Do you scale out of your trades?
- Thread starter funky
- Start date
Many good points on the thread and I'll chime in with my experience. Not implying that my way is the way, because I don't believe that there is only one correct manner to manage a trade, but for me this has worked for many years.
My basic scenario is that I enter a trade with my full load and then exit half at a predetermined profit using a limit order. The second half I let run with a trailing stop. Should my initial target not be reached, then I lose on the entire position.
Overall, the second unit is slightly more profitable than the first, but the variability of the equity curve is much larger and psychologically harder for me to deal with. At first glance it may appear that I am giving up some absolute profit for piece of mind, but my view is that by being more at ease I am trading better and will have a longer trading career which will hopefully translate into more profits in the long run.
Just my 2 cents
My basic scenario is that I enter a trade with my full load and then exit half at a predetermined profit using a limit order. The second half I let run with a trailing stop. Should my initial target not be reached, then I lose on the entire position.
Overall, the second unit is slightly more profitable than the first, but the variability of the equity curve is much larger and psychologically harder for me to deal with. At first glance it may appear that I am giving up some absolute profit for piece of mind, but my view is that by being more at ease I am trading better and will have a longer trading career which will hopefully translate into more profits in the long run.
Just my 2 cents
...sorry, not trying to be cryptic, just didn't want to belabor the point. From the top, I optimize as follows:
- test an NQ entry strategy which can trigger any time between 9:30 and 10:30 ET with commission and experienced slippage, using no stops and no profit targets
- if its expectation is $25 per trade or more and trades at least once every two weeks, I attempt to optimize it to more than $50 per trade
- next I a "too late to enter a trade" time and parametrically vary that to see if it helps
- then I vary the stop time to see if that helps
- then I introduce a profit target and parametrically vary that
- next I introduce a stop and parametrically vary that without a profit target
- next to last I iterate the stop and the profit target in the neighborhood of their separate optima
- finally I try a reverse at the optimum stop, and reject the system if it benefits from reversing.
I also discriminate against strategies which do not support my existing systems, unless they are powerful harbingers that the trades previously put on are going sour.
The end result is that there are three outcomes to an entered trade: it runs to the time stop, gets taken out by the stop loss, or gets taken out by the profit target.
And BTW, I would never think that you are dull. I have learned a hell of a lot from your posts over the years.
- test an NQ entry strategy which can trigger any time between 9:30 and 10:30 ET with commission and experienced slippage, using no stops and no profit targets
- if its expectation is $25 per trade or more and trades at least once every two weeks, I attempt to optimize it to more than $50 per trade
- next I a "too late to enter a trade" time and parametrically vary that to see if it helps
- then I vary the stop time to see if that helps
- then I introduce a profit target and parametrically vary that
- next I introduce a stop and parametrically vary that without a profit target
- next to last I iterate the stop and the profit target in the neighborhood of their separate optima
- finally I try a reverse at the optimum stop, and reject the system if it benefits from reversing.
I also discriminate against strategies which do not support my existing systems, unless they are powerful harbingers that the trades previously put on are going sour.
The end result is that there are three outcomes to an entered trade: it runs to the time stop, gets taken out by the stop loss, or gets taken out by the profit target.
And BTW, I would never think that you are dull. I have learned a hell of a lot from your posts over the years.
Below is a quote form the book with insignificant omissions
From the pure statistical point of view, partialling in is better way to trade than partialling out.... By partialling out you limit your profit potential. Also... entry point is usually the point of lowest confidence. As the trade develops, your confidence level rises with new confirmations that your initial idea is right. It makes perfect sense to enter just part of your position while you have no confirmation and add to it as the market tells you that you are right. If the trade doesn't work, you get stopped with just part of your position. In real day trader's life, hovewer, other factors come into play. Fast intraday trades do not always allow for mutliple entries along with trend... Futhermore, from a purely psychological standpoint it's much easier to keep your position when part of your profit is secured... Most of us have to maintain mental balance in order not to burned out from everyday's tension. Partialling out serves this purpose well. In some ways partialling out functions like insurance. In most cases insurance is a waste of money because nothing happns. But in rare cases when something does happen, insurance saves you from a hard hit. What is it that you insure against when you scale out? It's a trade turning around halfway and hitting your stop. By scaling out, you decrease the size of your win and increase the number of your winners, while maintaining part of your position in case a stock continues moving in your favor. Insurance also buys you piece of mind, which could be as important as monetary issues... Again, as with all trading choices it's a matter of personal preference.
