Ea That Trades Previous Candlestick (Question)

 

I paid a programmer to create a EA for me (I am currently testing the EA) that places trades in the direction of the previous candle or bar.

The ea has time settings adjustments so you can target specific highly probably candles at the same time every day if you want.

Does anyone know of any highly probably candles/bars (on the H1 chart) where it is almost always at least 2 bullish candles in a row or 2 bearish candles in a row?

I would aim the EA to trade the second candle in the direction of the previous candle.

If there are any such consistent candles (2 in a row) which currency pair?

The ea has a pending order option which can be turned on, so if the price goes a certain amount of pips in the opposite direction the pending order will be triggered. But the pending option only works with brokers that allow hedging.

 
hifromeddie:
......

Does anyone know of any highly probably candles/bars (on the H1 chart) where it is almost always at least 2 bullish candles in a row or 2 bearish candles in a row?

I would aim the EA to trade the second candle in the direction of the previous candle.

If there are any such consistent candles (2 in a row) which currency pair?

.........

Hi Hifromeddie,

If you say high probability you probably mean clearly higher then 50% or higher then flipping a coin.

If somebody would know whatever kind of element or happening that could occur with a certainty clearly higher then 50% then he would be richer then Warren Buffet in no time.

Price moves in a random way. So there is nothing that can be predicted.

Sorry, there are caracteristiques that can be predicted with a certainty clearly higher then 50%. But that will not show you the direction. For ex. one can predict that the UK and US session will be more volatyle then the night sessions But that is like knowing that more people will play on the roulette in the evening then in the morning. But it want help you to predict if red or black will turn up more then 50% of the times.

Friendly regards... iGoR

 

Thank you igor, that's what I thought. But just checking since I am new to Forex and am not too familiar with it yet.

I am hoping to add a new addon feature to the EA that ignores the stop loss and take profit, and instead opens a trade in the direction of the previous candle, and then closes when the candle closes no matter if it is a profit or loss...and then opens a new trade on the next candle.

A reverse option will also be added that trades in opposite direction of the previous candle.

I am curious to see if the reverse has more profits then the not reverse...or if it will be a 50/50 split etc.

 
hifromeddie:
Thank you igor, that's what I thought. But just checking since I am new to Forex and am not too familiar with it yet.

I am hoping to add a new addon feature to the EA that ignores the stop loss and take profit, and instead opens a trade in the direction of the previous candle, and then closes when the candle closes no matter if it is a profit or loss...and then opens a new trade on the next candle.

A reverse option will also be added that trades in opposite direction of the previous candle.

I am curious to see if the reverse has more profits then the not reverse...or if it will be a 50/50 split etc.

Hi hifromeddie,

I have done a LOT of backtest working with these kind of principle's.

For ex: If 1 bar is up then next bar long or if 2 bars up then next bar long or 3 bars up then 4th bar go long or 2 bars down then 3rd bar go long or 3 bars up and 2 bars down then next bar long etc etc... and all of this without any S/L or T/P so that they would not interfear with the probability.

The conclusion was after doing this on each and every pair was that there is no consistant logic or prediction element in it.

There were for ex a couple of years that a certain principle worked well on the eurjpy but then it was losing money for severall years. And on the same period that it was making money on the eurjpy it was losing big time on the eurgbp...etc...and did these tests on 1H,4H and daily charts with exel spreadsheets.

Or in other words it gave an other proof to me that markets move in a plain random way.

Friendly regards....iGoR

PS. For indicators counts the same rule, you will never find an indicator that will give you a clear better edge then flipping a coin. How unbelievable this may sound and with all the indicators that people have build over the years.

 

Thank you Igor.

I actually already paid the programmer to add the above mentioned features (trading in the direction of the previous candle, and if on reverse trading in the opposite direction of the previous candle).

But I also asked him to include a setting for days of the week (and it also has hours), as well as a "TP Pips Ahead" input.

This input is something I thought up, and works like this:

When the take profit (TP) exceeds the losses by X amount of pips then the EA shuts down, and does not trade again unitl the next day.

For example, if the input is 10, then the EA will stop trading as soon as the profits exceed the losses by 10 pips.

This means even if the first candle for the day hits 10 pips the EA will shut down. If $10 a pip is used that is $100. And $100 a day consistently 5 days a week is $2000 a month. 20 pips brings it up to $4000 a month and so on.

In backtesting EA's there are often times when the profits exceed the losses, but then the losses take over. Or the TP is too high and more stop losses get hit then take profits, or it is easy to become greedy and try to get as many pips as possible a day, and trade as many currency pairs as possible (I have done it all on demos only to figure out that there are times the profits are ahead of the losses and it would have been nice if the ea stopped there for the day).

When this EA is finished I am thinking of testing EURUSD. There isn't really a need to trade multiple pairs.

I can even use different settings in the ea to trade EURUSD on 4 different charts.

(For example have the reverse option for slow ranging hours on EURUSD, and another setting for the trending hours after London open, but back it up on another chart with a different setting etc.)

For example, if I trade EURUSD near the US close and New Zealand open when there is little movement, and keep the setting on reverse, it could potentially pick up 10 pips easily, and then shut down until the next day. Or I can gamble and aim for more and more pips, but then the losses are in risk of taking over. (Also if a few losses happen first it will take a larger candle or a few candles to get the profits 10 pips ahead, and then force a shutdown until the next trading day).

I need to backtest it first.

With a small deposit ($1000) $10 a pip would be too risky because if the previous candle was bearish, and then a trade is opened on the next candle which ends up shooting up (bullish) 80 pips (since no sl is used) that would be a -$800 loss if $10 a pip was used, and if the setting was not on reverse logic.

I will be testing it live on a micro account with 10 cents pips, and $100 deposit first, but on a demo with a massive deposit I can use $10 a pip or more.

