.2

That's 53 pips harvested, and I shoot for 10 per day! (And doubling the pips with the 0.02 lots is the equivalent of 74 pips of 0.01 lot size).

My last entry was the H4 coming into play; it should go a LOT more (100 pips is reasonable) than the 10 pips that I targeted.

What is going on, gang?

I have been on the road without much time to reply, but I have been able to somewhat follow the thread.

Regarding the EA, it works exactly as it should and as I intended the original strategy. Right now, I am trading the demo account on 6 major pairs and will keep you updated as soon as I close some trades. I want to play with the settings of the MACD - I was thinking about using H1 with 4x MACD values (basically to get H4 signals on H1 chart), and also got few more ideas I want to try.

I will definitely let you know what my findings will be and if they are interesting, we can further continue the discussion. Until then, I would say let's work on Dr.Dave's idea.

Dave, can you send us your trading template that you are using right now? Possibly with some entries marked as well. That would be a good start for me to catch up. Post #67 has the rules, right? I am asking because I can see the RSI on your chart as well.

Regarding the entry of the sell or buy zones (crossing level 50), and then exiting, what do you think about simply just following the previous bar value of the indicator? That means, if the indicator is going down and the values go down, we are in sell zone, and as it changes and it raises, we are in buy zone again. If we are using longer periods, we should not have that many ups/downs in the overall sentiment.

Have a great weekend everyone and happy trading!

PS: Great results, Dave!!! keep it up

Hey folks, are you trading AUDJPY today? NO???? Wha-a-a-a?? C'mon! It's another gravy train! I already sold today and banked 15 pips, exceeding my 10 pip goal. 41 minutes, and I am free for the rest of the day. Well, actually, I was free right after I put on the trade, because I just walked away anyway, and when I returned--KA-CHING!

(Actually, I could have skipped trading today altogether because I already met my weekly goal yesterday.)

U Can Do IT!

C'mon already! It's like stealing candy from a baby!

Still on AUDJPY m5. Two more little 5 pip moves in price with 2 lots = 20 pips total profit. KA-CHING! KA-CHING!

U Can Do IT!

Attached Image (click to enlarge)

I attached a ZIP file with a different template and some indicators. I left on the MACD for anyone that wants to make comparisons with the other indicators. I also added a Hull moving average. Maybe the one I use is a bit complicated for most people. You can just use a standard Hull moving average 12, median price instead.

I was looking at the stochastics some more and found that lower settings are much better.
(8,3,3)
(32,8,1) Turn off the %k line.
RSI(13)
I was comparing the RSI crossing 50 to the stoch32 crossing 50 to see if the RSI can be eliminated.
when looking at the H1 chart, this is essentially using the H4 stoch(8,3,3) to indicate direction on H1. stoch32 or rsi13 > 50 is buying; < 50 is selling. They have very good tracking, but because I saw a few times of more crossings on one than the other, I am leaving the RSI(13) on the chart for now.

That is a reasonable idea to a point. With HA candles, many of the small retracement moves seen as obvious with simple price bars will get smoothed out, so the HA color does not change and you stay in the current direction. However, you must accept that for a solid up leg on H4 that might run 15 bars with no color change in the HA candles comprises on H1 a cycle of 1-2-3, up-down-up. So H1 WILL give a definite color change during the down retracement, signaling that you should exit the trade. But NORMALLY, you should NOT exit on this signal, becasue it is a RETRACEMENT, not a direction change of the H4. On H4, that retracement just makes a few small candles that do not change color. When the H1 retracement ends and the continuation leg goes back up, of course H4 continues as well. Only when the retracement goes down "too far" and is threatening to break the up trend of the H4 should you exit the trade. This is where you make a decision by looking at the other indicators.

For now, you can look at H1 and set the trade direction window using the RSI(13) crossing 50; that is like using H4 to set the window. Then use the candle color changes and the stoch8 swings to make add-on trades. That's the basics. I'll add later more details that include the Hull Moving Average and how to "predict" that the direction window is likely to change direction. (I was a little busy the last couple of days with pip harvesting.)

