1 (edited by hannahis 2015-07-10 08:22:59)

Topic: Longer Time Frame: How Fast Do You Want to React to Market Changes?

How Do you make your trading decisions?

Do you prefer your EA to:

1A. Able to scan all the various time charts to calculate the best possible entry position, or
1B. Observe only 1 time chart such as H4 and enter into a trade position based on 1 time chart calculation, i.e H4?

When a market started to move significantly, do your prefer your EA to:
2A.  Able to scan every minute and enter a trade any minute when it is right, or
2B.  Only make the decision every 4hr only when Bar opens/close, if during the current 4th hour, the market is not ready, wait till the next 4hr later to make another decision.  And in between if the market move significantly, don't do anything but wait for the next 4th hour to decide.

As for me, I would like my EA to watch market on various time charts and make a decision anytime the market is ready and not wait till the next 4hr when the bar opens or close.  Hence, wouldn't anyone who uses a H4 time frame in FSB Pro be waiting for every 4th hour to decide when to open a trade or not? 

Would a manual trader look at his clock and allow himself to trade only when the clock strike every 4 hours and he uses only 1 time frame to make his decision?  That sounds odd to me.  Hence, wouldn't anyone who uses H4 FSB Pro time frame to generate strategies be doing just that, delaying their trading decision every 4 hours instead of be ready every minute when the market is ready?

Please correct me if my assumption on how the H4 time chart on a FSB Pro opens trade only on a 4 hourly basis is wrong.

If anyone wonder whether can 1min FSB Pro strategies make profitable EAs, take a look at this EURUSA Best EA portfolio, it's make up of mostly 1min strategies and it make 70k in 4 days.  All EA with opens with 1 lot.  DD 15% Profit 70%. 

http://www.myfxbook.com/members/HANNAHI … ea/1297917

As for me, I use 1min chart with many Longer Time Frames so that my EA scan all most time chart and ready to react any minute when the market is ready.  Hence, I wonder why would anyone restrict their EA to a react slower when using higher time chart to generate their strategies.

My other portfolio has hundreds of EAs (both good and bad ones) yet still making money.

Hannah

2

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

How many slots to you use in FSB to OPEN positions and EXIT?

3 (edited by hannahis 2015-07-10 12:07:30)

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Hi GD,

It depends, it's about 5 to 12 slots for open and about 1 (such as pv pt, or trailing, or close n reverse) to 5 slots for closing.

1. I usually use H4 (LTF) indicators as the 1st input and then I put H4 for closing. 
2. Then next I add in 15min indicators both open n close and then
3. default (1min) both open and close. 
4. add in other time frames if you want to make it stricter or eliminate sideway trending or false signals.

I haven't experience using the Signal repeat and shift repeat but I think it is useful in 1min charts to reduce entering too early or eliminate false signals.

There 3 time frame (H4, 15min and 1min) forms the main "structure" of the trading theory.  For each time frame e.g. H4, I may use 1 or 2 indicators, especially those with "higher than" options.

Because if H4 is ready (i.e. Higher than), the 15min would have already be ready (also higher than).  And the default is to make sure you enter at the best optimal price).

If you want to make your opening conditions stricter, then I would add in H1, 30m, 5min or D1 (rising)

Try to change the parameters of these 3 time frame till you see reasonable profits.  I use BE 150-200, TP 800 to 1300 and SL 450- 500 and Winner 4 to 8 slots, (try 4 slots 1st, if you increase slots, then decrease TP because with each new open trade, your SL move up when your new trade reach the BE mark, hence it is a good way to protect my profits).


As for Closing, don't make it as strict as Opening conditions.  As long as your Longer Time frame is ready, it would be fine.  Usually it is only the main 3 "structure" time frames that I would use i.e. H4, 15m and default. 

Another alternative methods of Closing would be using Pv Pt, Trailing, or any other "dynamic" closing that is not rigid.  I'm waiting for Popov's new upgrade whereby we have more closing options to make our closing conditions more "dynamic" and responsive.

Hannah

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Demo or Real account  ?

My 'secret' goal is to push EA Studio until I can net 3000 pips per day....

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Hannahis, I wanted to ask you about your methodology earlier, and now you covered this a bit, allow me then to dig deeper. So, for higher timeframes you choose looser logics? By looser I mean "rise,fall, higher,lower" group. And for more precise entry in lower timeframe(s) the logic is stricter like a cross or direction change?

