Continuing on... Again, this is just not correct. One does not always expect or anticipate what a "90/10" trade will be. All your trades will be initiated with +EV expectation, and some of them will initiate with higher conviction than others. But you do not know in advance where the "monsters" will be. A simple example: Short $AUDUSD. This looks like a good risk:reward trade. We are making good money on it so far. But how much profit will the trade generate for P&L at the end of the year? We don't know, because we don't know how the scenarios will unfold. The RBA could initiate a surprise rate cut, or base metals could collapse at a faster rate than expected, and AUDUSD could turn into a huge trade. Multiple pyramid opportunities could set up along the way, making it the trade of the year. Or, things could unfold a different way and it could wind up being a modest sized trade, with "ok" gains but nothing fantastic (as happened with short euro in 2011). The point is, we don't know and this is a common case. The world has too many complex variables for traders to know beforehand how the full story of their convictions will unfold. It is not always clear -- and in fact frequently NOT clear -- which trades will reveal themselves as home runs and which will be singles, bunts etc. Not how it works -- at least not for the discretionary global macro style that combines technicals and fundamentals, of which PTJ is a leading light. You focus on your best ideas, yes, but you don't necessarily know which ones will blossom from the start. Again, read Grant... or Invisible Hands by Drobny, a series of anonymous interviews with top practitioners. If I recall correctly "The Equity Trader," which I'm fairly certain is Steve Cohen, speaks further to this. Yes! You started off correctly here... it's not quite "logically bet the same size for all trades," but it is "logically bet the same size for MOST trades" (relative to certain benchmarks and equity curve considerations having to do with the state of one's portfolio and the current market landscape - as always glossing over some important nuance for the sake of time). So where does the "varying size massively" come from? It comes in as conviction develops. This is exactly the point I have been trying to communicate all along. The great traders start small, wait for premium situations to develop, and then "vary their size massively" as a favorable scenario unfolds and the market tips its hand. Think of a graph in which the Y axis is size and the X axis is certainty. The idea is to start in the bottom left quadrant of the graph -- small size and low certainty. You often have to start here because, by the time you have a great deal of certainty, so do plenty of other market participants. This is precisely where the skill of the great traders, and the value of this technique, comes into play. The great trader has excellent information and excellent intuitive skills, but he does NOT have some magic crystal ball that always tells him in advance what the biggest moves are going to be. What he does have, is the ability to "vary size massively" as you put it, by dialing up from small exposure to large -- sometimes very quickly -- as his OWN certainty develops, thus moving from the lower left quadrant of the size / certainty graph to the upper right. Great traders don't think this way at all, for all the above reasons stated. You are always looking for the big trades and needle-moving themes -- but they come about by way of a powerful methodological process, not a decision to "only" go for the top 10%. If venture capitalists could "only" invest in the Facebooks of the world and movie studios could "only" produce the Avatars, I'm sure they would do it that way too.
p.s. One last point: Why would the great traders "vary size massively" in the first place -- as you concede -- unless there was clear methodological value in starting small?
I would agree with the technique of 'adding to winners' or testing the markets and adding more later on IF its based on the fact the trader all of sudden thinks the trade has increased in expectation, specially if it derives from a decreased chance of a loss(And by Loss I mean the price changing from the day the ADD was put on, not the original entry point which is irrelevant) This would be similar to doubling down(doubling your bet) on blackjack when you are dealt an 11 against a dealer 6 I'm afraid some people add to winners for other reasons though. Sometimes they just feel more confident because they have a gain and thus have 'room' to risk more, among other psychological biases If the expectation doesn't change, having 100% of the position on the initial entry makes more money in the long-run obviously I see daytraders all the time having all kinds of irrational rules, calling that 'risk management' and continuing on with their style
Yes exactly -- this is how it's done (or at least how we've learned to do it). The "pyramiding" of a position comes under two general circumstances: * On a marked increase in conviction as the market "tips its hand," for nstance, after a period of vacillation when a data point, market reaction, or other catalyst makes the new direction clear * On a favorable price action development that allows for adding to a profitable position as such that only accrued profits are risked -- a willingness to take a "breakeven" result on a profitable trade, because the implied odds of large gains on an acceleration of the move are attractive. Certainly. This is art as much as science, and the skill component explains why there is alpha to extract in the first place. As a side note, we have zero interest in daytrading... given our style and approach, it's hard to see the point. And the most profitable daytrader I know (an ex-S&P floor trader with 30 years of experience) virtually never pyramids.
