 |
Stops and Exits |
Let me start out by saying that there are almost as many ways of using stops as there are trading methods. In my opinion it is the most personal part of your trading system. Each trader needs to determine which method they want to use for stops and still be within their own comfort zone. If you are outside your comfort zone, then emotions are in position to take over and you will lose the focus you need to manage the trade correctly.
You should never enter a trade without having a predetermined place at which you will cut your losses. Yes, you will get stopped out only to watch price take off like you thought it would, but that is part of the game we are playing. If fact, if you do not have the right mindset for taking a loss, it will hinder you from becoming a good trader. Getting upset over it only takes away from you ability to focus on what the market is telling you.
Many traders also feel that getting stopped out is a loss. It really isn’t. In fact, you have won. By being stopped out, you have won most of your money back. Preservation of capital is so important to be able to continue to play this game. You also have the ability to reenter if price does eventually go the way you thought it would. Lastly, you have eliminated the frustration of not being disciplined by getting out when your predefined stop hit. One of the signs of a successful trader is having small losses and not huge ones. Think of that the next time you do not exit immediately and without hesitation when your stop is hit. Losing properly requires skill, discipline and maturity as a trader.
As a trader, you need to ask yourself if you are going to use the same time frame for signals, entries and exits or go to a lower time frame for the entries and /or exits. This is a common practice but hard for beginning traders to do as they tend to focus on the lower time frame instead of waiting for the signal to setup on the main time frame. Once you have learned to do it though, it should improve your results.
One of the most common ways to determine your stop loss is to use the preceding price bar or the price bar you were filled on. Which one is used is nothing more than a matter of traders preference.
Two_Methods_Initial_Stops9.gif
Two_Methods_Initial_Stops_PRT.gif
If you are entering a long above the high of the previous bar you may,
- Use the prior bar stop – place an initial stop 1 or 2 ticks below the previous bar’s low.
- Use the current bar stop – place an initial stop 1 or 2 ticks below the current bar.
Some traders use a combination of both by normally using the prior bar and then if the risk is too high for their rules, instead of passing on the trade, seeing if the current bar method brings the risk back into range for their trading rules.
p_8: Just additional comment. Some say stop is like an insurance. A trade without a stop is a gamble.
cluetrain: jimmer suggests that bar high/low no more "meaningful" than another point for entries....what are your thoughts relative to these highs/lows as stop points?
Whether bar high/low no more meaningful than another point for entries depends entirely on the trader. For Jimmer, this is true. For others it would not be true.
Using this method for when price is in a range:
The stop can either be at the low of the range for long, high of the range if short or if risk too high the H/L of the breakout bar can be used.
Another problem can be created by runaway price creating a long bar when you are in the trade and using the H/L of the bar would be giving a lot back to the market. For this situation, I still use what Mike Bruns taught, which is not to let price get more than 2 pts away from your stop. This assures you of locking in profits while still giving it some room.
Those you use Fibonacci levels might use price targets to exit with. Once again Scorp has posted an excellent chart showing this along with orange lines marking measured moves that show how price moves the same distance after pausing.
This is a chart posted by Scorp. It shows two measured moves (the orange lines) and also shows how Fibonacci projection of 127% worked again today.
QSuzy: buffy .... what do you mean when you say "measured move"?
A measured move is when price moves up or down then pauses either with retracement or sideways consolidation and then takes off again in the same direction. The first move up or down more often than not matches the second move up or down.
Lightfoot500: where is the measured move tool in ensign?
Ensign Software: lightfoot, the example is using the
Ensign Software: Elliott wave tool, which was renamed to be Formations
Ensign Software: on the property window for the tool he has three lines checked
Ensign Software: the B-C leg (2nd leg retracement)
Ensign Software: and the C-D leg which is a measured leg with the parameter set at 100 so it is drawn at 100% of the size of the A-B leg
Ensign Software: you draw three points, A, B, C and the auto measure is the leg from C to D
prakso: I must be blind or something - cant find "draw tools"
Ensign Software: prakso, draw tools is a panel with all the buttons for the various draw tools, one of which is the Formations button
elroch: On long term charts, bar high lows can be swing high lows on short term chart
elroch: Not always, of course
yodu_de: on your first chart, do exit once the stop price is hit or when the 2min candle is complete which close at or beyond the stop level. remember, these candles are changing dynamically.
I personally do not wait for the bar to complete but then I am not trading off of the 2M and if the low of the prior bar is my stop then I am out.
yodu_de: ok. yeah I guess you need to gauge the momentum by watching time and sales
Some traders watch that but many do not.
prakso: what’s a hard stop ??