U Can Do It!

Here is an example of how nearly ANYBODY, including a 9 year old child, can make 10 pips a day!

Attached are the EURUSD H1 and m15 charts.
The bottom window has stochastics. The yellow line shows the stochastic(8,3,3) of the H4 chart, so I do not even need to look at H4 chart to know if buying or selling is favored.
The price bars are the standard Heiken-Ashi indicator that comes with MT4.
The moving average is a Hull Moving Average (12) coded by mladen (attached).

On H1, looking at the left side, clearly, this is a strong down move, and the yellow stochastic line shows me that the H4 went into strong down momentum along with this H1 chart. I am now looking for rallies to sell.

On H1 at the purple vertical, the candle before it is the very first candle that closed colored white, and the HMA has changed color to green. The stoch8 has gone up nearly to 50 with a steep angle. So it appears that a rally has started. I want to sell this rally, if it really is one, as soon as possible after a CONFIRMATION that it has stopped going up. I COULD continue watching the H1 for confirmation, which requires at least 1 red candle closed below the HMA line and the HMA changed from green to yellow. That signal comes at the close of the red candle just before the 2nd purple vertical. So we could sell at the open of the candle with the purple vertical. However... WE CAN DO BETTER! Check the m15 chart.

I marked the m15 with 2 purple verticals at the same times as shown of the H1 chart. We can be extremely confident at the left vertical that we are indeed in a rally because there are 3 white candles that closed above the HMA, the stoch8 has reached all the way to 80, and the yellow stoch is starting up. Is this a rally, or is it a trend reversal? NOBODY KNOWS! We can only determine that in hindsight. But it does not matter, because we are not in buy mode anyway; we are in WAITING mode for the rally to end to make a sell. There is no guessing or hoping involved--simply OBSERVING, and if the sell signal is observed, take action! If the sell signal never appears, there is never a sell made.

So now we look for the m15 to display CONFIRMATION that the rally has ended. The signal: At least 1 red candle closed below the HMA; the HMA has turned yellow; the stoch8 has turned down. That signal is actually 2 bars before the red vertical, so the sell order is at the open of the candle before the red vertical. I placed the red vertical one more candle later where the stoch8 has crossed 50 and the yellow line has turned down--there is no stronger evidence for a continuation sell than that combination of signals!

Now look at the difference in entries by using the open of the bar marked by the red vertical compared to the purple vertical after it that marks the entry if using the H1 chart. Bar open m15 entry: 1.13448. Bar open H1 entry: 1.13234. That's 21.4 pips!!! We are TARGETING a mere 10 pips! Plus, entering on the m15 gets you in a trade 1 hour 45 minutes before the H1 entry, so you get to LEAVE THE COMPUTER and do whatever else you want to do almost 2 hours sooner, knowing that there will be pips in your account when you return.

And if you were worried about the H4 crashing your pip party, remember to simply look at the H1 yellow line--look how it barely got above 20--absolutely no chance that it is going up for a new trend. (After m15 goes down, how can H4 POSSIBLY go up, if price is going DOWN?? DUH!)

Not seeing any rallies on the H1? Open the m15 and m5. Look for a rally to start on m15, and do the same confirmations on m5 as above. HOWEVER, keep H1 open, because if the rally you see starting on m15 ALSO appears on H1, m5 is now the wrong chart for making the sell--close it. You must move to the m15/H1 set again and do exacty as above.
A new month is just starting--let the pip harvest begin!

Hello all, how is everybody doing?

Sorry, I've been busy lately (not harvesting pips lol).

