for es example 1. use a 1 point stop and reverse but do not trail. 2. When market finally moves away from your stop, then trail a very very wide stop. 3. When wide stop is triggered go back to step 1. *Inventors notes: This is an original system I just stole. Optimiztion and curve fitting for the wide trailing stop shouldn't adversely affect the integrity of the system as long as you stick to the basic line in the sand principal.
*More inventors notes. I've traded step one of this system using mkt orders and was able to keep the average loss at under 1.5. So after you test, add 1 tick per trade for a total of 1.5 per whipsaw (That is standard procedure but it is fun to write stuff like I am a scientist.) possible guidlines for wide stop would be n times number of whipsaws resulting from step one. For example if n=1.5 and you got whipsawed 6 times you'd be $450 in the hole so you could set your trailing stop at a minimum of 9 points. Following that strategy would give you wider stops in a choppy mkt and tighter stops in a trend. Is that what you want? You could do it backwards with some kind of countdown system. Say no whip=ts 20 1 whip=ts19 2 whip=ts18 etc. That gives you wide stops in a trend and tight stops in a chop which is what I like personally. And then there is a lot of stuff availabe using percentages to set trailing stops. (not to mention atr) So really, the only thing different is betting that the tight line in sand stop will not get whipsawed enough to underperform just a straight fixed trailing stop and reverse system, and also that the tight line stop will enable you to use a much wider trailing stop than you would with a straight trailing stop and reverse.
See, the idea is to quit complaining about whipsaws in always in trend systems and to just turn around and incorporate it into the system. Instead of saying, I hope I don't get whipsawed You are now saying, I wonder how many times I will get whipsawed.
Plenty! That is why even if you can get this to backtest at 75% return, it will be more like 25% because you will need to allocate a lot of margin money. You could do away with the secondary whipsaws from step 2 by just assigning a profit target. For instance 20 times the number of whipsaws. Or even 2 times the whip. Or some set percentage of your equity. This is just an entry system with no setup, but in some strange way it is even more random than a coin toss. Now all you have to do is optimize your exit. And somewhere someplace, there should be a study comparing all the various exits, because that is the brickwall. It is apparently very easy to come up with a system that will eventually show you a profit, if not overall, at least on one trade. But then what do you do when you are sitting at your computer staring at an unrealized profit? Even doing nothing is only a temporary solution.
I'll tell you what I do when I am trying to trade system and I have an unrealized profit. I say, "System or no system, this market is turning and I'll be danged if I am gonna sit around and wait for the system to figure it out." What I need on my computer is one of those dollar bill feeders so I can say, "Here computer, feel this! We aint just trading paper anymore. I want to see some nervousness and anxiety from you so I know you're serious!"
Profitseer, Rather than have a trailing stop, maybe its better to exit into strength and draw a new line in the sand at a location when the market is likely to counter trend. You could identify this point by a daily RSI figure, or something similar. Each time the market becomes stretched, relative to normal, you take profits. Runningbear
Although not addressed to me I'd like to comment - The idea of a price being stretched from the norm is a good one. Even in an Oliver Velez book from Pristine.com he comments on a price that "seems" to be extended beyond the 20 sma. Another contributor several weeks ago was looking for a way to consistently quantify that distention. STD DEV was talked about. I mentioned just use a ruler. Simple, but darn if it didn't seem to be true for him that when a stock was about an inch from the ma is when he wanted to exit. If only to mechanize, and become the Corona commercial!
Inandlong, I know Profitseer, stole the line in the sand from you but did you get it from anywhere else? Or was it one of your own? Runningbear
Runningbear, Profitseer applied the lovely title. It was my response to a thread two or three weeks ago. Believe it or not, it might have been a thread started by sunnyskies, my buddy, asking about a system that really worked. I don't really remember for sure. Anyway, I usually trade using the closing price of the daily chart. I want to capture the bigger move. In a discussion with my dad, he being ever the doubter, I said that using the closing price on a daily chart, one could simply decide to be long above it, and short below it. On a closing basis. If you look at any of the blue chip nyse type stox, vs the nasdaq once at 250 now at 2.50 types, you will see that overall, you are in a pretty good profit position at some time. And I mean more than a few points. Look at stocks like mmm, ibm, ge, gm, mo etc. If you just use the closing price on any chart, the longer time frame the better in my opinion, and say from here I am going to be short below it and long above it, you will at some time be in a nice profit position. As sunnyskies so sweetly points out, the exit is the tough part. Indeed, you, me, profitseer, and anyone who has ever written a book about trading has a tough time deciding when to get out. So as an experiment, since I presume you will be here for awhile, pick some stox and draw a horizontal line at today's closing price. Tomoorow when it closes, you are papertrade long or short. And sit back and watch for yourself what happens. For quicker gratification, americans in the 21st century that we are, use the hourly chart, or the 30 minute. I'm sure it works on any chart. I don't imagine the cheap stox will give you the feed back you want. For fun, I've thought recently - actually after reading your posts and profit's posts, that drawing the line just above or just below an area of current consolidation would be neat. Either way. Hey how about this. I'll pick 4 stox right now, UNP, PG, GM, and WMT. You pick 4 too. I will create a screen with my 4 daily charts of these stox and draw a horizontal line at today's closing price. And let's just watch and see. Right now we seem to be in a sideways market so this will be good. If you pick 4 and want to share them I'll watch them too. Anyway, there are my five. Let's see how it works out. Along the way, you and me and profit and anyone else can decide when we should get out. But tomorrow's close will decide that we are in. The beauty of daily charts and longer using the close only is, we only have to check it once a day. I hope you play.