A stop that is actually entered isn’t of mental. Until you can trust yourself to execute a mental stop without hesitation, I recommend you use hard stops.
elroch: Great to have when the power goes down
Price target can also be done with price patterns. Some examples are HS, triangle, wedge,
http://www.dacharts.com/archives/TrendLines/Head_and_Shoulders/scorp_ES_headshoulders_08-30-02
Shows how to project a target from a head and shoulders price pattern. Many price patterns have their own rules for projecting price targets.
Other traders use stops to determine when to exit a trade. We will mention many of them but I am also sure there are many more out there with new ways being devised all the time.
Some prefer to use trailing stops. In my opinion, for this to work well, you need to find a "smooth" chart. By this I mean where most of the bars in an uptrend have higher highs – HH – and higher lows – HL. This will help prevent you from getting stopped out on small price movements within the trend. Which chart is "smooth" can change during the course of the trading day.
Rice21: One answer to Buffy's point is a volatility based stop of some sort. Has anyone considered coding a stop like Cynthia Kase's DevStop or LeBeau's Chandelier and ATR ratchet? I know Buffy has provided the link to LeBeau's site.
Rice21: Wider with 'unsmooth' charts, tighter with smooth.
Richard_: unsmooth = noisy
brandon_131: Rice will you describe DevStop. chandelier and ATR ratchet
Once you determine this, it is now possible to use the prior bar or two bars ago H/L as a place to put your stop. If the prior bar, you made want to add a tick or two if the trend is strong. You can also use a larger stop in trailing price.
Another technique is to drop to a lower time frame and put a SMA on the highs and another one on the lows. The chart example uses a 5 period SMA on each.
The upper red line which is a 5 SMA of the highs would be used as a stop for a short.
The lower green line which is a 5 SMA of the lows would be used as a stop for a long.
Another method is to use the last swing high for shorts and last swing low for longs.
The following chart gives an example of using last swing highs for a short.
Many charting softwares have a study called Parabolic stops. This chart shows an example.
A big volume spike is reason enough for some to exit as this happens when there are traders taking their profits because they feel the move is over and other traders are buying to play the price movement in the other direction.
For many traders, the candlestick pattern gives them the signal to exit. These reversal patterns can be found at www.litwick.com under glossary
When google can do her discussion on exits she should be able to add more to this list of ways to exit. I for one am looking forward to it.
As you can tell, there are many options and hopefully something that catches your eye has been mentioned here today for you to make your own.
Buffy_04364: now I am sure I didn't think of all the ways
Buffy_04364: does anyone have their favorite stop method they would like to share now?
cluetrain: the make me happy right now or you're GONE" stop
milesov: indicator exits
elroch: One simple one not mentioned was the high low stop, which much of the time is the last swing high/low but tightens on a longer move
elroch: Defined as high or low of last N bars
elroch: It's in Ensign
Ensign Software: elroch mentioned it, and I want to emphasis it that the High/Low stop tool in Ensign which is the highest high or lowest low of the last N bars plus or minus a slack offset is
Ensign Software: one of the most powerful and effective stops to use. Suggest you all give it a thorough investigation
elroch: Thanks for pointing out the feature I didn't mention, Howard
brandon_131: Howard what is "slack offset"
elroch: brandon - (high of last N bars) + offset
elroch: is the stop
brandon_131: I see, thanks el
Buffy_04364: what do you usually use for N bars
elroch: I must admit I have only experimented theoretically with it
Buffy_04364: ok np
cluetrain: prakso, the stop is entered with the initial order and is triggered when the initial order is filled
cluetrain: these add on programs for IB's TWS perform same function
elroch: That's a good feature for peace of mind
p_8: dunnigan suggested 2 type of profit taking stop
p_8: one is exit as soon as price hit last swing extreme
p_8: other is to move stop loss to B/E after price move to your favor and reach 1x StopLoss distance
p_8: then after price move 2x StopLoss distance, use trailing stop by give price room to go down 60% from the high
Buffy_04364: 60% would be too much for some traders - that is why each trader needs to find their own style for stops IMO
p_8: that way you give price room to move in a case there is a big trend
p_8: yes, Buffy, perhaps you do that when trading multiple contract
p_8: and use the 60% for the last contract
Buffy_04364: perfect use for it p_8
NZ_Kiwi: can also wait for a regular divergence to form against your direction
cluetrain: have always preferred the tiny "get back in 3 times" type of stop to the wider "sit there and watch while your stomach churns" type.
Buffy_04364: that is me too cluetrain
Buffy_04364: would rather reenter
cluetrain: yip
p_8: btw, with the 60% stop, you get out as soon as there is a reversal pattern
Richard_: has anyone mentioned the `cover you costs at the first point or two by taking profits on half' ?
elroch: For example on the chart Buffy displayed, exit would be in a very similar place with a 10 bar stop with no offset. With an offset or more bars, the short trade could continue
p_8: nope, Richard, could you elaborate on that?
Richard_: NQoos approach, p8.