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

It's Demo, when it is real...Popov will know cos I would like to contribute smile

I usually use "higher than" for almost all time frame.  Because in my opinion, if I use a crossover condition and if other time frames not ready when 1 time frame crossover, then I missed that whole opportunity.  So choosing "higher than" is a better option for me.  For D1, I use "rising"

Hannah

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

I see that the win rate is  53 per cent......

My 'secret' goal is to push EA Studio until I can net 3000 pips per day....

8 (edited by hannahis 2015-07-11 13:41:14)

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Blaiserboy wrote:

I see that the win rate is  53 per cent......

Hi Dave,

Yes, this portfolio has about 70 EAs, still work in progress, far from perfect.  I'm still observing these group of EA to see their performance consistency.  Because FSB Pro historical data for 1min chart has only about 1.5 months of historical data to work on, I'm not sure how consistent these EA would perform over a long time.  So far, it's been good for about 2mths. 

I'm now starting to download historical data on my own for 1min time chart, wish next time FSB Pro can offer long historical data for users to work on without having to download on their own.

I'm still trying to work on the closing conditions.  There are a few EA that have 70% to 100% entry accuracy but low exit accuracy.  If anyone has any suggestions for any good closing strategies, do share your tips here for the benefits of the FSB community (don't have to share your trading secrets but at least point others to some directions or suitable indicators instead of searching through tons of information). 

Ideally, I wish I could set my closing conditions with All the following conditions.

1. With Trailing stop
2. Pv Points
3. Close and Reverse
4. My trading theory closing conditions

One more question:

If my assumption about how the FSB Pro would only open a trade at every 4th hour during Bar opening for a H4 time chart is correct, then why would anyone still want to use such a slow responding strategy? 

Wouldn't it be beneficial for everyone to use the 1min or 5 min chart because it also offer all the benefits of using all the other time frames (utilising Longer Time Frame features) calculation?

The only limitation for 1min and 5min charts is the short historical data for one to work on to develop a consistent EA that stand the test of time.  So wouldn't it be good for FSB Pro to offer longer historical data for 1min and 5 min chart and do away with the rest of the time charts (because seriously, who want to have such laggy EA that respond so slow to market changes?). 

If the 1min chart is too responsive for some indicators, then just use the signal repeat or shift repeat features to reduce the "sensitivity" of the EA and reduce false signal.

It seems like people mix up FSB 1min chart with trading on a MT4 1min chart's indicators.  Yes, if I trade based on the MT4 1min chart indicators, I will end up with many "unwanted and too frequent" trades.

It seems like most people has the wrong notion of how the FSB Pro 1min chart works.  FSB pro 1min chart means that the EA is checking all my opening and closing conditions every 1 minute to see whether the current market condition fulfill my strategy conditions.  And likewise, a 5min FSB strategy means, your EA will check whether there is any possible opening trades every 5min.  Hence, how fast or frequent do you want your EA to check the market, depends on which chart you use. 

So for EA that is based on 1 min chart, the EA is checking the market conditions every 1 min whether the H4, 15min and 1min indicators ready for entry.

Hannah

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

I wonder if you have traded this manually using real money.

Actually, using the 4 hour chart may give you a much higher win rate and a longer life strategy with much greater profitability.

You are using data points from the higher time frames........ meaning you are relying on the 'lagging'.

I am not so sure you have complete understanding of higher time frame trading. On one hand you want to use the higher time frame data and on the other hand you call it lagging.  So you are using the lagging data....

I was of the impression that FSB is checking ticks, have I got that wrong?

Are you sure about the way FSBPro is checking the data?

My 'secret' goal is to push EA Studio until I can net 3000 pips per day....

10 (edited by hannahis 2015-07-11 14:33:05)

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Thanks Dave,

I'm not sure whether you understood me correctly about the 1min time chart compared to H4 time chart.  Data and respond rate are 2 different issues. 

1min time chart is the responsive level and when I use the H4 Longer Time Frame in the 1min time chart, I'm using the H4 data.  These two are different aspect that I'm talking about (data and rate of respond, i.e. open trade at every Bar opening).

Am I right?

1min time chart check whether to open a trade at every 1 min Bar opening.

H4 time chart check whether to open a trade at every 4 hour Bar opening.


