daily range Strategy

Mark.....

What would the stop loss be on this new grid system? You mentioned that we would place a trade in the other direction when we are stopped out, but I didn't catch what the stop loss would be.

I've played with grids quite a bit with only so-so results.

The reason I find the daily range methods interesting is because we have used some historical performance figures on which to base our "trigger" points.

Usually on the GBPUSD, if the price moves 50 or 55 points from the opening, it will not come all the way back. I think that Vegas quotes that around 70% of the time, for that pair, that is true. So that is where we start. Somebody here experimented with 60 or 61 also and found it workable. What we need to do is find the best range for the TP in order to smooth out our equity curve.

We could indeed straddle "breakouts" from anyplace on the charts, but I think for them to succeed more often than fail, they should be triggered using some historical patterns to gain any kind of edge.

I have the past few weeks had remarkable success using Micardo's trigger of 50 either way from the FXDD opening price. To a lesser extent with the Kamanda method of a smaller trigger on shorts.

I don't plan to trade on Thursday or Friday this week due to the NFP on Friday. Don't expect much movement on Thursday and the wrong kind of movement on Friday!

Well the new idea would have to be worked out between each individual person. When looking for such massive movements on a chart, it probably wouldn't be wise to use a 30-40 SL unless we're looking for only 30-50 TP. A larger 100 SL may be needed.

I haven't played with any numbers yet. Should the straddle be only 150? Should we only take 30-50 TP? 100? I think a 200 straddle with 100 TP is asking for too much movement. If you want 100 TP then lower the straddle, but then you risk getting stuck in a ranging market.

You're right there is the flaw that it doesn't necessarily use historical data that is easy to play with. It is going to be a game of guess and check, or manually calculating the amplitude in a ranging market and the amiplitude of a trending market.

But I can say the daily range strategy is also inherently flawed too. Its a price action system that relies much on time. If price gradually rises, the point to base the straddle off of moves with each time, and the overall movement is missed when no triggers are hit. It's also possible for the system to enter trades counter to the direction of the trend.

I'm simply basing this theory off the idea that for price to move so much it must be trending in that direction and therefore more likely to give us pips. It's filtering out the noise just like the original daily strategy.

Only problem I'm really seeing is the lack of trades made with a 200 straddle. From 7/12-9/11 it would have only made 8 trades. 6 winners and 2 losers. This was a 200 straddle, 50 TP, 100 SL. I doubled up on the trades following losers to bring the total pips to 200 for the 2 months. 100 pips a month. Not to impressive, but oh well.

The point though, is that perhaps we don't need a time frame on the chart. In fact I think that may hurt us. What if the market ranges and the opens for several candles are very close together? You're stuck in the range. If you abandon time frames and using opening prices for straddle points, you can use random points which could inherently move your straddle outside of a range, or at least be able to profit from it.

This just something to be discussed. Who's mark? Matt

Thanks for your well thought reply, Matt!

I think Mark is on another breakout thread..... I'm getting too old for this

Yes it is possible to get "stuck" in ranges. But then I think that's why we call it a "breakout" strategy rather than a grid. I think that is why it is also a good idea to let the trades, when correct, run for more than 30 pips. We could have several losers in ranging days, but then when you catch a couple of 100 or 150 pip runners that all goes away!

Just thinking out loud. Thanks for your continuing thoughts. It's good to bounce these around I think.

How do we let winners run though? When do we know they'll be winners? How long do we let them run? I don't think a system as simple as this one can answer these questions. In my line of thinking it may be evolving that way though.

I'm trying to think of different ways of implementing the system. What if we could use hindsight to our advantage? How would it work? With a large 200 pip straddle it's possible. Rather than taking random entry points, what if we waited for a peak or valley to form on the chart, then base our entry point off of that high or low? Most likely, on an intraday chart, it is easy to see a peak or valley before it moves 200 pips. But then, how do we judge if a trend will continue for 700 pips and we can take advantage of it?

It might require that we take out profits from playing the peak/valley observation idea, PVO method (hehe), as well as the original idea of placing a 200 pip straddle towards the direction of the current trend. If the trend doesn't reverse you'll enter another trade in the same direction. If it reverses, which it most likely will, you look for the peak/valley, PVO, to base a new 200 pip straddle. Just more thoughts to throw in the game.

