Reader question: How to handle exits?

Glenn comes back today with another good question. Here’s his full, unedited question:

Your many hats, your one job, the one and only thing that matters. For what it is worth these most recent posts may be the most important. The more I trade the more I truly understand what you are trying to say. However there is one problem.

If your system has you trading a certain way and you can execute those trades with precision life is great right( huh maybe not) …. So let say you job is to enter a trade and you have clearly defined exits. Great right?  But what happens in trade management is often hard to put into the box. Take a look at this recent breakout trade.

I went short at the blue line for a nice breakout trade…. I have a clear stop loss and a clear take profit when I entered the trade however managing the recent long pullback is hard to put into this system. The rules and or thinking are not so clear, should I move my stop loss to my entry? Should I wait, if I am a swing trader and my job is to take one clear trade out of the market then how much give back is too much give back…. Will it take thousands of trades to develop more of a clear system and or should discretion be used in the interim? I might have an edge and I might be able to exploit that yet managing the trade and or moving stops seems to be my biggest problem currently… 

pastedImage

Ok, so this is a good and deep question, and there is not a simple answer up front. First, let’s acknowledge that trade management is difficult, and let’s think about why that is true. For one thing, there are just more decisions to be made. When you’re not in a trade, you basically have a decision to put the trade on or to do nothing. (Some traders may have the option to put on part of the position.) When you’re in a trade, at every moment, there’s a temptation to micromanage–add, reduce, hedge, tweak, move a stop, move a target, tweak some more… the list goes on.

There’s also the psychological component. Trade entry can be stressful, as traders deal with the possibility of taking a loss, booking a win, or maybe missing the move, but at least you deal with these stresses without having risk on in the market. Having that risk, having your P&L engaged to market movements, adds another depth of emotion and can make many of the trade management issues so much more complicated. As I’ve said before, the market provokes mistakes, and it probably provokes more and more serious mistakes when we have positions on.

Let’s also take a look at a technical issue here: the breakout level. I knew a trader (who taught a lot of other traders) who believed that you would take a breakout trade and then you had to exit if the market came back to that level. The problem with this is that a breakout level is not a “real thing”–the market dgaf about your breakout level. If you doubt this, look at about a hundred example trades where the market deals with a breakout level. Sometimes it will hold just outside the level in textbook form. However, when the market comes back and breaks the level you will likely find that the trades fail or work about the same number of times. In my experience and testing, a breakout level just doesn’t work like most people say it does, or like most teachers teach.

To get us closer to answers, the first thought I would offer is that we need to simplify the options available. Think of the trade in a few “units” (perhaps two to four) and work with those units–if you have 1,000 shares of stock, you won’t be selling 50 here and then buying 25 back and then selling 100–that type of trade management is an invitation to make many mistakes, especially for the developing trader. First, simplify.

Second, decide when you will make decisions. Will you only make decisions and move stops after the market is closed? Is this always true, or are there exceptions? If there are exceptions, then they need to be explicitly written down, otherwise making exceptions is breaking your system and breaking your discipline. To my way of thinking, this is far worse than losing money.

Third, decide if your executions will be done with “resting orders” (note that I use quotes because many traders will prefer to not put actual stops in for many markets) or if you will ever “go to the market”. In other words, if you’re short like Glenn is in this chart, can you only be taken out at your stop price (which could be moved close to the market) or might you use a market-type order (marketable limit, limit on the offer, etc.) to get out? This is closely related to the second point, but needs to be defined clearly.

With those big questions answered–and notice that these questions drive us toward that bigger question of “who are you as a trader and what kind of trader are you?”–we can begin to think about actually answering Glenn’s question. Think of this in two extremes: On one side, we have the nearly completely systematic approach in which a trader will get into the trade, put in a stop and limit order, and just wait and see what happens. If it takes days or weeks for the orders to be hit, he waits. On the other hand, we can adjust the stop frequently (see point #2 above), move in and out with very small pieces (not if you read point #1), and can execute whenever we want (point #3).

There’s no one answer. In general, I personally prefer more of a laissez-faire approach to trade management, but this is very much a reflection of my psychology. I work for a deep comfort with my positions in the market, a clear distinction between the things I am responsible for the things for which I am not responsible, and a healthy respect for the factors outside of my control. This style evolved over many years, and, honestly, I couldn’t have worked like this in my early years as a trader. The questions of how much give back is appropriate, if that is different for a trade that has profit vs. a trade at the entry price, etc.–these questions all need to be answered, but they need to be answered by you and for yourself.

Obviously, this post is targeted toward the discretionary trader. Though some hard-core engineering types (in the comments on my blog 🙂 ) will sometimes argue that it is impossible to be discretionary and systematic, this is a semantic argument at best. Discretion gives us flexibility and vision that we might not be able to capture in systematic trading rules. This can be a good or a bad thing, but, even for a discretionary trader, we want to drive toward being as consistent as possible. Glenn, your answers for this trade example are not just for your NYSE:MUR trade; your answers must apply to all trades of this type and asset class. Asking these questions (and, no, the answers will not take you thousands of trades) is a key part of developing and growing as a trader.

AdamHGrimes

Adam Grimes has over two decades of experience in the industry as a trader, analyst and system developer. The author of a best-selling trading book, he has traded for his own account, for a top prop firm, and spent several years at the New York Mercantile Exchange. He focuses on the intersection of quantitative analysis and discretionary trading, and has a talent for teaching and helping traders find their own way in the market.