So choosing a H4 time chart isn't it similar to checking the market every 4 hours as compare to checking every 1 minute for a 1min time chart?  If this is correct, won't that be laggy?  (I'm not referring to H4 data, I'm refering to H4 opens trade only at every 4th hour Bar opening).

I'm more than glad if anyone can point out to me that I got it all wrong and hence take this chance to clarify my understanding for my benefit.

Are you saying that a H4 time chart checks the market at every time/ticks and enters opens a trade anytime/ticks instead of at Bar Opening? (if at Bar opening, which Bar? H4 Bar opening or 1min Bar Opening?)  So do you mean to say the H4 time chart will open a trade at 1min Bar Opening?

Hannah

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Everything gets recalculated at every tick, positions though get opened when conditions are fulfilled i.e how the strategy is set up by the user. Using LTF means spreading one particular signal for more bars - let me illustrate, one H4 signal is valid for one bar in H4 TF, but if LTF is used, lets say on M1, then the very same signal is valid for 1x60x4=240 bars on M1 chart.

The starting point/starting time for the signal is the same, so saying LTF cuts lag is not correct. For instance, if a condition is fulfilled in 12:00 bar in H4 chart, the signal is given in 16:00 bar; if LTF is used, the very same signal is still given at 16:00; if M1 is used to find the same condition and it finds it at 11:30, the signal is given at 11:31.

12 (edited by hannahis 2015-07-11 16:00:56)

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Thanks Footon for joining in this discussion to help me understand LTF and H4 better.

So when we input our H4 strategy, it has "enter the market at the beginning of the bar", so you mean, it will enter at any timing instead of every 4 hour of the Bar opening in a MT4 H4 time chart?

Another question

Can I use a H4 time chart in FSB Pro to generate strategy and then export it as EA and use it in 1min MT4 chart or must it also be in a MT4 H4 chart?

Hannah

13 (edited by hannahis 2015-07-11 17:34:48)

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

footon wrote:

if a condition is fulfilled in 12:00 bar in H4 chart, the signal is given in 16:00 bar.

So between 12:00 and 16:00, isn't there a 4 hours lag or delay between condition fulfilled and a trade open?  Furthermore, wouldn't the market condition change by then at 16:00 (after 4 hours delay) and no longer the same condition as the 12:00 conditions?

Whereas in 1min chart, the signal (open a trade) is fulfilled in the next minute.  So the difference is only 1min compare to a H4 chart, which is 4 hours?

Hannah

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Can I suggest that you start the thing in MT4 and observe closely so that you gain understanding.

Try three or four different time frames and observe closely.

That will enhance your strategy bulding greatly.

My 'secret' goal is to push EA Studio until I can net 3000 pips per day....

15 (edited by hannahis 2015-07-11 17:42:42)

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

footon wrote:

Everything gets recalculated at every tick, positions though get opened when conditions are fulfilled i.e how the strategy is set up by the user.


So position opens when condition is fulfilled at every tick or only at "enter the market at the beginning of the Bar" (i.e. Bar opens)?

Because if position opens at any tick (when conditions are fulfilled), then why make the distinction between "Bar Opening" and "Bar Closing".  Since it's at every tick when the conditions are fulfilled, it wouldn't make any difference if Bar opening or Bar closing.  But based on your another explanation, for the H4 chart, it would enter into trade at 16:00 which is 4 hours later, not the next tick.

Hannah

16 (edited by hannahis 2015-07-11 18:28:54)

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Blaiserboy wrote:

Can I suggest that you start the thing in MT4 and observe closely so that you gain understanding.

Try three or four different time frames and observe closely.

That will enhance your strategy bulding greatly.


Hi Dave,

I think my question is not about how my strategy work, but how the FSB Pro work.  Whether the FSB Pro H4 time chart opens trade at every tick or every 4hrs and whether if I use the FSB Pro to generate H4 time chart strategy, does it mean I also export the EA and use it in a H4 MT4 chart.  Because these were my assumptions all the long and I wonder I was right to assume this way. 