I'm sure this could be implemented on many different time frames with varying SL, TP, and trigger levels. Matt

I think taking note of the strong trends is a very valid way of playing breakouts. With the peaks and valleys you would be setting up a sort of trend line system. Then, we could be back to trading the "breakout" of the trendlines. And there we are back to price action trading.

You know Matt, I think we are getting very close here to good things.

Ha. You're getting that feeling too? Just today I was taking a shower and thinking, man I think we're getting really close to a system that is consistent, profitable, and easy to manage. We're just one step away from something nice and easy to trade. I hate not being at that next step.

In regards to your two lot trading, it could be possible. Since we're looking for the breakouts or trends then if we're already in a trade it could provide us with the extra pips more often then not. I come back to the idea though that if we could find the efficient point to have all trades exit it would be more profitable than a system of one lot exiting early and one exiting late. The system pretty much guarantees 10 pips when looking for the 200 pip move. Does that mean we should only exit after 10 pips? Well, maybe if you place 50% of your account into the one trade to make up in dollars what you dont make in pip quantity. I wouldn't recommend it just for sake of mental stability.

This could be one of the options traders have to decide between. I think I like the one lot system just so I'm in and out of the trade looking for the next set up. If we look for 50 pips we'll say the system is 70% correct. (Not sure of that figure) Looking for 60 pips is 60%, 70 pips 50%. I'd rather take the 50 pips and always stay at 70% than introduce more risk into the system. If you create a two part position then you're averaging the risk. It would be just as effective as playing a single exit point for all lots that was equal to the average risk of looking for maybe 30 pips on the first lot and 80 pips on the second. Thats a decision that has to be made by each trader though.

Anyways, evolving into a price action system with trendlines. Is that so bad? Are we discussing something new and different or should we just abandon the idea of looking for pip movements and just play a trendline and S/R chart system? How are we different? I don't know yet. Matt

matt,

I have started coding the weekly idea you had, we had a major change at work, so I lost some extra time to work on it. But now you are not so certain of it. Should I stop trying?

Second, on your idea of a 200 wait or whatever and using any time in general, to me it would just be better to buy immediately and hope it goes your way. If it doesn't average out until you can get out. I just can't see how waiting that long for only 50 pips in the long run can be worth it without a starting point, BUT the wild idea sometimes ends up being the best, so I won't count it out yet.

Regarding the daily strategy, the 60 wait is a good start, but turns out 2004 and 2005 and Jan - April of 2006 it was not good overall. I may have some good numbers though for all years.

llambert -

Keep on trying. Always forward test though. I think what my results had found is that it probably does need to be an evolving system of changing the efficient levels for the trigger, sl, and tp. I'm trading for the long term. I don't want the system to work for a year and then go sour on me. So when backtesting a system using over a decades worth of data it showed that there were good years and there were bad years where it would only average out to 30 pips a month or so. This was using one SL,TP, and trigger over all the years. Your results may vary.

So if you are going to use the weekly idea then you should probably always be shifting the numbers over time. Only use a year or two worth of data and update the numbers week by week. Unfortunately that means you're only using 52 - 104 candles worth of data. Not sure how reliable it would be. The long time frame might be enough to give it reliability though.

Anyways, I am probably going to abandon it for the reasons I'm discussing now. I can do the same thing, irregardless of a time frame, correctly pick areas of S/R and probably increase my success rate.

You say average out till you can get out. Do you mean using the martingale method to add to your losing position since eventually the price will retrace? Sure it works so long as your bank roll is huge. The other idea is like I said of immediately entering a reverse position once you are stopped out. I'd rather double up on the next trade though. Martingale method involves placing trades to fight the current trend your already losing against. I'd rather get out, reset the system, then find the trend more accurately again. See the difference?

The idea of waiting for the 200 pips though is to prevent you from being placed in a ranging or reversing market. If you institute an always in system then there is just as much a chance of the play turning on you once you enter the reverse position. When you are out of the market and wait for it to absolutely move in one direction then you reduce your risk in that manner tremendously.

