Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Not sure how you see what was written as an argument.
It was in response to your classification of them being separate strategies.
What led me to favor the method was past inefficiencies in position building upon a strong trend without risking too much at a time.
If you, or anyone else, has an alternative to doing so - I'm all ears.
ETA:
In hopes of pleasing you a contradiction could be that the inferior strategy would have to be replaced by something and if you are just focusing on one instrument...
Truly a good argument, especially because it IS backed up mathematically (my backtesting likewise confirms that scaling out, in ALMOST all strats is suboptimal).
However, in real world trading, the approximate target we may have in mind (S/R or trend lines, etc.) may change as the actual trade unfolds, be it news related, an unexpected turn in the euro, whatever. Yes, this is up to the trader's discretion and is where the "art of trading" is needed but say you were shooting for AIAO 5 points and it has hit 3 points and a sideways chop, do you take off all of it there or just a half of your position and see if it pushes through or you get stopped out BE?
Another example, say you are in a few CL contracts short from 100.4 yesterday and it has just spiked down through support but showed a reversal bar at 9:01. Yes, you are perhaps still bearish on CL but slightly less because it has come in significantly and you think the current trend is hard to read due to the overnight action. So, do you get out of the entire position, or just half and see what it (and the Euro) does from there?
Logically, scaling out makes sense to me because the markets are fluid/uncertain and scaling out is likewise much more mentally "fluid" than the etched in stone ALL out.
Failing all these arguments, scaling out can be a nice psychological crutch and as the game is at least 60% psychological, IMHO, it is a good option
This statement is the best argument I see for using a scaling out strategy. If you take a trade with two positions and you close the two positions with different rules then you have effectively place two trades. If you use consistent (but different) rules for your two exits then you are effectively trading two strategies (as baruchs pointed out) - and one is likely going to have a higher expectancy than the other.
The idea that certain conditions may favor one exit criteria over another COULD lead to such a strategy (two different exit criteria on 2 unit trades) smoothing results over time. However, (and I can't back this up with math) I would guess that it is unlikely the two exit strategies could be configured in such a way as that they both lead to equal expectancies and so one will always be better than the other.
Personally I consider it a good trade-off to gain some smoothing of P/L (thus making it psychologically easier to stick to the plan) in exchange for absolute return (at the expense of less smooth P/L). However, I hope to evolve to being able to trade an AIAO strategy with its concurrent choppy P/L and higher absolute returns.
I absolutely agree that a smoothing of P/L is the most important aspect in trading. I even showed that if the correlation btw. two strategies outcome is negative you can be better off even if one of the strategies is a losing strategy. In scaling this is NOT the case. The "two" strategies are very very very high correlated, and thats why you will NOT get a P/L smoothing!!!
For example lets take a scaling in strategy. If we look for correlation of the outcomes from the scaling strategy it is 100% correlated to the initial strategy!! If you entered a trade in the scaling part and now you have 2 contracts so if you win you win on both and if you lose you lose on both. If you look for correlation from the first part then you will have some trades that are winners or losers (depending if your scaling in is when you up front or in lose) without a trade on the scaling part.
I want to say something about my trading if some people do not know it yet.
I believe only in automatic trading. All my statements are mathematical FACTS and I see if some addition or a change in strategy like in:
had an edge in backtesting.
As I see it discretionary trader is a gambler (its OK for some people, but not for me). That also mean that a discretionary trader may take a trade and on the next day (after he forgets about it) he replays the market action and he takes a different trade and again some days later. Each time he will have very strong arguments why to do it like he did and he even won't be confused to see that in the same situation he behaved differently. As you call it, he had different "gut feeling" each time.
To conclude:
You will get much better results if you trade several setups and/or on several instruments and/or on different TF, but scaling is the worst possible solution.
If you still want to do it at least take a note if each strategy (original and the scaling) is profitable.
I may really did not understand every thing, but what when only scaling in and always get out with the whole amount of contracts, when stop is hit or when taking profit ?
Baruchs - if I understand your statement correctly you are saying that the size of a profit target has no effect on the expectancy of a given trading strategy. I doubt this is what you're saying so I think I don't understand your statement.
If you initiate two positions in the same place with the same stop and have two different profit targets won't each of these trading strategies (assuming they are applied consistently) have a different expectancy?
As a oversimplified example, let's say you the two profit targets are 10 and 20 - in ranging markets the 10 pt target gets hit 2.5 times as often as the 20 pt target. In trending markets it gets hit exactly as often as the 20 pt target (by definition). If a market ranges for 2 weeks straight, the strategy with the 10 pt target will have a higher expectancy than the 20 pt target strategy for that 2 weeks. Thus when markets go from ranging to trending and back, the two profit targets will smooth returns.