 

Would you mind sharing the EA? if more people can test different times then the better.

I was also thinking of trading within the range of a previous candle.

 
iGoR:
Hi hifromeddie,

I have done a LOT of backtest working with these kind of principle's.

For ex: If 1 bar is up then next bar long or if 2 bars up then next bar long or 3 bars up then 4th bar go long or 2 bars down then 3rd bar go long or 3 bars up and 2 bars down then next bar long etc etc... and all of this without any S/L or T/P so that they would not interfear with the probability.

The conclusion was after doing this on each and every pair was that there is no consistant logic or prediction element in it.

There were for ex a couple of years that a certain principle worked well on the eurjpy but then it was losing money for severall years. And on the same period that it was making money on the eurjpy it was losing big time on the eurgbp...etc...and did these tests on 1H,4H and daily charts with exel spreadsheets.

Or in other words it gave an other proof to me that markets move in a plain random way.

Friendly regards....iGoR

PS. For indicators counts the same rule, you will never find an indicator that will give you a clear better edge then flipping a coin. How unbelievable this may sound and with all the indicators that people have build over the years.

Igor,

Do you not mean the forex is "mostly" random. If it was completely random it would not be possible to make a profit which you clearly can. If it is completely random what would trigger a trade? What do you use to trigger trades or are you relying on money management only?

So are you saying that indicators (including your own) are useless?

Kind Regards

 

sl/tp idea

Hi hifromeddie,

Can you adjust the parameters on the ea to open the order 5pips below or above the open of the "next" candle?

example= after bullish candle appears on 1hr chart, enter the trade 5pips below open

Now you can set tp to.... lets say 10pips

Your open trade will have the rest of the hour to reach only 5pips above the open for a 10pip win.

with this senario, you can apply 10pip SL and have at least a 1:1 risk reward ratio.

I believe this will come close to breaking the 50% hit rate, just by a quick visual backtest, I find that after bullish candles it is fairly likely that the next candle after that will have a 5pip wick at least(1hr chart or above), and vice versa on bearish candles. Howerver, I would not trade this during choppy times since I find after bullish candles you will see "wickless" bear candles during choppy times.........

What you think?

 
spekkiefx:
Igor,

Do you not mean the forex is "mostly" random. If it was completely random it would not be possible to make a profit which you clearly can. If it is completely random what would trigger a trade? What do you use to trigger trades or are you relying on money management only?

So are you saying that indicators (including your own) are useless?

Kind Regards

Hi Spekkiefx,

Not only forex but all markets move absolutly random.

There are already many topics here on this forum that have been opened on this subject but not 1 person could give proof that they are not random.

It is easy for me to say that they are random in the same way as if you can not show a certain God to me then he does not exist in a scientific way and scientist wil say that he does not exist ( as long as something can not be proven then it does not exist).

Be awere that you make the mistake as many many people do and that is confusing the difference between WHAT triggers price to move a certain direction and predicting WHICHdirection price is going to move to.

I had this discussion already many times in my trading room with people who doubt my point of view:

I don't know if you know the stock markets. If a company comes with their quaterly numbers then you don't know how the market is going to react on them. It is possible that a company comes out with very good numbers and still that stock will drop afterwards (because the recent up-moves were already calculated in the price and now the investors take the profit or that it will be very difficult to do better then the numbers they gave etc..). But it is also possible that the price shoots up according the good news.

It is also possible that they come out with very bad numbers and that this makes the stock to go up because people feel that this company had the worst behind them now and that from now on it can only get better...etc..etc....

So news are one of the elements that can trigger a move but that does not mean that you can predict the direction of the move.

Even if someone would come to me with a newsreport 1 hour before the release of that report and says to me that he has inside information I would never bet my farm on it to which direction price is going to move.

And an other mistake many many people make is that thay think that you need to be cappable to predict price with a certainty higher then 50% to be profitable.

That is absolutly not the case. Till 15 years ago 99% of the systems were trend following systems that have between 35-37% hitrate. But the profitable systems had for ex. profit target that is 3 x times bigger then the stoploss ( 35 x 30pips = +950pips and 65 x -10pips = -650pips....total profit is +300pips)

Important with this example is that money managementmakes the difference in being profitable and not predicting the direction of price.

I hope you understand what I try to explain here...

Friendly regards...iGoR

 

guessing game

iGoR:
But it want help you to predict if red or black will turn up more then 50% of the times.

since it is 50/50 chance to guess outcome of next candle, wouldn't it still be profitable if it always guessed just one scenario each time with 1:2 risk/reward? So instead of guess red after red and black after black, just always guess red with sl 10 and tp 20?

I think so, but problem is 1hr tf, 4hr or d1 tf sometimes retrace more than 10 pip before returning, so maybe stop will be hit more than 50% of time. Unless maybe ATR factored in each trade and also pending order to trigger the trade at 10pip below open on a buylimit?

Also, I experiment with similar system that attempts to guess outcome of next candle based upon open on 1hr tf or above. Indicator was 5period bollinger bands set with deviation=1, apply to median price with levels 1.4, -1.4, and 0. rules were to short after red candle if price opened within lower band, long after blue candle if the next open price was within upperband, and reverse if price opens outside the bands (see picture below). Although no real statistics were done on this, I think it put the guessing game above the 50/50 level but the real problem to the strategy was no good way to apply appropriate mm rules, so I would have 3-5 winners and one loss to help me loose, but now am experimenting with buy/sell limits 5pip above/below open which seeming to work very short term demo anyways.

bottom line, I think that ea can be profitable playing the guessing game of next candle, but like Igor say, it will come down to MM and maybe some other type of edge that may increase probability to higher than 50/50

FROM: Kinjaheed

Files:
mt4pic.png  174 kb