SAR system based on QQE Adv and MACD Platinum

Hi Buff,

Firstly cudos to your work, effort and generosity. Thank you.

I had an idea on your system to reduce the number of signals so had a play with the QQE and MACD settings.

I got some interesting results on the 30 minute chart with these settings.

MACD: 48,104,36
QQE: 4,32,12

You will get a lot fewer signals as expected but the big thing I noticed was that the number of losing sequences was vastly reduced.


Using these setting on the EUR/CHF since 31st December the longest losing run was 4 and only once.

What would also be a nice feature of this EA would be if you could set a TP on pips as well as monetary value as this changes from once currency pair to the next. If profit hit on first trade then the EA will shut that trade and wait for next signal, if TP is not hit then it goes into it's normal trading mode to break even as is already programmed. The reason for this is that I noticed on the EUR/CHF quite often a 50 - 70 pip value was hit on the first trade to close at either a loss or minimal gain.

Just a few ideas I've been playing with.

Buff something else I have been playing with is whilst trading the system manually once a signal is generated I look back about 30 bars, calculate the recent low and the recent high.

If a buy signal is generated above the average of the recent high and low then I would (on paper take the trade) if below then I'd ignore it and vice versa for the sells. So far I have noticed that this method reduces a lot of false signals and also negates the need for the TP option I mentioned in my previous post.

Also in the same vein I'm looking at making 40% to 60% also a no trade area.

So if a buy signal comes in, you look back 30 or so ticks, mark the recent high and recent low.

If say this were 100 pips, you look at the buy signal and see if it is at least 60 pips (equal to or above 60%) above the recent low, if so then you can ASSUME that you are trading with the recent trend thus enhancing your probability of a successful trade.

Kirk and Zennor, thanks both for the feedback. You both seem to be on the same track with adjusting the settings on the indicators. Also some good ideas there with regards to filtering out lower probability trades. You could really put some effort in and test quite a lot of variations here. I'll check out the ones you posted Zennor. Also there is a modification in the pipeline with regards to profit targets in pips, either on each sequence or within a specified trading session. So watch this space. I have also loaded the EA onto a few live charts now, both on several daily pairs where I can monitor them easily to make sure nothing gets out of hand, and also on the EUR 1hr. Done a lot of testing on the EUR 1hr running it without a profit target and it seems to produce some good results. Haven't really checked out many other pairs as I tend to stick with one pair when testing.

Trade update:

The EUR achieved the overall break even target so that sequence is now closed and a new one begins. So long 1x at 1.3488.

No other changes a this time. Cheers. Jim

Also I've been looking at the trades regarding the position of the trigger in relationship to the recent highs and lows and have come up with an idea to get on board so to speak if the trigger was actually a good one and got filtered out because of it's position.

The idea is to take the trade if the price goes in the direction of the trigger and breaks the recent high or low.

I'll see if I can get some pictures and post them up to explain more fully what I mean.

This is an example on a 30 minute chart with QQE and MACD settings as per my previous post.

As you can see from the image the sell signal was near a damn plumb in the middle of the recent high and recent low which would have prevented a losing trade.

In this image I discuss a possible method to prevent the previous posts method from getting in sync with the good signal trades.

Zennor,
I like your thinking and what you state makes some good sense. My only issue would be having to actually monitor the new signals when they appear and take the appropriate action in relation to past highs/lows. Probably something you could build into the EA but it would require some hard and fast rules. It looks more suited to manually monitoring the trades.

Trade update:

The EUR went short 1x at 1.3447, which resulted in a loss of -41 on the previous long. This gave me an overall break even target of 1.3406, which has since been well and truly taken out. So that sequence is closed at overall break even and a new sequence begins now with short 1x at 1.3406.

No other changes to the other pairs.