Sure, maybe its only 50 pips after a 200 pip move. (BTW I'm thinking a 150 pip move might be sufficient. Need to play with it.) I have asked a couple times, but received no answer. How many pips does one need to be considered trading a successful system? I've read threads about "10 pips a day to success" and other such ideas. Trading the simple system of doubling up after losing trades and waiting for 200 pips to take 50 produced 100 pips a month from 7/12-9/11. This is far from averaging 10 pips a day, but its a start.

I'll post a chart later to show how my new ideas would be traded as well as a year or two worth of manual 2 hour chart backtesting. How much will using hindsight to find peak/valleys work? Should we look for 100 TP? 50 TP? I'd love to be able to say taking 100 TP is consistent, because if we can do that and double up after the losers then we're trading it such that our account is as if the last trade was also successful. With a 50 TP 100SL we'd need to triple up after a losing trade to make it seem as if the losing trade actually profited. Up to the trader and their bank account really.

The problem that may arise from the daily/weekly strategy is that its being curve fit. If its constantly updated, can it last? I'm not sure. I'm only leaning in this direction because there is no need for excel spreadsheets to find equilibrium points. It's simple add/subtract from current price levels. Monitor the charts to find peaks/valleys. And with less work it could actually prove more effective since it finds support/resistance which are disregarded in the daily and weekly range strategies. I'm trying to play the same system, but simpler and more effective. Isn't that what we're looking for? Matt

My point I was trying to make about averaging a trade going against you was it would seem simpler and just as successfull as waiting 200 for 50 using ANY GIVEN point in time. In other words to me it just doesn't seem logical. If this were a weekly strategy with a beginning time to use I wouldn't have a problem with it. I was actually just comparing the two.

I want >5000 pips a year, so >400 a month would do - for me anyway.

Yes it seems doubling up should help in my other program it does anyway. Most lots open at one time since 2003 has been 6, meaning the 6th time you'd have to purchase 32 lots to get rid of the losses. Need at least 5k in a mini account to accomplish this. I think probably 10k in a standard.

If we're talking about the same thing there is a difference. If you're averaging the trade, i.e. using the martingale method, then you've lost the original trade and want to add to that position to bring your average price closer to the current price such that on a retrace you're likely to go in profit. Thats up to you to do. Every trader should determine what exit strategies and methods they're comfortable with and can use with their bank account.

The problem is needing large bankrolls if it doesn't retrace. I also don't like the idea that you see that it's a losing trade yet the strategy involves adding more positions to predict a retracement. You're fighting the trend. I'd rather get out of the market to allow a new chart setup to materialize so that I can try to position myself in the trend again. If I know my trades are 70% accurate I'd rather bet on the next trade to double up since my system has no accuracy at predicting where the market will turn.

It might be sooner than you think at waiting for 200 pips too. Say you waited 200 pips from 1.8800 and entered long at 1.9000 to exit at 1.9050. It hits a 100 SL at 1.8900. Most instances involve the turn in momentum to take place over time. There's a flat spot at the top of the chart showing resistance. So you enter a new order to enter a trade 200 pips from the high of that run. Most likely the original trade didn't turn against you as soon as it hit 1.9000. We'll say it ran to 1.9025. So if your trade gets stopped out you know the peak of that run was 1.9025 and to enter a sell order at 1.8825. When you get stopped out on the original trade price would have had to hit 1.8900 so you're only 75 pips away from entering the next trade to double up on. That shouldn't take very long to trigger.

Yes this system does work for any price on a chart. Set a price, wait 200 pips, enter and TP 50. This is what the new posts to this thread are discussing right now. I first introduced the idea that waiting for new candles and resetting the system isn't necessary and could actually hurt this system. I'm trying to present the idea instead that we should take a simple look at a chart to use the highs and lows of the peak/valleys as our points to straddle. How many ranging markets or bad trades resulting from hitting S/R can we eliminate by not basing our positions on a time frame and more on the actual price of the market?

You might have to work with the charts to accomplish 400 pips per month. That can be quite a lot of pips. Looking for more pips introduces more risk into the system. If I can get 150-250 pips a month to compound I'll be happy. Eventually by trading more lots I'll be able to get the same profit at 50 pips as I would trading a single lot for 100. That's up to you, again, to decide.