So, let's get to it. The first thing I want to discuss is the money management to trade using Lukass' original idea that I told you I was testing. I went ahead and made some changes to the original EA. Those are: I set a max order number, so no more than four orders are open at a time in one direction. Even if the EA receives more signals, it won't open new orders. The second thing I changed about it is the lot size, it is 0.05 lots per $1,000 of account balance for the first order, add-one order #1 is Lots*2, add-one order #2 is Lots*3 and add-one order #3 is Lots*4, so we're increasing the lot size. Now, the trading here will be more aggressive, meaning more money will be made when there are wining circles, and of course, more money will be lost when it turns against you, but according to my back-testing earlier, I noticed that there are more winning months than there are losing months, so I said let's give it a shot and see how it goes. The test that I tried required kind of a large capital, and it's a long term thing, not a fast money making machine. So, the capital that I used for the testing was $10,000. We start with that and we create two accounts, one is called Reserve for example and the other is called Active. On day 1 of the month, Reserve account must hold 90% of our overall capital (in this case $9,000) and the Active account holds the rest 10% (in this case $1,000). We let the system run aggressively on the Active account for a month. On the last day, we close all trades win or loss, and we see what is our overall capital now (Reserve account + Active account), and we redistribute our balance so now 90% of the new capital is in the Reserve account to be safe, and 10% percent is in the Active account. This way, each month, if we're winning, we're moving some of the profit to our Reserve account, and if we are losing, we are adding more balance to our Active account to stay in the game and make up for the losses as we keep trading. During winning months, we can win anywhere from 1% percent of our overall capital up to 32.90%, and if the worst happened and we blew our account, we're only losing 10% of our capital. So, this is not a very original idea, but I wanted to see if it works.

According to my back-testing, starting from January 1, 2014 to July 31, 2019, this would've been a winning system, and here are the results.

We started 2014 with $10,000.00 overall capital that is balanced monthly, and ended the year with $14,614.99. That is a 46.15% increase.
We started 2015 with $14,614.99 overall capital that is balanced monthly, and ended the year with $17,568.56. That is a 20.21% yearly increase, and 75.69% overall.
We started 2016 with $17,568.56 overall capital that is balanced monthly, and ended the year with $28,365.15. That is a 61.45% yearly increase, and 183.65% overall.
We started 2017 with $28,365.15 overall capital that is balanced monthly, and ended the year with $39,487.33. That is a 39.21% yearly increase, and 294.87% overall.
We started 2018 with $39,487.33 overall capital that is balanced monthly, and ended the year with $43,516.50. That is a 10.20% yearly increase, and 335.16% overall.
We started 2019 with $43,516.5, but so far we're losing and on July 31st, our capital should've been $36,255.52, that is a -14.05% loss. But, our overall capital is still 262.56% higher than our initial capital, which was $10,000.

Here's a graph of the overall capital growth.

Now, moving to DrDave's MACD idea, I've finished the EA, and did some quick testing to see if it does what it's supposed to do, and it works. Regarding the profitability of it, I didn't get happy results, but we can run more back-testing and fine tune it to see if we can turn it into a profitable one. But so far, it doesn't look good.

So, here are the settings that you can change:


"Buy/Sell zone initiated by:" As you can see, you can choose the Signal Line to indicates the buy/sell zone, or you can change it to use the MACD Line.
"Would you like it to open an initial order?" You can choose Yes to open an order whenever the buy/sell zone starts, or No to ignore that first order.
"Would you like to set a stop loss?" When I did a quick test to see if it worked, it seemed like the stop loss was causing a lot of losses as sometimes it doesn't allow the price to turn, so I've added this option for you to test it with stop loss or without a stop loss. If you choose no, there will be no stop loss, and all open orders will be closed when the zone for the opposite direction starts.
"Lot size:" Adding a lot calculator to calculate lots based on risk wasn't a very smart idea, because especially for initial orders, the risk (pips between open price and stop loss) is high mostly, so the lot size will be smaller, and then for add-on orders, that risk is lower so the lot size is bigger. So, even if you have the initial order making a good profit in pips, your add-on could wipe that out because it has larger lot size, so I took it back to a fixed lot size that can be set externally.

So, without further ado, here's V2.00 of the EA. I would suggest turning on Visual mode when you back-test it as it helps you see what is happening and also gives you ideas on how to improve the trading style.

DrDave, this is a really interesting system and it seems easy to follow. My only question is, what is the exit strategy here? When to take the profit and close the order, and when to take the loss?