These questions have nothing to do with my strategy.  it's about understanding how FSB work.  It can be answered straight away without me wasting my time observing and figuring it out.  And if my assumption about H4 strategy opens trades at 4 hours interval when a condition is fulfilled, is correct, isn't it right to call that laggy.  Who on earth trade that way manually, see a condition fulfilled but wait 4 hours later till the next Bar opening to trade? (as Footon mentioned condition fulfilled at 12:00 but signal to open trade given at 16:00, which is 4 hours later).  I wouldn't trade that manually neither would I ever imagine who in their right mind would ever want to generate a H4 EA to behave that way.  How profitable can that EA be, it's always 4 hours behind time.  The market moves very fast...within 4 hours many things can have happened

Hannah

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

well, load some strategies into  Metatrader and observe closely how it reacts to the indicators, it is all laid out on the screen for you.  It is constantly evaluating.

Just load a couple and watch and you will see!

I have an idea that your thinking about how the program works and higher time frames is way off base, I think you should do some actual testing to observe with a simple chart. Just place an indicator on the chart and figure out what  that same indicator would be from a higher time frame on that same chart.

I am sure that you will find you can have simple strategies with less indicators by understanding that you can use single time frame and benefit from what happens on longer periods by adjusting the lengths in your indicators.  Once you can see that, then you can add higher time frame stuff, first you have to comprehend the relationships between time frames.

For example  Ma 500 on a one minute chart would approximate Ma 100 on a five minute chart.

There is no need to complex things about waiting for bars to open.

You can rest assured that Popov has taken this stuff apart and reassembled it many times to get at the root of bar structure.

If you can adjust your thinking a bit you will see far better results and be far less confused.

Enjoy the ride!!

My 'secret' goal is to push EA Studio until I can net 3000 pips per day....

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

hannahis wrote:

So when we input our H4 strategy, it has "enter the market at the beginning of the bar", so you mean, it will enter at any timing instead of every 4 hour of the Bar opening in a MT4 H4 time chart?

How's that possible? You set your opening point of position, be it bar opening or bar closing or whatever technical indi, and off it goes according to the selected indi and logic, that's how FSB works. If it doesn't do that, please report because it might be a fault of some kind.

So position opens when condition is fulfilled at every tick or only at "enter the market at the beginning of the Bar" (i.e. Bar opens)?

I don't see I talked about position openings, I talked about recalculation, indi value recalculation at every new tick, it shouldn't come as a surprise I assume.

So between 12:00 and 16:00, isn't there a 4 hours lag or delay between condition fulfilled and a trade open?  Furthermore, wouldn't the market condition change by then at 16:00 (after 4 hours delay) and no longer the same condition as the 12:00 conditions?

Bar time marks bar opening time, depending on some aspects or parameters previous bar value (look it up in the manual!) is used or not used, generally it is needed. So, if previous bar value is used, it has to wait until the close of the bar to verify if condition is truly fulfilled or not, you can call it lag and in some ways it makes sense, but in terms of reliable backtesting and ability to reproduce trading logic after the fact and vice versa previous bar value is used, which means signal is given at the next bar. In my example bar opens 1200, closes 1559, condition is fulfilled during the bar, signal is given at 1600. That way trading and backtest are alike, which is what everybody wants, isn't it?

19 (edited by hannahis 2015-07-12 04:23:59)

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Hi Footon,

Thanks for your clarification.

If I use H4 time chart, I want my EA to able to enter into trade, i.e. open a position anytime at the next tick whenever the conditions are fulfilled.  How should I do that? 

So in order to prevent my H4 EA to respond with a 4 hours "lag", I should avoid using Bar Opening in my opening conditions and Bar Closing in my closing condition and also dis-select the "use previous bar" option, right? (How to remove this "use previous bar" option?)

But if I don't choose Bar closing as my closing conditions, then I'm left with very limited preset conditions/options (which I can only choose 1) on the left hand menu and I can't input my own trading conditions i.e. add more closing conditions.  Hence, I ended up with only 1 closing option to input.

Hannah

20 (edited by hannahis 2015-07-12 04:37:46)

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

In my example bar opens 1200, closes 1559, condition is fulfilled during the bar, signal is given at 1600. That way trading and backtest are alike, which is what everybody wants, isn't it?

I definitely wouldn't want to trade that way (be it manually or with an EA).  If the condition is fulfilled at 12:00, I definitely want to enter in trade at the next tick or at the next minute, definitely not 4 hours delay.  Don't think anyone out there wants to trade that way. 

