This part of the system, that is the portion that relates to risk of massive divergence from the first entry position is clearly the location of the biggest risk. As someone who uses a system similar than this and has suffered a black swan I have come to believe in creating a risk matrix that explains up front the maximal drawdown that is sustainable for my arb. It was a hard lesson learnt but one that will stay with me. A great place to start is to look at the history over the past x years and establish a range so you know the playing field.
So, EURCHF ranges by lets say 2700. We'll ignore the fact for now that it very well could go further. A simple spreadsheet will show you the position size that should be used with your capital so that a maximal drawdown event does not blow your account. Spieler accounts for this by 'working the system' and from what he has said, he is willing to cut a loss on a position by using the winnings from other position within that 'trade'. He also diminishes the % of DD of that bad position by continuing to trade in the face of a DD.
One risk of this system is that you need to keep careful trace of the total % renormalisation that has occurred at each daily roll from the inception of each trade. This can easily be calculated in an EA providing you mark trades as pairs in some way (magic number, comment, date-time correlation).
Spieler, something that I did was create a table of statistics of the locations within the spread where statistically the price is likely to be. After a while hotspots will appear. I used that to help create a matrix of position sizes (apportioned of the maximal drawdown even described earlier) so that I could use larger position where I thought I would make the most money and smaller positions where the market rarely moves to. The benefit here is obvious, though there of course is the risk for the market to just not play where I it has in the past and subsequently not generate above average returns. (I can of course rebalance my position size matrix any time the arb passes zero when all of the positions close.
Also, as advice to other here, I suggest you write some little scripts that open positions with a size you nominate up front, say a % of your account. I think spieler suggests 1%+1.5% position sizes. It's terribly easy to greed and not enforce strict risk management when you are on a winning streak. The black swan in these systems has the same probability of occurring every day though despite your winning history. (Unless you cut your losses in some way.)
So, EURCHF ranges by lets say 2700. We'll ignore the fact for now that it very well could go further. A simple spreadsheet will show you the position size that should be used with your capital so that a maximal drawdown event does not blow your account. Spieler accounts for this by 'working the system' and from what he has said, he is willing to cut a loss on a position by using the winnings from other position within that 'trade'. He also diminishes the % of DD of that bad position by continuing to trade in the face of a DD.
One risk of this system is that you need to keep careful trace of the total % renormalisation that has occurred at each daily roll from the inception of each trade. This can easily be calculated in an EA providing you mark trades as pairs in some way (magic number, comment, date-time correlation).
Spieler, something that I did was create a table of statistics of the locations within the spread where statistically the price is likely to be. After a while hotspots will appear. I used that to help create a matrix of position sizes (apportioned of the maximal drawdown even described earlier) so that I could use larger position where I thought I would make the most money and smaller positions where the market rarely moves to. The benefit here is obvious, though there of course is the risk for the market to just not play where I it has in the past and subsequently not generate above average returns. (I can of course rebalance my position size matrix any time the arb passes zero when all of the positions close.
Also, as advice to other here, I suggest you write some little scripts that open positions with a size you nominate up front, say a % of your account. I think spieler suggests 1%+1.5% position sizes. It's terribly easy to greed and not enforce strict risk management when you are on a winning streak. The black swan in these systems has the same probability of occurring every day though despite your winning history. (Unless you cut your losses in some way.)