Again, let us clarify something; RISK CAN BE INCREASED WITHOUT INCREASING SIZE OF TRADE. Risk can also be determined by stop distance.
My model places stop distance in consideration of volatility calculations (higher volatility = further stops & proportionaly smaller size to maintain desired risk... vice versa for lower volatility). Target is then calculated according to a combination of both valatility (larger targets in higher volatility) AND accumulated preceeding losses in the "series". The target = at least enough distance to gain back trice as much as the accumulated preceeding losses. Another thing, I don't take 100% of the trade off the table at the target. This is based on another principle upon which my model is based upon; We never know just how far the trend will continue. Therefore, only part of the trade is closed at the fist target, stops are tightened on another % of the trade. This permits me ti ride long trends despite knowing in advance where I must take some profits to get back preceeding losses and lock in profits.
Anything can work until it doesn't. Many accounts have been blown using all sorts of different technics. If I were loking for a gaurantee of some sort I certainly would not be trading FX.
If I may get slightly off topic here and evoke part of the technical entry paart of my system (which is realy not all that important compared to the risk management aspect), I have 2 questions for all of you. 1) What do you ALWAYS find imediately after a top when looking at a candlestick chart ? 2) What do you ALWAYS find imediately after a bottom when looking at a candlestick chart ? ... no magic or anything, look hard
JJ, Just to correct a misconception you may have : You said "...But OK, if you increase your RISK based on volatility... I do NOT increase risk because of higher volatility... I just increase the stop distance.... and because of the greater distance on the stop, I reduce trade size in order to maintain the desired risk.
The principle here is that "gains must not only be larger that accumulated losses, but gains must also be accumulated at a faster rate that preceeding loses. Lets look at a simple (and grosly exagerated) example: FIXED RISK METHOD WITH A 5 TRADE LOSING STREAK : - lose 20 pips - lose 20 pips - lose 20 pips - lose 20 pips - lose 20 pips TOTAL: - 100 pips BREAK EVEN LEVEL: +100 pips OBJECTIVE TO RECUP AT LEAST EQUIVALANT OF 2X LOSES: +200 pips RISK INCREASE METHOD WITH A 5 TRADE LOSING STREAK : - lose 20 pips - lose 20 pips - lose 20 pips - lose 20 pips - lose 20 pips TOTAL: - 100 pips BREAK EVEN LEVEL: possibly far less than 100 (depending how risk was increased) OBJECTIVE TO RECUP AT LEAST EQUIVALANT OF 2X LOSES: possibly far less than 200 (depending how risk was increased) Because the break even level and target in our second example are not as far out, we can regain our highwatermark faster. This is important because after a bad losing streak using a fixed risk mothod, the break even level can become so far out that iy may take litteraly months just to break even. The counterpart, of course, using the risk increase method, is that you have limited capital and can not eternaly increase risk. This is why is is absolutely necesary to only use such a method with a conservative aproach. Personaly, my trades often begin risking only about 0.25% of my capital (see previous posts with EUR/NOK example). I cosider my risk "rather large" when risk on a given trade gets near 1.5% of capital. This leads us to the "rate of risk increase" problem (see previous posts).
Let me try to add something to this thread. Acrary in one of his very insightful posts talked about "drag" that can be introduced in an equity curve by reducing trade size after losses. As a result I always keep my trade size constant until after a new equity high, when I am free to increase or decrease my size. The basic premise is that if you go for a 10% profit target and a 10% stop, and you base your trade size on your equity you can introduce "equity drag" into the equation like so: (alternating 10% losses followed by 10% wins). 10,000 9,000 9,900 8,910 9,801 8,821 9,703 8,733 9,606 8,645 9,510 8,559 9,415 8,473 9,321 8,389 9,227 8,305 9,135 8,222 9,044 By keeping trade size constant you end up with 10,000 at the end of this series instead of 9,044 - which in my book is a big difference. By keeping trade size constant you do increase your risk after a loss. I haven't found a good justification for increasing trade size after a loss but I'm comfortable keeping trade size constant after a loss or series of losses to avoid the "equity drag" phenomenon.
I agree with this totally. The previous trades or your P&L have nothing to do with what is happening in the market right now and the best strategy for playing it. The P&L is obviously important but the best way to increase that is to forget the last trade in regards to it's P&L and just trade what is NOW keeping your risk management in sight.
To illustrate your example, you will have to be more precise. If you maintain costant size of trade, you are using proportionaly larger size trades after a loss. For example, 1K trade size of a 10K account = 10% trade size. After a 10% loss, account is 9K ... a 1K trade is now 11% trade size This point you have included in your example, ok, clear. The point you did NOT include in your example is the risk% on capital. with a starting capital of 10K, after a 10% loss, the account becomes 9K. Now, what are you going to risk on the next trade ? 10% of 10K (initial capital), or 10% of 9K (remaining capital)? If you risk 10% of 9K you will most likely have equity curve drag. If you risk 10% of 10K (which was your starting capital, but now is reduced to 9K), you ARE in fact increasing risk because 1K risk on 9K remaining capital is now a real risk of 11%.
See post on page 6 at 10-18-06 10:05 PM If you "forget the last trade in regards to its P&L" after a loss, your net profit will be significantly reduced unless your targets become significantly further out. Also, maintaining targets that are reasonably attainable is in fact part of managing risk.