i agree with everything you said ,but your decision to get back in is discretionary,and you could lose another 10% ,point was that as a discretionary trader, sometimes, no matter how long you've been at it , sometimes it's worse because you've been at it for so long without getting burnt seriously for so long, just like a stop, you have to honor your ability to get taken (stay humble) because you are being foolish or the new game is beating you, that stop being a mental note that has kept you in the game thru thick and thin, should be a wakeup call, either you are losing a step , the games changing , or you were just wrong, i have a backer and that has always been the risk kicker ,a gentleman's agreement to max loss, and having it be ever present, gives me a discipline that my ego may be biased to overide,, ..there 's a point in one's carreer where the ego hasn't been a problem for ages,and you might want to can the stop limit idea,or take the stop and get back in, aside from that,i agree with your point
Agreed. My only complaint is that this should never be called 'adding to winners', because this is not the reason you are adding to the trade - it is quite possible, for example on a mean reversion trade, that your trade expectation goes DOWN as it becomes a significant winner, and vice versa i.e. you should add to a loser, not a winner (e.g. overextended market getting even more overextended and providing better odds of a bounce back). Even the term 'adding' is incorrect, because you might be flat, or might be too large in your position and need to cut it somewhat. What you should do is 'sizing your position correctly given the new trade odds'. NOTHING else matters except the trade odds, your current capital, and your risk tolerance - if you know those 3, you know the correct trading size. It's irrelevant whether you have a winner or loser, or whether you have to add or cut your position to get the optimal size. These are all forms of mental accounting, and have no rational or logical justification whatsoever. Most of these rules of thumb are frequently misleading and encourage unsound trade decisions with no basis in fact. You should only use a rule of thumb if you understand its limitations, and find it to still be useful (e.g. in a high pressure situation where complicated thought is difficult or too slow). Otherwise, without understanding, they are just as bad and misleading as trying to argue by analogy. Traders should argue and reach conclusions by using provable axioms and rigorous logic only, anything else is simply vaporising capital, increasing risk, and embracing ignorance.
Yes I agree as a rule of thumb it is helpful to prevent going on tilt, to recover your mental stability. Just that once you are experienced, it is better simply not to go on tilt in the first place, or to realise when you are on tilt, and just stop then. Stopping based on losses is good if they have destabilised your equilibrium, but ONLY because they have destabilised your equilibrium, not because you have suffered losses. Therefore the correct approach is to monitor your mental equilibrium rather than your losses per se, and stop when you are disturbed (which could be after a big winning streak too, or a small loss) and keep trading so long as you are not disturbed (you could lose 20% in one day and still be perfectly calm and able to trade well). So, the optimal way to handle this is as follows: 1. Know what triggers destabilisation of your mental equilibrium. E.g. large losses, large gains, emotional disturbances from outside work. 2. Whenever a trigger occurs, examine your mental state. If you are disturbed, then stop trading, reduce your risk on open positions, and work on regaining your composure. Do not trade until your get back to a calm and rational state of mind. Thus you should do this CHECK (not a mandatory 'stop trading' rule) any time you might be unstable (not just when you are down X%). This approach is superior to an automatic monthly stop loss, because it addresses the problem directly (going on tilt), rather than relying on a semi-arbitrary 2nd derivative of the problem (i.e. losing X%).
