Open/Close Crossover Method

I haven't thought of a cool name for it, but I will ;-)

Basically, I was studying "The Wave", and noticed something useful.. So I yanked things out, borrowed a few things from everybody on the board, and came up with a system that seems to backtest quite well, very high probability of winners vs losers, intelligent stoploss locations, and fairly high profit. So I'm going to describe the system below, in hopes that someone can backtest it and independently verify for me, that the system works. Also, these types of discussions (hopefully) lead to tweaks and mods for the betterment of everybody on this board.

I've programmed this in VTTrader, if someone wants it, PM me. I am not a MetaTrader user, but hopefully someone else will program it based on the simple instructions below.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The Open/Close CrossOver Method version 1.0 by Twoblink

(This Seems to work best on a 4H chart.. so I'll talk about in the 4H chart for now)

1) Open up a GBP/USD 4H chart.
2) Add EMA(34,Close)
3) Add EMA(34,Open)

Scenerio 1) When the EMA(34,C) crosses below the EMA(34,O) then your trend is DOWN, and you enter 3Units of short positions at the point of crossover. Your stoploss is placed 13 pips ABOVE the EMA(34,O) (Thinking in Tunnel Terms, your stoploss is 13 points on the other side of the tunnel).
When the position is +55 pips profitable, take one unit off. When the position is +89 pips profitable, then take the second unit off. You let the last one ride out until you are signalled out OR you get a signal for reversal of Trend and you manually close it out. REMEMBER to keep moving your stops, as the stop is a trailing stop. So always 13 pips above the EMA(34,O) line.

Scenerio 2) The exact opposite in every way to Scenerio 1. When the EMA(34,C) crosses above the EMA(34,O) then your trend is UP and you enter 3Units of long positions at the point of the crossover. Your stoploss is placed 13 pips BELOW the EMA(34,O) (again, on the other side of the tunnel). When the position is +55 pips profitable, take one unit off. When the positions is +89 pips profitable, take a second unit off. The last one, let it ride out until it hits your stop, or the trend has reversed. Remember to keep moving your stoploss so it's always 13 pips below the EMA(34,O) line.

When I say "+55 pips profitable" I mean when you are +55 pips from the positions you entered, not the total pips of profitability. So if you entered 3 Units, and you were +55 Pips profitable (each), then you'd be at +165 pips in reality.

If you've taken off 2 units, you should be +55 + 89 = +144 pips locked in profit already.

The reason I use the 13 pips stop loss on the other side of the tunnel is, if ou use the EMA(34,O) as stop loss, often times, the price hits it and bounces right off again, in the correct direction.

This works extremely well for a trending market. Small ranging markets don't work as well.

The other thing about this system is, you are ALWAYS in the market, because the market is either up or down, so in concept, almost like the Parabolic SARS, you are always in the market, and the indicator tells you which side to be on for maximized probability of profitability. (Say that 3 times fast!!)

Upon backtesting on the on the GBP/USD4H and the EUR/USD4H, this yields incredible results. So incredible, that I'm thinking (because it's 3am here, and I've been working on tweaking this for the last 5 hours) that it might be the sleep getting to me and so would love to have someone else independently verify my findings for me.

Also, if someone can program this into Expert Advisor in MetaTrader, I'd greatly appreciate it!

I will probably add the semi-hard rule of even though the stop is 13 pips on the other side of the tunnel, I will set a 89 pip stoploss limit, whichever is less.

Also, if I've taken 2 units off, I will probably move the stoploss to the +89 pip point or the original 13 pip on the other side of the tunnel, whichever yields the smaller loss.

It might be harder to spot a trade in real time since the lines are so close together. 13 SL away from the cross sounds great only if you can enter trades exactly where they cross. I'm afraid this won't work cuz the price will go so far away from the cross before you can confirm that it is a real cross.

Hi,

Just picking up on what nitman was saying Is the entry price in your program the crossover price or the bar close price?

Best regards
Alan

Hi nitman,

The crossing point can be made much easier to see with a custom indicator below the chart with a single line showing the difference between the two ma's along with a zero crossover line.

After a solid crossing by price, look for the first pullback to the 34 before entering. Then you can set a smaller stop and consequently put on a larger size.

Dial

nitman, this would be true on paper, but now we have software, so it tells us all ;-)

The "exact" point is probably not that necessary, but indicators can give alerts in almost all software.

firehorse, I'm not exactly sure your question....

Just woke up.. so still a bit groggy. I will post a few screenshots, that will clear my head and yours as well.

Attaching Screenshots, one with detailed explaination, the other just a vanilla version of the same screenshot.

I can basically visually backtest fairly quickly..

I am trying to figure out how to get rid of the rocky transitions, so any suggestions would be great!

It seems that to "protect profit", I might have to move that stoploss to +1 pip after it hits the +55 pip line..

Rough version of the method in VTTrader attached. Take it for a testdrive and let me know!

Thanks.

''blink

This is interesting. Do you have any stats on it yet? You have been busy havn't you? Cool

Scott

This is the first system I've seen that involves two of the same moving average except that one is applied to the open and one is applied to the close. The uniqueness might explain the effectiveness.

I almost want to say that this technique is similar to comparing EMAs that are 1 price bar different. Of course, this isn't true. It's more like zooming in on faster time levels. I want to study this, as I'm pretty sure there are MA systems that closely correspond to this one mathematically. I just need to think about it. I don't care about how to display the stuff in MetaTrader or whatever, just how the math works.

Hey Mike,

We have Bulls and we have Bears.

Assuming bulls, then we can assume that the close is higher than the open.

Assuming bears, we can assume that the close is lower than the open.

If you EMA on the same number of bars, with Open vs Close, then "in theory" you can tell if we are heavy on the bulls, or on the bears.

That's the "theory" anyways. When the "scale" is almost balanced, you are in a choppy market, or if the scale is tipping back and forth in a small range.

I think a small modification I can make to the system is like what Vegas has for his tunnel. If the MA_O, MA_C, and price is all within 20pips or something like that, then hold off.

That will probably be a great filter to add.

Again, any help to improve the system would be greatly appreciated..

Hi TB, interesting thread we have here, thx for the original post and your comments/ideas. I look forward to the developement of the thread as i was captivated a while back by the PA on the 34ema, then found Raghee's site, saw the "Wave" and now i'm hooked. I'm going to add the open to the "Wave" and track things per your suggestions. . . and hope to add 2pips worth to the thread at some time.

Cheers,
Thom

I have chatted with twoblink and he suggested I modify the Ts to trade. I have changed the profit display to show it more easily and added a counter modification along with changing the MA to triangle which removes choppy bad trades. This is a quote from VT author Chris on how to use the counter:

When you add the trading system to the chart there are 3 possibilities:

1) You have no open positions in the market.

- In this case, leave both OB and OS settings at 0 and start the trading system in auto-trade mode; do not alter these settings while it's running.

2) You have a long position open in the market.

- In this case, you would set the OB variable to 1 and leave the OS variable at 0 and then start the trading system in auto-trade mode; do not alter the trading system while it's running. This tells the trading system that you had a buy position already open when the trading system was started; this means you don't have to exit your existing position before starting the trading system to ensure it's in sync with the current open position.

3) You have a short position open in the market.

- In this case, you would leave the OB variable at 0 and set the OS variable to 1 and then start the trading system in auto-trade mode; do not alter the trading system while it's running. This tells the trading system that you had a sell position already open when the trading system was started; this means you don't have to exit your existing position before starting the trading system to ensure it's in sync with the current open position.

It is currently set to trade on Crossup and Crossdown. The trailing stop doesn't work yet but is merely on the chart for reference. It is set for 5 minute trading and does very well on all charts at the moment. Either there are big profits during trends or breakeven on choppy trading. Trading 5 min charts gives less risk and stops traded at close of the bar (when available) will kick in sooner. It needs further work but it should trade properly. It is a simple system and hopefully can be ported to MT4. Thanks to twoblink for his excellent TS.

Hi Rich,

Great to chat with you, will take a look at your changes and play around with it a bit.

The concept of using the triangular MA to counter the choppiness is a new idea to me, so I will study it further.

Also, I don't trade on the 5M, but will take a peek at it. I will also take a look at VT to see how it can automagically move the trailing stop.

SOMEONE has got to be able to move this to EA on MT fairly quickly...

http://www.forexfactory.com/forexfor...achmentid=5692

Hi Rich,

From this pic, I see that your graph confirms what I first thought. I see 2 (perhaps 3) touches to the MA lines, and then a bounce back. That's why I thought scooting the stop line to the other side of the tunnel was a good idea and prevented the MA touch and bounces.

I will officially name this method:

OCC (Open/Close/Crossover) for a lack of a better name at the moment. It's shorter to type :-)

Here is a new version of Open Close Cross that includes stops, a timer on/off, counter, and limit. Trading is still based on MA open close cross. To use on JPY pairs profit needs to be divided by 100. I prefer short times and twoblink likes longer ones but both can be done by changing Periods setting. Remember that trades including stops are made at close of the bar so stops on a 4hr chart may take 4 hours to kick in. More work will be done to filter out bad trades so it is a work in progress.

Profit display has been fixed to include stops

Hi Rich


Thanks for the code...

I have apaperbased version of StrategyRunner which I'm demoing ...

How do I get your strategy to be available in the "Add Strategy" window?

Thanks
BAKROB99

I haven't used Strategy Runner but I see its for futures trading. VTrader is for Forex only and has a built in simulator thats free and is the best way to run this system.

Hi Rich,

This OCC system seems so be quite interesting.I am having some problems with the setting though,so i would greatly appreciated if you could share your short term setting for the system.

Thanks,

Sebastian

This system is both very simple and complex with stops, limits, counter, and timer. Counter details are in post #12 and all options are off by default. Only the periods now is adjustable, and the options are numerous so you will have to experiment for yourself. VT crashed at 830am Friday during major news which is common, so the day's charts are flawed. Any positions that are open then will have to be closed manually after stopping the system and restarting it. It currently has problems with reversals like 830am Friday and only trailing stops will fix that. Twoblink has a different trailing stop that seems better and may be changed in the TS. It is still being tested and modified.

I have included my XMA TS here that uses a similar system that works good on GBP US 10min charts with an initial stop of 30 so each trade will only lose 30 pips. Again testing is needed to see your best settings and may be modified at times.