I'll end by noting the idea of using the highs/lows of runs has a built in safety device. Most likely prices don't stall out on 1.9000 or 1.8400; even numbers. This means when we place our orders off of them it could automatically protect us from placing orders and SL's at levels that could get stop hunted. Whats 200 pips from 1.8847? 1.8647. Whats 150 pips from 1.8847? 1.8697. This gives us a little more peace of mind knowing it makes it easy for us to avoid being part of a typical trades mindset and entering trades at round numbers like 1.8650 or 1.8700. Matt

Here is an EA I had programmed. I have not done much with it yet and did not program it. In the distance field put the distance which is where the limit orders will be placed. When one is hit, the other is cancelled. The cancel at the end of day field is for daily charts so it will start over. Good luck. Matt

Ok doke. I'm going to attach a picture below of how my new idea would play out.

Since I was looking at 2 months of data, 7/11-9/11, I had to compress the chart. Therefore I couldn't put text on it. I'll explain by arrow colors.

First yellow arrow - on this chart this would be the first level to base a trigger off of. Since this is a peak it means we base sell trigger off of this number. That would be 1.8476-200 = 1.8247 which is marked by the second yellow arrow and a TP 50 is marked by the bottom yellow arrow. If we played the trade perfectly we could have banked 92 pips if we let the trade run. (not worth it IMO compared to the potential los on other trades.)

Green arrow marks the new low. This is where we base our buy trigger 200 pips from. 2nd arrow is the trigger, 3rd is the TP.

As you can see the price continued upwards for another 150-200 pips. I've been trying to think of ways to trade continuing trends. Right now I'm looking at looking for the retrace to see where the new base forms. Use that as the new low to base the buy trigger off of. In this instance that point is marked by the white arrow at 1.8414. So the trigger is 1.8616. That continued uptrend did not reach that point so no trigger was hit and I would have cancelled it.

Red arrow designates new high formed in that trend at 1.8597. Sell trigger is then shown by second red arrow at 1.8397. This trade would have ended in loss.

That same candle to trigger the losing trade would later be our point to base our long trigger off of. The point was 1.8383 shown by turqoise arrow. Next turqoise arrow is the trigger and then the next is TP 50 point.

The high off of that trend did not go down 200 pips to trigger anything. So we find the low to continue the trend. This is shown by the yellow arrow at 1.8543. Trigger at 1.8743 and TP at 1.8793.

Uptrend continues, no sell triggers. Look for new low to base buy trigger. This point marked by green arrow at 1.8699. Trigger at 1.8899 with TP at 8949.

I would not have made a new trade since the upmove was in only one candle and would be too volitile to judge. There was no new low to base a buy trigger off of.

As time passes by on the chart you will place your triggers off of different points and change it accordingly. So the candle after the big run up formed a high for several candles. Then later a higher high was formed and that would be the new point to base the trigger off of.

It was a high of 1.9144 designated by the white arrow. Trigger is 1.8944 with TP at 1.8894.

This is where things get rough for the system. The market ranges drastically. It stays out for most of the time and then finally the moves are so violent that it allows a trigger to be hit. We base a long trigger off the low designated by the red arrow. Price is 1.8775, trigger is 1.8975 and TP 1.9025. As you can see the market hits our trigger and then reverses hard. Hindsight shows the market will range for quite a bit longer. So there is the 2nd losing trade.

The next trade designated by the turqoise would have been a tough one that could have gone either way. The base was 1.8828, trigger 1.9028 and TP at 1.9078. The market went to 1.9027 and then dropped 131 pips. We would have narrowly missed a loss by 1 pip. This could have gone either way really. We lucked out in this situation. That would have been 2 losing trades in a row. This is why I say doubling up on trades may not be for everyone. To break even on the last 2 trades you'd need to put four times your usual size into the next play since you had doubled up on the last one as well.

Anyways, next trade is profitable anyways. This is why I say I like the idea of doubling up on the next trade. If the system is very successful then I will trust eventually things will work out. It may be rough if you're unwisely using large lot sizes to play with for every trade. If you only use 2% per trade you would of had to use 4%, 8%, 16%, 32%, and 64% depending on if you lose that many in a row. I think this system is very unlikely to hit 5 trades in a row that lose.

The next trade is shown by the yellow arrows. High is 1.9091. Trigger at 1.8891 and TP at 1.8841. SUCCESS!

