My trading method is mainly about finding an entry in the direction of the prevailing trend... So I am entering when I see some hints that the pullback might end...
However I often take entries just to see my trade stopped out a short term later because the pullback hasnt ended... A lot ot times the pullback ends afterwards and price goes in the original direction...
So how do you make sure that a pullback has indeed ended?
I use retracement trendline breaks, candlestick patterns, stochastics in extreme zones... however i am not really satisfied with the results...
Last edited by ikeaboy; August 24th, 2010 at 04:46 PM.
The following 2 users say Thank You to ikeaboy for this post:
You can never know for sure when a retracement will end. You can only use prior information and make some educated guess based on what is available. For example, in a trening market price will move in a stair-step fashion. So you can expect price to retrace and test a previous level. In a downtrend, what was previously support becomes resistance and vice versa. So you can expect price to test this resistance in its was down. You have this stair step pattern when two moving averages open up as in the case of a 21 and 89 EMA. When these two EMA get flat or converge then you can expect a deeper pullback. Some of these pullbacks might take the form of a channel in the opposite direction of the trend. In this case i would tend to select one side of this channel to enter, the side that will give you the lowest risk and greatest reward if the trend resume. You can measure a previous pullback using a line and expect price will retrace about the same distance on the next swing. Usually such price action will produce a clear channel or at least one side of the channel will be clearly defined.
Here are some of these ideas in action (see picture)
The following user says Thank You to trendisyourfriend for this post:
As trendisyourfriend stated, there is no clear cut way to determine if a pullback has completed. There
are only signs. One way is if you use a MA, if price is in an up trend and you want to go long on a pullback,
let price get close to the MA and then look to enter on a bull trend bar. If you trade price action, there is a
good thread on a book written by Al Brooks. He is a price action trader and I try to follow his set ups. If you don't do this already, you might want to try and let the market take you into a trade instead of entering at a specific price. This way you have the market sweep you into the trade. It is not fool proof, but then again nothing is.
The following user says Thank You to tomcav56 for this post:
There will always be a tradeoff between going too early, or too late, and then it will all depend on you tolerance for stops and how far you are willing to run something if you get onto the trend. Al Brooks has some good info on it - there is a thread here.
Alternatively use a Ross Hook method to enter.
Given your frustration with the stops, possibly change your thinking..... work out where you think the pullback is more than a pullback, work out where your stop is first and then work out what a fair entry level (or zone) would be.
Of course you may miss some pullbacks but then you might choose to have a failsafe entry level for those shallow pullbacks.
Unfortunately its all a trade off and always will be.
thanks for your help... it is a tricky topic... i personally think that if you see a horizontal S/R level holding price, it is one of the strongest signs that the retrace might end...
i also want the retrace to be at least 38,2 %, rather 50 % (fib levels) from the last move...
one thing i stopped using are diagonal trendlines... because they are never the same on the different charts... depending on the chart style (volume, tick, minute, range,...) they are different... so i dont use them anymore... i feel most comfortable with horizontal levels.. but still it is hard for me to find entries which are close to the end of the retrace but also make sure that the retrace has indeed ended...
Platform: NinjaTrader (It's a love/hate relationship)
Favorite Futures: CL, TF, 6E
Posts: 176 since May 2010
Thanks: 60 given,
I use volume. Can't sum it up in this thread, but when buyers step in to resume the trend, it usually will leave it's footprint in the volume. I've been using volume for about a year now, and feel like I am only scratching the surface. I can't even imagine trying to trade off a chart without volume now.
One of the important things about volume is not what it is doing, but where it is doing it. I look for volume at points of support and resistance. For example, in the retrace, I would be hesitant to think the pullback has finished if I don't see any significant change in volume at key S&R. I'd expect a response at 38.2 and then a bigger one at 50%. If not, it shows a lack of conviction on the part of sellers or buyers.
The following user says Thank You to Slipknot511 for this post:
i often tried to incorporate volume in my trading because everything sounds so logical and meaningful in the books... but when i apply it on the charts it just doesnt really confirm the stuff from the books... sometimes it does yes... but more often than not it would keep me out of good trades... and to sum it up i would probably better off without it..
nevertheless how do you use it? just the normal volume bars on the chart or OBV or anything else?
btw, do you use it for intraday trading? say on 2-5 minute charts?
(2) Look for a short term oversold or overbought situation (use oscillator such as RSI or Stochastics)
(3) Watch out for a logical point for the price to reverse (pivot, Fib retracement, trend channel)
(4) Wait for price action to confirm the retracement (for a long entry price should be above the high of the prior bar, for a short entry below the low of the prior bar)
The difficult part
Now comes the difficult part. The correction prior to the retracement may have 1 leg or 2 legs, and it could go further than you like. So you need to decide with some discretion, whether your entry setup is valid or whether it will stall.
In some rare case you will get a churn bar showing support or resistance, but usually volume is not very helpful to enter a retracement trade.
So you need to develop a scenario. Some points to look at:
Is the confirmed trend young or mature? Do I still expect a push? Has there been a climax type situation, so that a retest of the high or low can be expected? After three pushes up or down, you can expect a two-legged correction, so you may not want to enter at the first occasion.
Is there any logical target for the correction (measured move or ABC pattern)?
Where is the trendline? Has the trendline already been broken, or will the current correction end above (for a long setup) or below (for a short setup) the trendline?
What is the time of the day? Is it trend continuation or grim reaper time? Is volatility rising or settling down?
For index futures, what does the NYSE Tick show, is is still strong enough to enter long or weak enough to enter short?
The Reward-to-risk or R-Multiple
The retracement trade is based on the reward-to-risk ratio. Some of these trades can go very far and the initial risk is limited to the range of the setup bar of the setup bar plus a few ticks. Even, if half of your entries goes wrong, the setup still should be profitable, if you let your winners run.
The following 4 users say Thank You to Fat Tails for this post:
ikeaboy - check out how Im using Macd to find retracements and LLs / HHs to enter into the prevailing trend . This might help you develop an objective way to quantify when a retracement is turning back towards strength . I find that when I try to pin down something by " knowing " if its over or its started Im left pulling my hair out , and I dont have much hair . Rather , I do better when I recognize when the odds are leaning in my favor and the evidence is strong enough to make the trade . Strong evidence plus trading towards strength is a pretty high odds scenario .