This Post Has 6 Comments

  1. Matt Haines

    Personally, I don’t think you should ever change your exit plan AFTER you’ve entered the trade. You’ve spent all that time backtesting and verifying with out-of-sample data to make sure your system works. So why then would you change it in the middle of a trade?

    When I take a win or a loss and conform to my system, I feel I’ve been successful (although it can be a painful sort of success when it’s a big loss). If I break my system and exit at a different moment than was pre-planned, it feels wrong. I just basically pulled a random stunt.

    I have one mean-reversion system that exits at a close > 6%, with a hard stop loss of 30%. It has a high hit ratio and can be very satisfying. But it’s very hard to watch the stock fall 30% (which doesn’t happen that often), and dreading that trade-confirmation email. But if I bailed early, then I’m just trading randomly and can no longer have confidence that my system has any evidence of success.

    1. Erik

      Hi Matt,

      So does that mean that all your exits are at -1R? Say a trader never moves the stop loss and always takes losses at -1R, but scales out half his position at +1R, and the other half at +2R, wouldn’t that negatively impact his overall results? Unless of course the trader is an all in-all out type of individual.

      1. Matt Haines

        Hi Erik. I’m not sure what “R” is in this instance. I didn’t go into the complete details about my exit (it has a time limit of five days, and the entry point is critical), and there’s no scaling out. It’s a relatively short-term MR system. The details aren’t important though.

        My point is that it’s been tested, and these are values I stick by no matter what.

        There may in reality be some discretionary “wiggle room” in those values that could theoretically improve the results. But if I go beyond my tested system and start improvising, I can never feel good about what I’m doing. If I stick with the system, at least when I hit a loss (even an occasional big one) I can say “well, that’s to be expected in this system.”

        1. Erik

          Hi Matt,

          R in this case would be the Risk (amount of money you are willing to lose on this trade). Judging by your answer, it just goes to show how important it is to know what kind of a trader you are! I’m personally a swing trader so I scale out of my positions at R multiples. It happens quite often that I take 1R losses, but, I aggressively move my stop as the trade goes in my favor or if it doesn’t do what I anticipate it should do. Sometimes this works, other times it doesn’t, but that’s part of being a discretionary trader I suppose. I do agree with the fact that there should be some method to the madness so you’re not reacting to every little move, but are reacting according to the trade management rules you’ve set out for yourself.

  2. Edward Jenner

    Well, I can see the setup, but entering the trade I would think about where are the most likely places to have problems. In my mind you have the 200dma just below and you are close to the 20period lower Bolinger Band (Of course what you look at depends on your system, that’s just something I look at). So anyway, my exit plan would be something like ‘if it finds support at the 200dma or the lower band’ then exit. Price went through he 200dma on the 1st and then down to 24.50 on the 2nd – looking good. However the lower 20day band was at 24.90 on the 2nd and rather than closing near that price, the lower price was strongly rejected. While I don’t trade candle patterns as such, this would be a warning to watch it closely the next day. At this stage, once it regains the 200dma for any length of time, and certainly a close above, I would be out.

    Also, perhaps much more importantly, this is an oil stock, so you are basically trading oil (or at least it can have a large influence). On the 2nd oil hit an important potential resistance level (look at oil prices, not USO) at the low after the mid-east summit on April 18th, so that would be an area to make a decision.

    So one question, is what do you think oil will do? It’s been moving, so whatever it does may have more impact on your stock than any technicals. Do you want to use technical analysis on a stock driven by a commodity price? Personally I would not unless the commodity is stable or I want to trade that commodity (there are long term issues with trading USO or UNG ETFs.

  3. Edward Jenner

    Well, I can see the setup, but entering the trade I would think about
    where are the most likely places to have problems. In my mind you have
    the 200dma just below and you are close to the 20period lower Bolinger
    Band (Of course what you look at depends on your system, that’s just
    something I look at). So anyway, my exit plan would be something like
    ‘if it finds support at the 200dma or the lower band’ then exit. Price
    went through he 200dma on the 1st and then down to 24.50 on the 2nd –
    looking good. However the lower 20day band was at 24.90 on the 2nd and
    rather than closing near that price, the lower price was strongly
    rejected. While I don’t trade candle patterns as such, this would be a
    warning to watch it closely the next day. At this stage, once it
    regains the 200dma for any length of time, and certainly a close above, as it did the next day, I
    would be out.

    Also, perhaps much more importantly, this is an
    oil stock, so you are basically trading oil (or at least it can have a
    large influence). On the 2nd oil hit an important potential resistance
    level (look at oil prices, not USO) at the low after the mid-east summit
    on April 18th, so that would be an area to make a decision.

    So
    one question, is what do you think oil will do? It’s been moving, so
    whatever it does may have more impact on your stock than any
    technicals. Do you want to use technical analysis on a stock driven by a
    commodity price? Maybe. I sometimes trade commodities though ETFs based on technicals. But IMO care is needed.

    I would also ask yourself if you really want to be holding stocks through earnings when trading on this timescale. Most swing trades will not. If you are holding for months, then I think it is a different matter. Obviously a personal decision though.

Comments are closed.