Is there something wrong with my thinking here? honest question - I'm still learning...
In my explanation I refereed to scaling in. This was the example someone posted. You talk about scaling out. Its the same sh.., but the arithmetic is a little different. I don't intend to give here private lessons, so don't ask me what if I scale in and then scale out and sometimes move my stop to BE and sometimes move my PT to BE and trail my stop and and...
All I'm saying is: Treat scaling as two separate strategies. First check that each strategy is PROFITABLE. Then look if one is not inferior to the other.
My point is that no matter how you do it those two strategies will be much much more correlated in their outcomes than any other two strategies(not scaling type)!!!!!
If this is true then the smoothing of P/L will be much much much smaller than with any other two strategies.
Absolutely true, trading IS gambling in that you are putting money on an UNCERTAIN future. Whether automated or discretionary, it is a game of probability, pure and simple.
EVERY day of trading is unique but the detected similarities that we have experienced 50 times are what we are relying on to make us some coin. If a trader fails to see these similarities, as you said, "he forgets", then cannot possibly be a profitable trader because how can you possibly see your edge or the probabilities of a successful trade if you can't recall whatever your setups are or what they or the market dynamics look like in real time?
Also, I am not saying "discretionary" is to have no rules at all but to include some flexibility in your rules because, as i wrote in a prior post, the markets are fluid/malleable. I know of no traders that are consistently bringing in money that trade purely on "gut instinct" and everyone I know that does well has a method/system/strategy (or several), one is totally automated, a couple are more gray box and a few more are analyzing everything themselves and using their own discretion to a certain extent at different times.
Say your system has you in a CL contract from 97.05 and the rule(automated) to sell your CL contract when it hits 98.48 but when it has climbed up to 98.43, news comes out that Gadaffi has been killed and it starts to tank. The trader then uses their "discretion" to decide if they want to sell the position a little early or wait until they are stopped out at BE (not have to wait to long). I am not saying that you cannot make money with AIAO automation because clearly, you can. I am saying that people can make just as much or more if they really know market dynamics, have mastered their emotions and are great at getting good production, i.e. scaling out, getting back in with size or more size when called for...
Oh yeah, scaling out does not necessarily mean that you are out of that portion of the position for good as you can always add it back, especially if you are good at reading PA..."OK, it should come back in here so let's dump a couple of contracts. OK, it has come in 20 ticks, let's put them back on again with that 20 ticks price improvement. Surely, it doesn't take much of an analysis to see that today's bottom to top CL move of just over 200 ticks had within it an easy 900 ticks of potential production, short here, cover there, reverse entire position, etc.
Anyway, just saying there are at least 34 ways to skin a cat and scaling out can be a valuable scalpel if you know how to use it (unfortunately, my fingers were removed in a weird garbage disposal accident as a child so...well...)
With individual stocks I just put the money in there and would sell if it went down about 7% or if my opinion changed. Adding to the position was usually a result of ending positions in other stocks and not having a better idea of something that would be likely to perform better % wise.
With futures being so leveraged a question is then (or please anyone tweak if you have a better idea):
1. Would it be better to put the total sum planned in the future, when you think you found a bottom, then try to get out if you lost 7% or so of the money allocated to that investment.
Or
2. Build on that investment by risking something like 1.5% at a time.
With the risk of losing more than the intended 7% and uncertainty of not losing it again the next time around I can't justify not just risking 1.5% at a time. Others input would be very much appreciated.
FWIW, I screwed myself out of a lot of money recently by selling 1/2 the futures contracts prematurely, instead of waiting to be stopped out. I don't even want to call it scaling out - it was panic selling where an irrational fear of loss is justified at the time by not wanting to "get greedy". The remaining contracts were stopped out on the day it hit a 30 year high.
Making investments when the outcome is uncertain, but you believe the odds of profitability are in your favor, is speculation.
Putting money towards something where you know the odds aren't in your favor is gambling.
If a person thinks the odds are in their favor but are wrong and lose money they are an unsuccessful investor.
Top poker players and blackjack card counters are gambling and they are increasing the size of their bets according to when the odds/probability ARE in their favor much like when a high PROBABILITY setup calls for you to throw in your digital chips and roll the dice (hopefully loaded in your favor;-)
Call it gambling or speculation, it's just semantics, bro