First example is not pyramiding - you would also, I presume, add to a losing position if you knew that the trade odds had shifted massively in your favour (let's say you short gold on small size, it rallies 1% against you in 1 week, still not near your stop, then the Fed say they are tightening and it immediately drops 1% in 1 minute - you are still down moderately on the position, but it's now a much better short). Equally, if you short gold, it breaks support giving you nice gains, then simultaneously Bernanke says he is going to double the money supply, presumably you would not add to your shorts despite the fact that they were profitable? So, you are "adding size when the trade becomes better", not pyramiding winners. On the second point - trading with 'the markets money' is a pure fallacy, its mental accounting. If you lose 10k of open profit, you lost just the same as if you lose 10k of realised gains. Your entire trading capital is accrued profits since you started trading, and there is no logical reason to treat $1 different to $1 just because it was created by an open position or a prior closed position. Your entry point, by definition, has nothing to do with subsequent trade odds or trade decisions - the market does not know where you entered. If entry price is irrelevant to subsequent trade odds, then rules like exiting at break-even are also wholly irrational. We can illustrate this easily: Imagine trader X buys a breakout as soon as it ticks to a new yearly high. Trader Y, using identical trade strategy, has an execution glitch and by the time it is solved he is faced with buying 1% higher. What should he do? Well, this is easy - what's trader X's position? Long 1 unit. What should trader Y's position be, if he is trading the same strategy? Identical - long 1 unit. He has already 'lost' the 1% profit he would have made had he been able to execute correctly, that loss is gone already, it's in the past, the correct position is based on the *current* trade odds, and it (according to the system) is long. Now, after the market rallied 1%, and trader X had already entered 1% lower, and trader Y got in, let's say it rallies another 3%. Great confirmation of a breakout. According to the strategy, if the trade is good, it should not now re-enter the trading range, so trader X raises his stop to break-even (or maybe a tad below break-even). Now the market falls back to the 1% level. Trader X has 1% profit, and it has not yet hit his stop which is 1% lower. Trader Y is back to break-even. So, the logical thing is for both trader X and Y to have the same stop - 1% lower. Yet if trader Y is following a lazy heuristic like "don't let a good winner become a loser", he will exit based purely on his entry price. Yet if the strategy is profitable, this is by definition a mistake - the correct trailing stop point is the original breakout price, not 1% above it. Trader X is still long and trader Y is now stopped out, despite having the same strategy, purely because Y followed a general rule of thumb without understanding the deeper reasoning (or lack thereof) behind it. It can NEVER EVER be rational for two traders following the same system, with the same capital and risk tolerance, to have anything other than identical positions at the current market price. Either the current opportunity is net profitable or it isn't. Notice that entry price, prior positions, current positions, open P&L etc are NOT part of that equation. Therefore those things are never relevant to what your current position should be. The only inputs that are ever allowable are: current capital, current risk preference, current trade odds. That's it. Anything else in there, and you are by definition making a trading error.
Sorry but this is totally wrong, it is mental accounting which is a total noob error that has no justification at all. It also doesn't matter if you see it differently, if you were taught that way, or if God himself uses that method. What matters is which method is rational and optimal to use. If your method is indeed correct, we should be able to identify *why* it is correct, same for my method. Firstly, just because a good trader does something does not mean it's correct. People are not always immune to illogical reasoning or mistaken beliefs just because they are good at a certain skill set. So although it's worth paying attention to what top performers say, one should certainly not take their word as gospel without investigating the reasons why those beliefs may (or may not) be true. With your last post you are no longer using successful traders to illustrate a point, you are stretching into the realm of arguing from authority, a well-known logical fallacy. An example may clarify this: Jones and Smith are both trading with 91k and identical risk tolerance. Jones follows your approach and runs 91k up to 130k (30% winner) riding a trend, then loses it all back (30% drawdown) due to having pyramided on the way up and run into a major trend reversal. These are the facts: he has just lost 39k, a 30% drawdown. Smith made a separate trade in a different market, and ran his 91k up to 130k, then booked his profits due to getting an exit signal. At the same time as Jones has 130k, 39k of which is open profit based on an open position with 39k risk, Smith has 130k in cash. Smith now meets Jones in the local trader bar, and listens as Jones explains his trade rationale. Smith accepts the reasoning - the market, whilst somewhat extended, is in a clear bull trend. So, how much should Smith risk on the position? Smith questions Jones - what's my risk here, at what point would you place your stop? Jones tells him - 20 handles away. Smiths' eyes open wide as he asks Jones "but you'd lose 30% if you get stopped out, that's insane! I never risk more than 3% on a setup with these odds" Jones shoots back - "but that's ok, it's *open profit*" and proceeds to claim that his risk is actually not 30%, but in fact ZERO! Jones has found the holy grail - a trade with NO RISK WHATSOEVER! Jones and Smith have identical capital. They have identical trade odds. They have identical risk tolerance when opening a position. How then can Jones justify risking 30%, when if he was fresh back from holiday and seeing the EXACT SAME SETUP, he would only risk 3? How can he justify risking 10 times as much? How can he claim a risk of 30% loss is actually a ZERO RISK position? The answer is of course that he can't. If you don't believe me, we can make an even easier reductio ad absurdum. Instead of making it 30% risk of open profit, let's make it 99% risk. Imagine Jones had punted his entire life savings on AAPL at $7 back in 2001. Now he is up almost 900 fold. Smith asks where his stop is. Jones says "at break-even"! Is this anything other than total insanity? According to your theory, Jones is taking not just a sane risk, he is taking NO risk. He could lose 99% of his net worth but according to you he didn't lose a dime. Sorry but this is nonsense. Losses are losses, there is no difference between dropping 1 million of open profit, and 1 million of realised profit.