I think I'll be calling a halt to this forward demo test on the 4hr charts. I have done a fair bit of backtesting on the 4hr and daily charts with the EA, and the way it is at the moment, I can't seem to consistently produce a decent profit. But having said that, if I just used the EA for entries and then manually monitored the trades and adjusted the stops etc, then that is a completely different story. Also an advantage of being more hands on, is that I can dictate my position size a little better as sometimes I can just look at a chart and just know that I shouldn't be in a trade in that particular direction. The results on the 1hr charts seem to be quiet good, and it is a little easier to drill down particular trading times also. Then there is the option of profit targets etc, where Christina is currently working on a slightly modified version of the EA, so hopefully that will be in my hot little hand soon. So there is plenty happening. Cheers. Jim

Buff,

For sure the methodology of my proposed method is very much geared to the manual trading of your system and not the EA side of things. By smoothing out the QQE and MACD you get far fewer signals, just 3-5 per week sometimes but the results look very promising. Now to see how it does on the 4H charts with the smoothing which I suspect will need a bit more given that the time frame is 8 hours longer.

The reason for the proposed method regarding recent highs and lows was that the system does generate a lot of signals where the market just turns round on getting you out of a profitable previous signal and losing you money as well as adding additional stages to the recovery phase of your system.

Here is a continuation of yesterday's trade on the EUR/GBP pair and clearly demonstrates that by waiting and giving the signals a chance to be right you can turn a 50 pip loss into a 140 gain.

I've been talking with Jim about modifying some rules. The principles are the same, just worked out different entry and exit strategy based on different conditions.

I'm about to upload new video for new versions of EA, to prevent any confusion of the rules, here I'm listing all of the new rules under 4 possible conditions.

----------------------------------------------------------------

1. Nil trading times & nil profit target

Trading is conducted as per the original concept, from signal to signal using the fib sequence for position size. Always in the market.

2. Trading times & nil profit target

Multiple sequence each session, after close each sequence, new sequence of the same session starts immediately.
Closing options 1)Immediately 2)Next signal 3)Let it run.
When a session's trade finish, new session start on next signal.

3. Nil trading times & profit target

If first order reach target, close order and start new sequence at next signal.
If first order does not reach target but profitable, close order and start new sequence at signal.
If first order is a loss, let it run and finish sequence at break even. Next sequence will start immediately.

4. Trading times & profit target

Multiple sequence each session, but after a sequence makes any profit (on first trade) stop session trading. If during the same session one sequence end up break even, then start next immediately.

If by session end time, there is an order:

a). if this order is the first of sequence and it's profitable, close right away and finish session.
b). if this order is first of sequence and it's at loss when session close time, close it when order break even.
c). if it's not the first of sequence, means the sequence must at a loss by session end time, run it until break even.

New session trading always wait for the first available signal after previous session trading finish.

----------------------------------------------------------------

New videos will be uploaded shortly. Thanks.

Here is how EA supposed to behave under the current rules with 4 different conditions:

1. Nil trading times & nil profit target

http://screencast.com/t/YXgSJLSnqFe

2. Trading times & nil profit target

http://screencast.com/t/FA3wAGgD

3. Nil trading times & profit target

http://screencast.com/t/0SbvpZfaq3tv

4. Trading times & profit target

http://screencast.com/t/x5CzFM6MPVkI

Thanks.

Buff and Minmin830,

Cudos to the both of you, very nice indeed and I think the session will have a big impact if the EA is trading mostly during the EU/US trading times. Very nice.

Here is a clear demonstration of what I perceive to be a positive effect on the smoothing out of the QQE and MACD.

The second chart due to being uninterrupted by signals went on to post over 200 pips in gain whereas the first chart is trying to just break even after a host of losing signals.

One problem I have noticed with the EA is that firstly changing the QQE and MACD to a more smoothed out bunch of settings leads to the EA crashing, I personally have not managed to keep the EA going more than 24 hours on the smoothed setting. also if the EA has to be left live will it not automatically close all trades on the close of the Friday session leaving you no chance to continue with your recovery legs on Sunday/Monday on Market open?

"Fib numbers made up nonsense"

This is nonsense

Christina,

Thanks for the updated version and the videos to go with it. Done a bit of testing and having some good results so far. It is hard to nail down the smaller time frames though, but I must admit most of my testing to date has been on the 1hr and above and mainly on the EUR/USD. Might drop down to the 15 min charts when I can actually sit down and monitor the entire session just to see how it performs live. The videos were excellent as usual, and I like the new versions profit target being in base lot pips instead of $. Looking forward to more testing.

