Money management from a Math God? Id like one

Recently I downloaded the history of CABLE for the hourly. I got 3630 figures which work out to be 151 days after I subtracted out new years and Christmas as well as the weekends from the data for non trading days. I did a bunch of IF() functions to see what type of results I would have if I tried to base my trades on certain things. You can probably easily figure out what I did with what I am about to tell you:

Winning/Breaking Even:
I found that 85.5% (+/ - 3.8%) of the time I at least break even with 4 pips. The “potential gain” ranges from 0 to 155 and works almost exponentially from 155 to 0 meaning that the lower values are more likely to happen than the higher values. The average is 19. The median value is 7. This means 50% of the 85.5 are above 7 pips and 50% are below.

Losing:
I found that 14.5% (+/- 3.8%) of the time I lose. Losers range from -1 to -155 with a median value of -10 and an average of -14. I noticed that these loses are also almost exponential growing from high to low Noted by the median value. (Very small large loses, a lot of small loses.)

I concluded that as I attempt to make a profit larger than 0, the odds of winning become less and obviously my losses will rise. I am looking for someone that can help me analyze the data to come up with some sort of money management to take advantage of the range I have given. I’d like to talk through MSN ([email protected]). So if I told you 85.5% of the time, I can at least break even which is about 8% of that. I want to be able to maximize the profits! If you have any questions concerning more percentages, throw your question here. I will check this thread every day.



-Liquid11

I could send you 1,000 consecutive days of hourly data if you like, 151 days that's not enough to draw any conclusion.

You will have to explain that sentence again, I am not sure what you mean by that exactly.


I have already talked about that almost magical number (155) in another thread. It seems to be appearing in every backtest I perform on the GBP/USD.

This idea is this: if you are a day trader and you notice that you have already earned 155 pips during the day, the best thing to do is to take your profits and close your position, you have already reached THE OPTIMUM take profit possible, there is no more "juice" left.

It all depends on your system of course but this 155 pip limit keeps showing up again and again in almost every backtest I do, at least for GBP/USD.

Id love that!!


I just mean that im using a spread of 4 pips per trade. If my trade doesnt move in my direction atleast 4 pips to cover, its a loser. If it does move 4 pips, its a winner. Just for the fact that it broke even, i consider it a winner.




155 pips is used because it is an actual data value. I'm worried that this value is bringing up the averages and making my whole test look promising. No where am i looking for a 155 pip win. If it happens, it would have to be with some sort of money management that exponentially closes positions in the same fashion that my winners showed pip profits(smaller wins close bigger positions, larger wins close very tiny positions). Its hard to make sense without at chart. I am going to make a chart later to show you what i mean by exponentially. If you can send that history, id be very appreciative! Thanks for the quick response FXT!

-Liquid11

No problem, I just emailed the hourly data for GBP/USD from 10/22/01 to 05/14/06 to your hotmail email account.

If you need more data just whistle 3 times

Happy backtesting.

PS: Please note that the prices are in this format, example 1.4349 is written as 14349, I backtest with C++ and floating variables do not always give exact values so I use the integer type instead. But if you like to get the data in another format let me know.