For those who have been following this post, do throw in your opinion.  Do you want your EA to be limited by such delay because of the closing conditions?  I'm not against reliable back testing, but as an end user/customer, I certainly wouldn't compromise my trading speed for such an option because, my trade isn't "reliable" anymore, by the time I enter a trade at 16:00, based on 12:00 conditions, at 16:00, the 12:00 conditions are "irrelevant, invalid and unreliable" input for me to trade at 16:00 because market conditions would have changed drastically by then.

Hannah

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Hello Hannahis,

Please take a look at the User Guide: Indicator's Advanced Properties => Use previous bar value, and read this more detailed article: Use previous bar value. If you have questions on the topic, we'll be happy to answer.

Here is another article that may help you understand better why we use "Bar Close" concept: Entering the Market at Crossover

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

hannahis wrote:

If the condition is fulfilled at 12:00, I definitely want to enter in trade at the next tick or at the next minute, definitely not 4 hours delay.  Don't think anyone out there wants to trade that way.

I want to trade that way. Conditions can occur multiple times a bar, to me it is whipsaw. Secondly, if I believe in backtest, which I do, I can have better understanding of strategy's performance in a reliable manner.

For those who have been following this post, do throw in your opinion.  Do you want your EA to be limited by such delay because of the closing conditions?  I'm not against reliable back testing, but as an end user/customer, I certainly wouldn't compromise my trading speed for such an option because, my trade isn't "reliable" anymore, by the time I enter a trade at 16:00, based on 12:00 conditions, at 16:00, the 12:00 conditions are "irrelevant, invalid and unreliable" input for me to trade at 16:00 because market conditions would have changed drastically by then.

It can go both ways, maybe the first signals are false signals, market conditions might turn out to be not favourable by the end of the bar. And you have to remember that not every condition appears on the first minute of H4 bar, it can also be in the middle or at the end of the bar, reducing the so-called lag.

Thirdly, I do understand your concern, and I have a suggestion. Lower TFs build higher TFs, hence it's possible to track higher TF in the making while having fixed price points (lower TF bars) for reliable backtest as per FSB concept. Therefore, why not select a lower TF with indicator settings following higher TF? It's not 100% to the point in some cases, but usually it is close enough. It would look like this:
objective: track H4 with MA 10 on M30;
solution: 240/30=8  10x8=80  -------> MA 80 on M30 tracks H4 bar.

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

hannahis wrote:

Hi Footon,

Thanks for your clarification.

If I use H4 time chart, I want my EA to able to enter into trade, i.e. open a position anytime at the next tick whenever the conditions are fulfilled.  How should I do that? 

So in order to prevent my H4 EA to respond with a 4 hours "lag", I should avoid using Bar Opening in my opening conditions and Bar Closing in my closing condition and also dis-select the "use previous bar" option, right? (How to remove this "use previous bar" option?)

But if I don't choose Bar closing as my closing conditions, then I'm left with very limited preset conditions/options (which I can only choose 1) on the left hand menu and I can't input my own trading conditions i.e. add more closing conditions.  Hence, I ended up with only 1 closing option to input.

PBV was user-selectable in the previous FSB legacy, but in Pro it is not. Unfortunately some users are unable or have no intention of learning the concept, they took "the easy away out" by deselecting PBV because essentially backtest was using future data and had spectacular equity curve. Of course live trading is not even close to these results and depression, anger and resignation takes over. So, to avoid this, KISS principle is used to protect one from oneself. Advanced users can modify the indicators to remove PBV, but one must know exactly what it can bring.

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

Thanks Footon

Hannah

25 (edited by hannahis 2015-07-12 17:01:09)

Re: Longer Time Frame: How Fast Do You Want to React to Market Changes?

I do understand your concern, and I have a suggestion. Lower TFs build higher TFs, hence it's possible to track higher TF in the making while having fixed price points (lower TF bars) for reliable backtest as per FSB concept. Therefore, why not select a lower TF with indicator settings following higher TF? It's not 100% to the point in some cases, but usually it is close enough. It would look like this:
objective: track H4 with MA 10 on M30;
solution: 240/30=8  10x8=80  -------> MA 80 on M30 tracks H4 bar.


Thanks Footon, your suggestion is such a relief to me, finally at least someone is able to follow my trains of thoughts and understand the points I'm trying to present here and offer very practical and helpful solutions.  And at the same time confirmed my understanding on how the H4 opens a trade on the next signal.

Hannah