Richard_: and a number of others
p_8: I think Joe Ross mentioned something like that too
Richard_: many seem to like it.
p_8: not recommended by Van Tharp
p_8: though
p_8: Van Tharp suggested that you should be in your biggest position when you are right
p_8: but getting out partially is opposite to that idea, you will be at the fewest position when you are right
p_8: I can see Van Tharp point but psychologically, it's hard to do
cluetrain: worth observing woodie to see such cost covering stops in action
elroch: Richard - on IB that would mean setting a limit trade at 0.25 over your entry!
Richard_: it would mean writing a little program to do it for you, elroch :-)
elroch: Unless you add on costs of computer, data feed, good lunches and perks of other types
Richard_: normally it's 1.5 or 2 points on ES
Richard_: then let the rest run.
Buffy_04364: yep many schools of thought
p_8: lol
Buffy_04364: and if you are an all in all out trader not a concern
Buffy_04364: :-)
cluetrain: just produces a different sort of equity curve p8
elroch: Partialling should improve consistency. Adding should improve max profits.
cluetrain: but over time, working on the same "edge", you get to same place I think
Buffy_04364: true elroch but achieves opposite effect if outside comfort zone
elroch: With one contract, choices are simpler.
p_8: it's a good backtesting material if you have tradestation and a programmer
cluetrain: y
prakso: Like Angell puts it "send one Soldier to the battlefield an see how he is going to be treated - and then send another one - maybe _
cluetrain: I am more comfortable with all in/all out approach
Buffy: Jimmer and I do that also cluetrain. Last I knew Teresa Lo also was all in or all out.
p_8: I have a hunch Van Tharp is true, in the long run you make more by not partially getting out, but, not sure, and I for one can't do it.
elroch: What I meant was partialling is out of the question with 1 contract
CJACK4: I am wishing for some kind of toggled alert (I toggle it on after entry and off after exit), set to my specs, that would show me visually, up on the chart, where my exit point is.
elroch: But could use a stochastic approach. Toss a coin to decide to take profits at 2 points or close stop to break even and let it run
NQoos: great.. hi Buffy
cluetrain: that would be cool, CJACK
NQoos: Good trading to all
elroch: Sounds entirely valid
elroch: Hi NQoos - a famous visitor
Richard_: If you're profitable trading one contract then logic dictates that you can probably make 10 times as much with 10 contracts ... reality doesn't always follow logic.
_8: yep, and your mind don't treat it equal either, Richard
Richard_: yep, the mind is not logical
Buffy: It is emotions taking over with the added monetary risk not being perceived as a percentage but as dollar value.
cluetrain: however you do it, you must find a way to get rid of losers fast and try to hang on to winners a bit longer
Lightfoot500: what happens with 10 that doesn’t happen with 1, Richard
cluetrain: sheer panic, lightfoot
Lightfoot500: well maybe lack of discipline?
Lightfoot500: or ego.
Buffy_04364: does that help explain it lightfoot?
Lightfoot500: ya, mind over matter. :-)
p_8: what's your favorite stop loss and profit taking stop, Nqoos?
p_8: that's the topic we are discussing right now
NQoos: 2 pt stop entry 1 off 2
NQoos: get 1 off stop B/E
NQoos: my minimum target 2 pts
NQoos: not target....just will take partial
NQoos: p-8 always have targets.. as trades moves targets can change
p_8: then what do you do with the rest, Nqoos? how do you decide to pull the plug?
NQoos: and I risk more when up
NQoos: and when down I risk more to get back..so I can be risky
NQoos: lol
Not recommended for new traders IMO
p_8: Nqoos, make sense, so, if target is not hit, you get out ?
cluetrain: nqoos, all these playing with the house's money notions don't work for me
NQoos: no at target I am deciding if I can see a farther target..looking at last retrace or measured move to decide if I want to hold longer
NQoos: even if I see signal to go other way..
NQoos: this style with trend
elroch: Really, NQoos - you don't SAR?
NQoos: so let trend work with me
NQoos: yes sar sometimes
elroch: I thought divergence trades were great for SAR
p_8: ok, so if target is not hit and retrace, you let it retrace to last swing low (when long)?
NQoos: usually close one then open new other direction
p_8: as long as trend is still valid
NQoos: today hod passed my target..twice held..
Richard_: helps to have s&r supporting your sar
NQoos: trend not reverse way I see it till 802 area...
elroch: I think it's a mistake to let trades hit your stop when you've lost confidence in them - better to close at a smaller loss
cluetrain: true elroch
Buffy_04364: agree elroch
cluetrain: spotting chop and pushing your chair back away from your order entry button is a crucial skill
For sure Clue.
Buffy_04364: Thanks everyone for coming and hopefully you have some new ideas to play with now
Please let Buffy or Dave know if you have a topic for a future discussion.