From the pure statistical point of view, partialling in is better way to trade than partialling out.... By partialling out you limit your profit potential. Also... entry point is usually the point of lowest confidence. As the trade develops, your confidence level rises with new confirmations that your initial idea is right. It makes perfect sense to enter just part of your position while you have no confirmation and add to it as the market tells you that you are right. If the trade doesn't work, you get stopped with just part of your position. In real day trader's life, hovewer, other factors come into play. Fast intraday trades do not always allow for mutliple entries along with trend... Futhermore, from a purely psychological standpoint it's much easier to keep your position when part of your profit is secured... Most of us have to maintain mental balance in order not to burned out from everyday's tension. Partialling out serves this purpose well. In some ways partialling out functions like insurance. In most cases insurance is a waste of money because nothing happns. But in rare cases when something does happen, insurance saves you from a hard hit. What is it that you insure against when you scale out? It's a trade turning around halfway and hitting your stop. By scaling out, you decrease the size of your win and increase the number of your winners, while maintaining part of your position in case a stock continues moving in your favor. Insurance also buys you piece of mind, which could be as important as monetary issues... Again, as with all trading choices it's a matter of personal preference.
...as one system trader to another, scaling in smacks to me of discretionary trading. I constantly remind myself while in a trade that any "discretionary" impulse for taking profit or stopping loss would constitute trading a non-tested, non-optimized system.
Also, I think the discussions so far have missed what is to me a valid way of scaling in, which is to buy optimized retraces in the loss direction away from the entry. While IMO such retrace entries do not qualify as tradeable systems on their own, because they don't occur often enough, they powerfully improve the profitability of the basic system. In its nerviest form, this means piling on one tick before the stop price.
Also, I think the discussions so far have missed what is to me a valid way of scaling in, which is to buy optimized retraces in the loss direction away from the entry. While IMO such retrace entries do not qualify as tradeable systems on their own, because they don't occur often enough, they powerfully improve the profitability of the basic system. In its nerviest form, this means piling on one tick before the stop price.
I can see making a case for not scaling with very thick futures contracts like the bonds or the minis. That makes some sense. It's painful, and I've had this happen much too often to be sitting on the bid, watching trades go off there, then have the market take off without you....still sitting on what used to be the bid. OUCH. That's a case of being right and missing the entry. But scaling or paying up for the offer can really hurt if you are scalping.
Now when it comes to stocks, especially the more volatile ones, I can't pick a level. Period. I generally know how I think the trade will progress, but there's no way of knowing that it'll come down to xx cents.
Let's say a stock is strong, and I want to be long. The market pulls in a bit. That's my entry. Not 5 cents from the previous high, etc. because it may pull back two cents, then the buyers come piling in again. I'll scale into these all the time. I'll think of my average entry price as say, 5 cents down, as in the above example, so I'll bid 2 cents down, 4 cents, 6 cents, and 8 cents. If I get filled on them all, then I'll be in the same position as if I had entered the whole deal at 5 cents. If I'm wrong, then my same rules apply about getting out. If I'm right, but my 5 cents is wrong, I'll at least have a chance to catch some of the position to make a profit.
Sure, if the trade always went 5 cents then took off, I'd only get half my position on my winning trades. No argument there. Of course, if I knew that, then I wouldn't scale. I scale because I cannot say what price the stock will trade to.
Now when it comes to stocks, especially the more volatile ones, I can't pick a level. Period. I generally know how I think the trade will progress, but there's no way of knowing that it'll come down to xx cents.
Let's say a stock is strong, and I want to be long. The market pulls in a bit. That's my entry. Not 5 cents from the previous high, etc. because it may pull back two cents, then the buyers come piling in again. I'll scale into these all the time. I'll think of my average entry price as say, 5 cents down, as in the above example, so I'll bid 2 cents down, 4 cents, 6 cents, and 8 cents. If I get filled on them all, then I'll be in the same position as if I had entered the whole deal at 5 cents. If I'm wrong, then my same rules apply about getting out. If I'm right, but my 5 cents is wrong, I'll at least have a chance to catch some of the position to make a profit.
Sure, if the trade always went 5 cents then took off, I'd only get half my position on my winning trades. No argument there. Of course, if I knew that, then I wouldn't scale. I scale because I cannot say what price the stock will trade to.