Scaling out has one distinct disadvantage, it exposes your entire position to the entire stop but reduces your exposure to the entire move.
Beyond that though, scaling out is a huge confidence booster, there's nothing like being in a free trade!
At some point, I believe scaling in is the ultimate answer. Lets say you trade 10 cars normally. A typical scale out maybe to take 3 cars off at 5 ticks, another 4 off at 10 and let 3 ride for as long as you can. This has the built in disadvantage of having your entire position exposed to the stop.
If you enter instead with 3 cars, add 4 more at 5 ticks, and another 3 at ten ticks and let all of them ride, in theory anyway, you are exposing only a small portion of your total position at the early stages of the trade when it is most likely to go against you and you are adding positions as the trade gets going in your favor. I read this somewhere and it seems to make a ton of sense. For me, if I took a small position at the beginning of a move, added to it on a retrace, and again on future retraces until my entire 10 cars was in play. Then all out at a predetermined target or when price action says the move is over. This seems to me the best way of all.
Of course, I have never done this nor do I really know how to it but at some point in the future, I'd like to try.
Simplicity is the ultimate sophistication, Leonardo da Vinci
Most people chose unhappiness over uncertainty, Tim Ferris
It is correct that you reduce your exposure when scaling out, but you could even further reduce your exposure, if you take the profits on the entire position, and never trade again. This is of course an oversimplified reasoning. But digging deeper I can imagine three reasons, why to scale out. And only the first one would be risk-related.
1. System Trading: Improving the Sharp Ratio
If you have made your homework, you know the expectancy for your trade in both versions, the first target and the second target (second target could be a trailing stop as well). The second target will have the higher expectancy per trade, because otherwise you would exit the whole position at the first target. However if you stay all in, you will have a larger dispersion of your results, as some of the positions will be stopped out. So you may opt to scale out, although this reduces your expectancy for the trade, because you may get a better Sharpe Ratio.
2. Discretionary Trading: Crowding Signals that the Move Could Be Over
Scale out after the market has signalled that the move might been over. If I am long and price is approaching resistance, then accelerating with a high volume wide ranging bar and hits resistance, I scale out and take part of my profits. A similar sign would be a churn bar and/or a narrow trading range. I am scaling out, because the probability that the move continues has changed, not because I want to reduce my risk.
3. Psychological Comfort and Increased Trading Opportunities
Just taking half of the profits, set stop to break-even, and you are out of the danger zone. Playing with the money of the house, you can let your profits run and catch a larger move then and now. Also you can now look for new opportunities. This is rarely taken into account, when backtesting a system. How long does the system stay in the market and how often does it trade? Ideally I want to have a large number of quick moves that make money, and not a slow moving position, which keeps me out of other trades. I want a quick rotation of my working capital.
Personally I have never so far used the option 1, I often push the exit button if resistance is approached - but these are planned scale-out targets, or if I feel uneasy because the market is not doing what I anticipated and my scenario is no longer valid.
The following 7 users say Thank You to Fat Tails for this post:
Expectancy.........hmmmmmm........the amount or distance you think a market move will take you........based on what? Your historic research? Your presumption? Your arrogance?
There is no trader alive that can tell how far any entry signal will take them. You can't set some arbitrary Risk/Reward ratio and expect the market to comply. All you can do is know when you are wrong (correct stop placement) and take profits (first and second targets) and trail your stop as the trade moves in your favor.
Is this intentionally reducing your profits for psychological relief? NO !! This is common sense.
The following 2 users say Thank You to Jeff Castille for this post:
Come on. If you have a backtested system with fixed rules, you do have an expectancy for each trade, which is the expected profit per trade of your backtest. You can also backtest a trailing stop. I am not setting arbitrary ratios. If you have two separate strategies for exits - the scale out (target 1) and the trailing stop - the risk profile of the combined strategy can be better than the risk profile of each individual one. So don't call this arrogance, please. I was just responding to the idea of Zoethecus to reduce risk.
And yes, I am often taking profits on half of my position for relief. Once I have done this I can reevaluate the situation without suffering from bias associated with the position. If need, I can also reenter the trade.
The following 2 users say Thank You to Fat Tails for this post:
Do you have plans to become the first-ever 100 lot crude oil scalper? I think there is a lot of research done on trade execution, a bunch of software companies focussing on trade execution algorithms to minimize market impact of large orders. Don't think that it is possible, to do this in the rather illquid markets you mentioned. But this is not something, I have thought of, I have never had such a problem.
There are a few statistical papers of how far a price will move based on the time frame you trade and the instrument. (I will try to find it and post a link)
If you trade a 1 min chart, dont expect to make 70 ticks per move, statisticly <10 is expected based on your system. If you trade an hourly or a daily chart, you can look for a larger move per trade. Its all statistics and math. See how Jim Simon build his funds. His systems trade miliseconds and take a fraction of penny. Soros traded the BP for months and took a billion on a trade.
So yes, a trader should have expectancy for each trade he takes.
My previous post was not in response to your post. None of what I said was directed at you.
My "expectancy" for any given entry signal is minimal......thus this minimal expectation is the place for my first target......I have "backtested" and have determined that any "edge" that I have will more often than not reach my first target. Beyond that my "backtesting" has also determined what a second target might be.......but I have NO assurance that this objective will be reached. I just want to make myself "available" for any further moves in my favor by having a second target and ideally trailing a stop from there if there is a really big move.
As far as "backtesting" goes........each moment in the market is unique......therefore what it has done in the past has NO bearing on what it might do on a current trade.......to think that we know how far the market might move or to say since we want a certain r/r that a certain target must be reached.......that's arrogant.....the market doesn't care what our desired r/r is.......we just need to take what it gives us
I believe that the psychological relief that we feel when early targets are reached and stops are moved up is a secondary benefit to scaling out NOT the primary objective.
The following 3 users say Thank You to Jeff Castille for this post: