Great entries, horrible exits

Discussion in 'Strategy Development' started by illiquid, Jun 21, 2005.

  1. I think I have a pretty good knack for finding decent entries. I trade on 60 minute and daily time-frames, but I'm still able to use very tight stops. I don't try to catch lows or highs but I can usually find a place to enter right before the highest momentum part of the move takes place.

    So what's the problem? My winning exits are pretty much horrible in retrospect (the losers are no-brainers). I will almost always exit way too soon or hold way too long. You'd think that if my timing was spot on for entries that at least my exits would be decent, but they're far from it. I've resorted lately to just taking profits based on percentage gain but I realize it's arbitrary and nothing like what I use to determine my entry points. I think I'm probably only taking about 30% of any given move on average, it's like once I have the position on all the newb instincts come out and I will invariably sell out on the lows of a pullback, or otherwise sit way too long and watch a 50%+ retracement on a great move take me out.

    Would appreciate some thoughts, especially from those who trade beyond intraday time frame, I know that taking profits is always hard but I think in my case things are way too asymmetrical.
  2. Are you able to identify what type of trade it will be before hand?

    If it is a trend then identify acceptable retrace levels. If it is range S/R based then identify inflections and position exits around those infections/breaks...

    Those are the two types of trade I really focus on personally and I will base my stop placement accordingly. I also change my stops frequently, both stop losses and stop profit losses (only in the direction of my price movement though).

    Flags (pennants, broken vs. completed), S/R breaks and the rest are much more difficult to quantify but, in general, S/R levels seem to be a self fullfilling prophecy IMO.
  3. Try scaling out of exits.
  4. 2 ideas:
    1) if long, use the Lowest Low approach....say a low is printed that is lower than the previous 3 lows, get out.
    For shorts, use highest high.

    2) use chandlier: whatever your peak profit in the position was, get out when the price falls to a % from that high water mark.
    Example: you're up 2% in a position, get out if it goes 0.5% against you. Scale accordingly: if up 1%, then get out if your position goes to +0.6% (0.4% against you), etc.
  5. Assuming you have a system for entries, what's the problem with systemitizing your exits? IMO, any exit strategy that leaves room for emotional decision making is a poor strategy. You'll just keep beating yourself up on every trade.

    Try setting a profit target, then exiting at the first down bar after that. For me, as a swing trader, it lets me stay in winning trades longer. If your system tells you that X stop is the most efficient, then go with it, unless you have a great reason to overide the system.

    Also, be creative in your exit signals. Try a technical indicator, or other approaches. As a swing trader buying pullbacks, for example, I wait for a certain level of stochastic, then let the stock run to the first down (daily) bar. So I let the stock take itself out at a certain level. Think outside the box.
  6. Most of the trades I do don't fall into specific categories, I only go as far as classifying by time frame that I looked at to get in the trade. I use charts but don't specifically trade off s/r lines or bullish/bearish patterns, when a certain number of factors line up I enter, I try not to anticipate exactly where my profit targets are before I see how the price is moving but maybe that's where my weakness is. But basically the longer the time frame I use as the basis for entry, the wider my stop is.
  7. Yeah this is the logical solution but when I tried it I still had problems determining the system by which I would portion out the exits -- as in, should I exit 1/3 every X amount of ticks in my favor, or divide it up into 1/3 systematic trail, 1/3 fixed target, 1/3 gut instinct, etc. It left me undecided most of the time so I just gave up on it, but I might have to try again.
  8. nitro


    Those things occur to you only in retrospect. It is the wrong way to look at trading.

    The correct way is to have a way to enter, and a way to exit that is 100% mechanical. Even in that case, you will always be looking to improve the entries and exits when your system gets killed, and not worry about it too much when it has a good day. It is all hindsight.

    The point is, anything can happen at any time, and once you enter, everything that happens in between your entry and exit or reverse is random noise from the point of view of your "system", even if the system uses discretion to enter/exit-reverse. Think about that last statement carefully for in it lies all truth, and it will allow you to let go of fear and greed...

    If you don't know when to exit and your system is unprofitable because of it, then you are not applying the strict discipline of systematic trading and have no system at all - just the illusion of one, based on hindsight.

  9. Wow. A philosophy of life. Birth, death, random noise. As good as anything I've seen.
  10. =======
    Illiquid ;
    Figure you are close to solving the problem because you have it pretty much pinpointed;
    mainly taking profits way too soon.

    On longs in a bull market give much better weight , focus, data recording, focus ,attention , study to, dailys , weeklys, monthlys, and yes look some , not mostly intraday. Mainly weekly, monthly

    Also look some at yearly charts;
    like 10 year SPY =10 candles

    Shorts are much trickier and usually, unless its crippled , just not as much there % on daily charts. My semi shorts have lost money lately.

    System swings/work better for me with ;
    liquid stocks
    NYSE more than derivatives
    monthly weekly, daily time
    longest strongest like homebuilders.oil stocks

    Dont think selling lows of pullback is your main problem;
    in a bull market ,;
    unless its overtrading ,
    youre looking at to much intraday,
    UNLESS you play sloppy trenders, like derivatives.

    Also lost a little bit on semi liquid MTH;
    so didnt play it but once, even though its highly rated by IBD

    #10     Jun 21, 2005