Zennor,

I have had a quick play with different settings but probably not enough to make any judgement call on performance, but I certainly will be looking into it further. There will be times of larger drawdown I assume and certainly not as much action. But at the end of the day, it is all about profit, so that's the end game. Work within decent money management rules and be patient, and all should be fine if you have a strategy that works long term. I also haven't had any problems with the EA closing trades at week's end, but I guess I'll find out Monday when the market opens. I do have the EA running live with 3x different brokers, so hopefully it won't be an issue. These are all on a VPS so that takes my computer out of the equation. I'll also set the EA up on a demo account with the smoothed out settings to see if I can replicate your crashing problem as I can't really comment on that at the moment.

Big call on stating Fib numbers are made up nonsense. Whether they work in trading is probably debatable, but Fib numbers in general are a huge part of life and make perfect sense to me.

Trade update:

This is it, the end of the road. All three pairs reversed to longs just before the weeks close, and all had profits on their previous shorts of approx 220 pips overall. Add it to the +3200 odd already achieved, then it has been a pretty good result in a pretty short time. Definitely not saying it would be like this all of the time as I was probably pretty lucky with some very good trending markets.

But that will do me, as everyone that is interested should now have a pretty good idea on the method itself and be able to make it your own if you so choose to. I have also provided a pretty basic EA that can do most of the work for you, where you can adjust a variety of settings to suit yourself. I will continue to add things to this thread in relation to testing and possibly other ideas that relate to this type of trading method. To those that have been participating and giving me feedback, thank you as it is greatly appreciated.

Also thanks to my Programmer, Christina, for participating here to show the process of an EA build. You can find her at Wise-EA if you require the services of a Programmer that knows her stuff.

So now onto some further testing and watching this EA perform on my live charts. This is where it gets a little more serious. Cheers. Jim

Hi Jim,

Actually the draw down is not as bad as you would expect as you will see when you try the new numbers.
Since December 18th 2012
EUR/GBP - 74 pips.
EUR/JPY - 240 pips
EUR/USD - 100 pips
GBP/CHF -120 pips

Regarding the EA, it not so much crashes , all the information to the left disappears, the smiley face remains, no new trades are taken on valid signals and current runs are ignored. This is on a 30 minute chart with the smoothed settings. I have set this up on a demo account the night before to see in the morning some trades have been taken but not all, all charts have lost the information to the left and the smiley is still there. I have seen this pretty much most mornings after loading the EA the night before.

I agree, anyone saying fibs are made up nonsense is either watching tick charts or just not paying attention

the QQE and MACD dual signal.mq4 (#97) compilation have an error due to the last #include at the end of program: missing bracket.
i cant get around that issue, and wonder that #include is declare at the end of program instead of the usual top lines.

dinofx, Obviously a question for Hanover re his indicator. It did have a problem initially but it was fixed, so not sure if this is a new issue identified by you. My indicator seems to be working fine. Contact me and I'll send you a copy of my working version if you would like.

Zennor, I have settled on the 1hr charts on multiple pairs using slightly extended settings of MACD 18,39,13 and QQE 2,12,5 and they seem to work fine. Testing them on a few live accounts now with a pretty low entry level just to make sure that it does work.

Just taking your idea a little further with cutting out the low probability trades, I was playing around with looking at a longer time frame with the same (or similar settings) and then only taking trades in that same higher time frames direction on the smaller time frame. For example, setting up the daily chart on the EUR/USD with the default settings, and then trading off the 1hr chart on the same pair setting the EA only to take trades in either the long or short only mode depending on which way the daily chart is heading. Hope that makes sense. Obviously a bit more work as I would have to check the trend at the open of each new day, but if it it was viable, it could be built into the EA at a future date. The fib sequence still works out ok if it is only taking longs or shorts also.

With regards to your problem with losing data on the chart overnight, maybe that is an issue with the version that is available on this thread. I have checked all of mine, and the data is all still clearly displayed. Either PM me or contact me through my website and I'll send you the updated version. Cheers. Jim