Last trade for this time period is the last set of green arrows. High of 1.8869 for a sell trigger at 1.8669 and TP at 1.8619. Another successful trade. The last white arrow is just showing the current point where we're basing our long trigger off of. Its the low of that valley.

I like using this time period to show how survivable this system could be. We had a long period of ranging. This is a 2 hour chart. It ranged in the middle there for 18 days! That's a long time to be stuck in the same 150-200 pips.

So lets look at the results. 9 winning trades versus 2 losing trades. For 2 months of time that equates to 350 pips profit or 175 pips a month. If you had tripled up on losers to breakeven on the loss as well as make 50 profit then it would have been like having 11 winning trades of 550 pips or 275 pips per month. The problem is if you were to get unlucky and had lost those two trades in a row where we luckily missed out by 1 pip. In that case you would have to triple up on your triple up to make those trades seem like profit, or 9 times your typical entry size to hope the third trade wins.

I hope I've explained all of this clearly to everyone. I'll be backtesting further, but wanted to get a chart on here for people to see. Time to discuss.

http://img295.imageshack.us/img295/8574/200piphh0.png

Just backtested 5/12 - 7/12 and it produced the following results.

13 wins with 2 losses. Both losses could have been nullified by doubling up on the next trade. So 550 pips profit for the 2 months. 275 a month. I'd take that. The beginning of the 5/12-7/12 chart was rough though. It ranged tremendously in a downward channel. The trades profited, but only by a few pips. Makes me wonder how much peace of mind it would give us to look for only 45 pips. It decreases our profits by 10% though. Up to each individual trader to determine if thats what they're willing to do.

Even if you didn't double up on trades after losses then you would have still made 450 pips or 225 pips a month. Not bad. Might be something to start out on just to avoid the risks involved with doubling down.

Since this is not an "always in" system it could be adapted to trade other pairs as well that could bring in profit. It literally takes a 1 minute look at a chart to set up all the trades. Matt

P.S. - I've already edited this post 3 times to add more and more to it. Just thought all of you should know.

Well, maybe the trigger could be at the 150 pip point, or 170 just to give more room for the 50 TP.

Are you using 100 for the SL?

I dont think the trigger should be moved to 150. Imagine how bad the system would have fallen apart during the extreme ranging in the chart I posted.

TP was 50 and yes SL was 100 for those numbers. SL of 50 would get stopped out too often. I was hoping a lot of the trades would be able to be moved to TP 100 such that when you double up after a losing trade it makes it seem like the loser was actually a winner too. Asking for 100 is too much though apparently.

I think I would say the 200 trigger filters out enough of the ranging market, but only leaves us with so much space to take profit. I'm happy with the results so far though. Matt

August is the classic ranging (or stagnate) month. Apparently a lot of Europeans take vacations then. It was great for Firebird (a ranging EA) and pitiful for trending systems in general.

That makes me feel better. If this trending strategy can come out on top during ranging months then I feel pretty good about it. What do you think? Am I missing something? Wish I was european to take vacations with them. Matt

I agree. I am surprised it did that well in August. I am encouraged.

By the way, I think your calc on profit was not quite right for the 7/11 to 9/11 period. You said you had 9 wins (50 pips) and 2 losses (100 ea?). So that would be 450 pips good and 200 bad for a total of 250 pips, or 125 pips per month. Not bad. I would trade for 125 pips per month as long as the drawdown was not too wild!

I should have clarified. The results I give are if you double up after the loser. This is because I trust the system to be much more right than it is wrong. So when you double it up, instead of losing on one trade and winning on the next, -100 +50 = -50, it actually loses on one and wins double on the next, -100 + 100 = 0. So a losing trade just cancels out a winning trade.


11 trades - 9 wins, 2 losses. So the total profit is actually just like having only 7 wins. So 50 X 7 = 350 pips for the two months. 175 pips per month. Sorry I didn't clarify that earlier.

For these months it would have produced 450-200 = 250 or 125 pips a month if you decided to not double up. Not as good, but not terrible for literally looking at a chart for maybe 2 minutes out of a day.

OK. Thanks.

Not bad for 2 or three minutes a day but one always must calculate the risk involved. Also, one must eventually calculate the time spent in endless study and analysis. Although that is the challenge!