I have been a "breakout" kind of trader forever. I have been wanting to add a "pullback" type strategy to my arsenal for quite some time. I have been thinking about the rules for entry, exit, and initial stop for a while and have come up with some ideas. I just thought I would type them out here to help clarify them in my mind and possibly get some input from anyone here. For entries I have a couple of ideas. One of them is the stalking entry on the primary chart. I was thinking of possibly trailing buy stops along descending bars for a long entry. From the top of the rally (peak bar) I would require two bars which each have lower highs and lower lows than the peak bar. They do not have to be consecutive. After this condition is met, I would start trailing the buy stops. Another entry idea is to drop down in time period charts. My primary chart is the daily chart, and I would drop down to a 15 minute chart to look at the "pullback". The pullback on the daily should appear as a down trend on the 15 minute chart. My entry idea would be to enter on change of trend OR a (down) trend line break on the 15 minute chart. The initial stop I'm going to place just under the prior reaction low (not the one that I am stalking). This of course means that I will always be attempting to buy a higher low. As far as exits, I'm not planning on doing swing trades or scalping. I plan on riding the trend on the primary time period chart from the entry, if it is a profitable trade. So I will just use my standard trend following tactics. I do want to move my stop under the reaction low if one forms around my entry once the position moves past the rally high (peak bar high). Another thought I had was instead of just moving to under the reaction low, I might want to move to breakeven. I don't know if this will interfere with my trend following though. Any thoughts? Banker
In response to your question on the other thread: I don't like entering pullbacks at all (it's a long story), but if I did, I'd want to be very careful that I wasn't entering the wrong way on a reversal. There are several ways of massaging the probabilities here, though you can never really know what the outcome is going to be. One has to do with the strength of the trend. Is it moving or is it half-assed? What's the volume like? Do you show marubozus? Another has to do with the age of the trend. Is it just beginning or has it been running for a while? Another has to do with the swing point. Did it hit R of some sort, either a pivot level or the PDH or the day's average range (if you're trading eminis) or whatever? If so, you're more likely looking at a reversal setup than a pullback. Using TLs can also help, though if you don't want to mess with continually re-drawing them, you can also calculate an MA that will fit whatever trend you're following. This will most likely not be the standard 20 or whatever. It may have to be 12 or 14 or some such. But the MA should look like a TL and serve the same function. It won't serve as S/R, but it will warn you of changes in the trend (which is all a TL is supposed to do anyway).
In response to your question on the other thread: --Thanks for taking the time. I don't like entering pullbacks at all (it's a long story), but if I did, I'd want to be very careful that I wasn't entering the wrong way on a reversal. --I will try to be as careful as possible. There are several ways of massaging the probabilities here, though you can never really know what the outcome is going to be. One has to do with the strength of the trend. Is it moving or is it half-assed? What's the volume like? Do you show marubozus? --I'm not familiar with marubozus. The trend would be of varying strengths. For stocks I'm looking at something similar to what O'Neil would call a base. A nice runup on the daily chart, I look for at least up 100% over the past 52 weeks. Then I look for a period of "basing" The stock stops setting 52 week highs for a number of weeks or months. This usually pulls the intermediate term trend down (left side of base) and then at some point there is a reversal of trend and the right side of the base starts to form. Think Trader Vic 1-2-3. Anyway I plan on waiting for the first higher high in most cases. Stalk the pullback from the high and jump in on a "reversal of pullback" or whatever. I'm not against buying higher up in the "base" or even buying a pullback above the base. I plan on being very careful about volume. I want to see a fair amount on the prior rally and a noticeable dry up on the pullback or reaction that I stalk. I would think the strength of trend will vary. I would guess probably fairly light near the bottom of the base and it should have a good amount of thrust by the time price reaches the prior highs. Another has to do with the age of the trend. Is it just beginning or has it been running for a while? --I plan on coming in fairly early in the trend. Although one never knows for certain if you are near the end. It could be short lived. Another has to do with the swing point. Did it hit R of some sort, either a pivot level or the PDH or the day's average range (if you're trading eminis) or whatever? If so, you're more likely looking at a reversal setup than a pullback. --Good point. If I'm buying inside the base, it will probably have hit some R at the swing point. I'm willing to take some losses if things don't pan out. Using TLs can also help, though if you don't want to mess with continually re-drawing them, you can also calculate an MA that will fit whatever trend you're following. This will most likely not be the standard 20 or whatever. It may have to be 12 or 14 or some such. But the MA should look like a TL and serve the same function. It won't serve as S/R, but it will warn you of changes in the trend (which is all a TL is supposed to do anyway). --I don't mind continually re-drawing trend lines. I prefer that to trying to find the proper moving average for each setup. This is all going to be long trades for now. I'm doing this in my IRA account. I have been trading changes of trend trying to enter and exit on 1-2-3's (down to up). I do want to add pullbacks to my arsenal though. I have a couple other rules which I will be using. I require the 52 week high to be greater than or equal to the highest high over the past two years. I also move my stop to breakeven any time a position trades 20% or more in my favor. I have been using fairly strict money management. I cap each entry at no more than 10% of my account value, and the difference between entry and stop loss has to be 1% or less (account value) for position sizing. That's about it for now. Thanks for your help. Banker
Thanks Wally. I love marubozus. I look for those types of bars especially. I didn't know they had a name. I'll have to study candles some time to learn some more jargon. Banker
Actually, you're probably better off learning less, since so many candles have so little "predictive" power. I like a class of candles including shooting stars, hammers, hanging men, gravestone dojis, blah blah. But they all say essentially the same thing, only in the reverse of each other, and are all fairly predictive. I just call them "spikes". Other than that, marubozus and dojis. That's about it.
So which do you think is better for entry? 1. Trail buy stops along descending highs on daily chart 2. Change of trend entry on 15 minute chart 3. Cross of trend line or moving average on 15 minute chart 4. Something I haven't thought of yet. I want to be able to set orders to enter, and attach the exit order to the entry order (no problem with TWS) while the market is closed in the evening. I'm leaning towards a Teresa Lo style. I might try them all and pick whichever works best over time. Banker
Since I daytrade eminis rather than interdaytrade stocks, I don't know that I can help you much. I spent more time on T's stalking tactic than I care to admit, and I found it relatively useless. I'd rather find the trigger for the resumption and enter there than wind up getting triggered long before the price is actually ready to continue its trend. The only way you can answer this is to look at pullbacks and see what works and when and where and why. You probably don't need to look at more than a few weeks' charts, particularly if you're looking at several stocks. You may find, for example, that the resumption is triggered by touching an MA of some sort, or by reaching the previous swing high or low level (though the point at which the price touches the MA ought to be about the same as the previous swing point if it's really going to move). Or you can look for some sort of divergence as the price nears the reversal point (assuming that it will actually reverse). An oscillator of some kind is good for this.