Profit Generator EA - page 69

 

Hi!

After 3 days running tick backtests on EURUSD (3 first weeks of march on H1 - PG 3.2.3), and before have hard analysed results, there seems to have good settings with longbar 5->15, BarOpenPercent 0.7-0.9, SL around 50 and TSactivation around 70 with trailingpips=5, obsolete minutes 30-60 with profit only and reverse=true.

Obsolete method brings a lot to profitability but mainly associated with profitonly=true.

It was just for info. I'll tell you more after deep analysis. Now I'm going to select parameters range that seems to work and then run longer backtests on those (3 months). I will also test some kind of trailing with fibo retracement %.

Also, even if history are the same, tick backtests results seems to differs a lot from MT4 tick simulation. Sure 3 weeks is not a lot and I will check again on longer history.

jojo

 
novafx99:
Holyguy, now that is some serious testing and some nice trading. That JPY is a nasty one, eh? Always the trickster. No, question that the different pairs have different patterns. Just be careful to not get caught finding settings that reflect a "snapshot" in time. The behavior of the pairs does change over the course of weeks/months. It may be difficult to keep up with and find the best settings as things change.

I agree, Nova, that is why we need to continue the work of testing and finding good settings for all currency indefinately. The problem with many trading systems is that they work great for a while then stop working because the conditions on the currencies have changes and now they "trend" different. I am seeing this problem right now with Support and Resistance. It used to work great last year but this year, it has "stopped working" and many professional traders which relyed upon it are in trouble having to figure something else out. The great thing about this EA is that it is not based on some magic line but instead based solely on price movements. As the price movements change, our EA can change along with it to keep up with what is going on now.

That is why I really don't think backtesting works beyond about 3 months because longer than that, the market has changed quite a bit.

Perhaps if somebody could program this EA for Tradestation and we can do accurate backtesting on that platform it would be an improvement or we can bug metaquotes to create an accurate backtester that will work with Metatrader 4.

I think we are in good shape here and, yes, I am trading this EA live starting this week and am down 130 pips. Not a big deal but I see rebound in sight now that the news is over.

Yes. I accually do like the JPY currency pairs a lot because they move a lot of pips very quickly. The downside is that they are very inpredictable sometimes. However, I believe going to a higher timeframe is the key with those pairs and using the reverse feature as it does appear to work very well.

Sorry, I have been away from the board for a while. I have had some things going on and needed to take care of them. Many could give up on this EA because of some news and tough trading or we could think of that as an opportunity.

IDEA:

We all know that in ranging markets that this EA works very well, however, when trending on news starts, it craps out pretty bad. Anybody have an idea on how if trending does start that the EA could automatically begin the "REVERSE" feature which would work well during trending markets?

 

Hey Nich, Your SCALPING SYSTEM WORKS!

At least it works great so far on the USDCAD. Check out this statement.

Folks, let's think outside the box here.

USDCAD (M15)

Longbar: 5

Longbar%: .6

Max Trades: 3

Stoploss: 25

Take Profit: 80

UseClose: True

TSActivation: 5

TrailingPips: 0

OneTradePerPeriod: True

No Obsolete

ADDED:

I added in a screenshot of how the scalping system is doing. Look at the trades yourself. Very Nice.

21 Wins

0 Losses

SWEET!!!!

 
holyguy7:
I agree, Nova, that is why we need to continue the work of testing and finding good settings for all currency indefinately. The problem with many trading systems is that they work great for a while then stop working because the conditions on the currencies have changes and now they "trend" different. I am seeing this problem right now with Support and Resistance. It used to work great last year but this year, it has "stopped working" and many professional traders which relyed upon it are in trouble having to figure something else out. The great thing about this EA is that it is not based on some magic line but instead based solely on price movements. As the price movements change, our EA can change along with it to keep up with what is going on now.

That is why I really don't think backtesting works beyond about 3 months because longer than that, the market has changed quite a bit.

Perhaps if somebody could program this EA for Tradestation and we can do accurate backtesting on that platform it would be an improvement or we can bug metaquotes to create an accurate backtester that will work with Metatrader 4.

I think we are in good shape here and, yes, I am trading this EA live starting this week and am down 130 pips. Not a big deal but I see rebound in sight now that the news is over.

Yes. I accually do like the JPY currency pairs a lot because they move a lot of pips very quickly. The downside is that they are very inpredictable sometimes. However, I believe going to a higher timeframe is the key with those pairs and using the reverse feature as it does appear to work very well.

Sorry, I have been away from the board for a while. I have had some things going on and needed to take care of them. Many could give up on this EA because of some news and tough trading or we could think of that as an opportunity.

IDEA:

We all know that in ranging markets that this EA works very well, however, when trending on news starts, it craps out pretty bad. Anybody have an idea on how if trending does start that the EA could automatically begin the "REVERSE" feature which would work well during trending markets?

Look, I'm not trying to start an argument and this has nothing to do with my S/R ideas and this EA but to say S/R doesn't work this year? Come on, Plot the pivots on your chart everyday. Price either trades in them, using the levels quite faithfully or it breaks out of the range. Breakouts usually happen about three times a month on average. Pivots have worked for the 10 years I've been using them and I'm sure they will work 10 years from now.

 

I agree with Nova on the pivots....

I've used them for a while and they seem to be the most consistant way to trade.

I do feel that it will be a tough EA to program being that price can bounce around pivots quite a bit triggering many false signals!!!

I think pivots are to be traded manually with discretion.

Billy

 

Hey Nich,

Any thoughts on that trend manager EA I directed you to? It has great potential and may be able to morph with this system. It will always keep us on the right side of the trades based on the price action of this EA.

It's worth looking at!

tell me what you think.

Billy

 

Hard to program, easy to trade

bagovino:
I agree with Nova on the pivots....

I've used them for a while and they seem to be the most consistant way to trade.

I do feel that it will be a tough EA to program being that price can bounce around pivots quite a bit triggering many false signals!!!

I think pivots are to be traded manually with discretion.

Billy

If you look at the filters for pivot trading I had in mind, I think it can work pretty well. Like if you said, We only go long if price is X pips above the pivot and we have X pips headroom to the next level, reverse for short. With proper stops and even trailing stops, this could probably work well. However, I'm definitely sure the programming wouldn't be a breeze.

 

Today's trading

Well, I had some drawdown overnight but I've had a great afternoon so I'm still in profit for the last two days. It seems that volatility of any type creates a string of losses. Prices have been working around the trading range from pivot point up for the last two days in the majors against the dollar. I do see on my charts that if the pivot filtering was used, I wouldn't have had the drawdown because the rules would have prevented the bad trades from having opened. Likewise, the same rules would have allowed the profit trades, that is if the signal was generated. That's really the big question about marrying Pro-Gen to a system strategy. If we introduce rules will we get enough signals generated. If so, I think we would have a winner that wouldn't need settings to be curve fitted to different pairs.

 
holyguy7:
All default settings except the following:

EURJPY (H4)

Longbar: 10

Take Profit: 15

Stoploss: 30

Reverse

No Obsolete

What do u mean by NO OBSOLETE...u mean USEOBSOLETEMETHOD=FALSE

Thanks

 
babarmughal:
What do u mean by NO OBSOLETE...u mean USEOBSOLETEMETHOD=FALSE Thanks

That would be correct. UseObsoleteMethod: False