Lost me. I was looking here http://www.bjmath.com/bjmath/kelly/kelly.htm to see if I could figure out what you're talking about, but nope I can't. To me the risk parameters as listed in the spreadsheet are "the deal" (and actually I was being generous with that 4 pt stop, normally I can tell long before then that a trade is going against me). Best Regards, JJ
Finally a topic for me. I increment after a loss and have found it a very good way to smooth the account equity. A few notes after years of working with it: 1). Know your pain threshold for drawdowns before figuring the progression. If you can't stand a 10% drawdown you'll need a different progression than a person that can stand a 40% drawdown. 2). Plan on hitting multiple max drawdowns in a row and incorporate that into your pain threshold. If you can only stand a total drawdown of 10% then your progression might stop at 3% max drawdown points. This allows you to have 3 failed progressions before you stop trading. 3). Figure the % chance of hitting your max drawdown per progression and keep the % to less than 5%. If you're hitting it more than that then your progression is too aggressive. Also, it can be pretty demoralizing to hit a max drawdown only to start with a new progression and head for the max again. 4). Once you hit a max drawdown, stop trading and check your strategy to make sure it's still valid. Wait for confirmation with winning paper trades before putting more money at risk. 5). It's easier on your mind if the strategy you're using with this has a high win %. If you only win 30% of the time the progression has to be very slow. Also with a low win rate it can be hard to figure out when the strategy has stopped working. With a high win rate you can be more agressive and it's pretty easy to tell when the method is broken. 44yotrader
My model involves increasing risk after a losing trade proportionaly to the increase of a volatility range. Therefore, for example, if I risk 0.2% at the begining of a "series", and then the market stops me out and increases my volatility range by X, my next trade might be risking say 0.243%. The increase of the risk takes market volatility into account, and is therefore not linear. For example, I have been having a difficult time attacking the short side of EUR/NOK lately and have been stoped out 4 consecutive times, and am presently short again. The risk (in terms of % of capital) on these five trades was : 0.2% 0.2045% 0.2099% 0.2269% 0.2449% My break even level is now 8.3540 EUR/NOK, and I have an order to close 25.87% of the position at 8.2420 which make up for 2x my preceeding accumulated losses. This is quite a conservative model. 2005 gains were a bit over 25% and this suite me fine.
Also, just to give you an idea of what my model can, and has, sustained, this summer I had 16 consecutive losses on gbp/jpy and came out with an absolute profit. At the worst point, my account sustained a draw down of just over 5%. ... it was a bit of a painful period for me, but I was able to sleep at night knowing there was very litle chance of me blowing up.
The problem that getting bigger after losers/drawdown is that often you begin to rely on this alone to "get you back", instead of focusing on improving your entries. All you do by increasing size is increase exposure to both loss and gain, it may "appear" that increasing size helps to speed up gains but that only "works" until it doesn't -- and that's when losses can accelerate.
Wow, you've leap-frogged ahead of me on a few of your concepts here. While I figured your model would take volatility into consideration, and I also figured you would look at different risk levels (or stop placement), I didn't think you would combine them both together. But OK, if you increase your RISK based on volatility, do you also increase your PROFIT TARGET based on VOL? Regards, JJ
seems to me this this the benefit and tradeoff. increasing size after a loss gets you back from drawdowns quicker, but it means you will have larger and more frequent drawdowns.
One of the principles on which my system is built is that I assume I don't know where the market is going to go. I have a simple technical method of assigning entries. In consequence, I don't try to "get better entries" (furthermore, I am of the oppinion that one would have to have some supernatural power to see the furure to get "better entries" as we don't know if our entry is good or not until after the fact). As far as size is concerned, you missed a point. Risk can be incrased without increasing size. Again, risk can be determined not only by size of trade, but also by distance of stop (read previous posts).