My limited experience has been that adding more to a winning trade compounds the loss of already earned profit if the trade moves against you suddenly... it can also turn winning trades into losing trades. IE: trading SPY or DIA, 500 shares per trade, positive .50 and thinking it looks good for another .50 so I double up and then it pulls back .25 - just turned a winner into a scratch trade and then have double exposure if it keeps going against me. In my opinion, raising my position cost Long or lowering it Short for more shares doesn't make sense. I prefer to set my trade size and then live with it until exit unless it goes MASSIVELY (like a 90% down day or similar strong move) in my direction in which case I'll double up for the remainder. This is what I've been doing lately and it works much better for me. Then again, I'm still learning. Paul
Most of the successful flat price traders I know scale out of their profits. That's a personal choice of style. The inescapable truth is that one should NEVER scale out of a loser.
Scaling out of a loser is dangerous. By doing this one increases unnecessary exposure to a bad trade.. All traders know that if the trade is not profitable immediately chances of a real winner drop off significantly. --MIKE
TrendFader, Perhaps in a game where 90% of players lose -- it may be a viable strategy to do what DOES NOT come "easy" psychologically to the majority. Are you somehow under the impression that our condition is in accord with her nature or the nature of successful trading? In most cases scaling out at a 1R-gain will not increase expectancy, it will only serve to indulge and perpetuate your psychological shortcomings as a trader. Reducing R-gains and increasing win% will not neccesarily make drawdowns smaller and the smoothness of the curve will only be seen on the microscopic levels. This is a game of odds and as such, the microscopic monetary results are nothing but noise. My equity curve spends 65% of the time in a drawdown or flat period BUT my draw-downs from peak to trough are 95% of the time contained to less than 8R's. In addition if you were to plot the line curve at say every 15-20 trades I would rarely ever have a downtick. My advice to you would be to 1) go play that cardtrick game that was posted a few weeks ago in the psychology forum and seriously ponder what sets that illusion in motion 2) spend less time trying to figure out what is going to be "easy" for you to do and more time trying to figure out what is in harmony with her. You are trying to conform the external aspects of the game inorder to accomodate your internal world. Flip the script bro. Instead of trying to bend the spoon ------- > BEND YOUR PERCEPTION OF IT!
J Commiso, I respect what you are saying. But if you ask any pro trader what they would rather prefer.. they will tell you a consistent income rather than a greater expectancy. I have spoken to some of the top NYC pro traders, and especially hedge fundies. Any attempt to smooth out the equity curve should be more important than increasing expectancy. If one scales in effect they are "attempting" to smooth their income by decreasing drawdown. Whether they succeed or not is regardless. The point is when you trade.... optimize for smoothness as opposed to expectancy. Here is the meat of my argument. Assuming all things being equal and we have two types of traders.. One trader's quest is to smooth out his equity curve and the other's is to optimize expectancy. Lets just say the trader that "tries" to smooth out their equity curve has a lower return rate on capital than the expectancy optimizer. However, the expectancy optimizer has a greater risk of drawdown and bit more wilder equity curve. Here is the point: Which one do you think would have an easier time increasing their capital and position size?? The psychological benefits far outweigh any other. The added value of having the ability to easily increase capital base, is a great plus.... which in turn would probably lead to greater $ gains in the very long run. Nightmare scenario: you increase capital base and size of trading only to go immediately through big drawdown. In reality it will slow you down. Having the smoothness will not eliminate this.. but make it less likely. Obviously the ideal situation in trading would be to have both.. max expectancy and max smoothness. But this is impossible because of the inherit nature of risk. I advocate smoothness over expectancy. In reality when it comes down to everyday life and making $ for a living not theoretical testing.. in my "opinion" smoothness is the key. Now, I am not saying I am right and any other method is wrong.. this is just my opinion and the opinion of some of NYC's top traders. In the very long run.. and if your really wanna make millions trading and be in the business for a long time this is something one must consider. --MIKE
mike i understand what you are saying, and in some ways i agree with you. the assumption you are making though, is that a smoothing out of the equity curve by scaling out is necessarily successful simply because the attempt is made to do it. ie, the trader goes for smaller $ amount gains (by scaling out), but gets them more consistently, simply by virtue of wanting a smoother equity curve. i don't think there is any reason to assume that just because that is what the trader wants (the pyschological easiness of a smoother equity curve) that he is necessarily going to get it by scaling out. and if the scaling out doesn't produce the assumed increase in number of winners, then the equity curve isn't gonna look all that impressive to me; "smoother" with a downward slant doesn't seem too enticing to me. the need for a pyschologically easier trading method is, as commisso says, an attempt made by the trader to make the market conform to the trader's needs. however, that's not to say that the market only rewards trading styles that go for "expectancy optimization" (to use same terms as you). of course scale out strategies can bring positive results (as evidenced by the people employing them), but it should not be assumed that just because this is the goal it will automatically happen. i agree with your point about increases in size being easier when the equity smoothness is achieved, and that, in the long term, these easier size increases compensate for the assumed lower overall expectancy (of "equity smoothness" systems/styles).
Mike, First off you are making a wrongful assumption that a higher win% / lower R-gain decreases the magnitude of draw-downs, wich in my experience is simply not true. It would seem that way because you are putting a considerable number of more upticks on the curve, but you are failing to see both sides of the coin -- the upticks are smaller, as such, when the drawdown does occur it takes a lot more wins to work your way out of them. I frequently put on 5-6 straight losers, but it is soooo easy to continue to perform at an optimal level because I know the ust one win will put a new high on the curve. Secondly, like every other word in the english language -- "consistent" is a subjective term. As I said in my prior post my account spends 65% of its time in a draw-down or flat period on an extreme "micro" level (trade to trade), but if you were to stretch it out to say every 10 trades there would rarely ever be a downtick on the curve. Lastly, you are again making a wrongful assumption that it is psychologically "easier" to perform well when you have an equity curve that spends most of its time on a micro level in an uptrend. You have to look at the motives behind employing such a method, the traders usually hate "losing" money so they employ a method that rarely does, but when the unavoidable does occur it may be harder for them to cope with it because they 1) don't spend much time in one and 2) because it is going against the very thing they were trying to avoid by employing the method in the first place. Perhaps it may be easier to deal with a draw-down when you are in one the majority of the time? This is just my own experience with the matter... Oh and one more thing, what constitutes a "top" trader and what does being in NYC have anything at all to do with it????
To scale or not should be viewed this way - The not scaled part, all by itself, should be profitable as a standalone trading strategy. The trailing part, all by itself, should be profitable as a standalone strategy as well. Otherwise, you would have wrong expectation what the performance actually is for each individual part. In short, you are trading 2 strategies that just by coincidence that they share the same entry points