Numbers/Percentages/Systems

[bangkoker]

Numbers/Percentages/Systems/Waffle

OK first off, my data does not go back far enough to be properly statistically significant as of yet but thoughts appreciated.

I am using a daily system, it is so far winning 47% of time.. erm doesn't sound great so far does it?

losing 53% (logically).

When it wins on average I make : 39 points
When I lose on average I make : -17 points

Therefore averaging circa 45 points a week or 9 a day (because presumably a little less than half of 39-17, half of the time is 9ish).

Am I wildly off here or is there some logic in it? Does anyone know what I mean?

Presumably if it maintained those numbers it wouldn't be a bad system (in my mind anything with a net profit has to be good,as can have multiple system uncorrelated and running different times of the day/markets etc)as half the time I am winning, I win twice as much as when I am losing... mathematically that makes sense.. although I am not happy about the fact I lose more than half the time.

Feel free to flame me, or join in on the thread. Would welcome views

 

[tomorton]

So far so good.

 

[DHB]

It's all about expectancy. If (win prob * avg. win size) - (loss prob * avg. loss size) > 0, then you basically have an edge, and over the long run, a profitable system.

Check out this random equity curve simulator:

http://www.hquotes.com/tradehard/simulator.html

With an avg. win size of 39, and an avg. loss size of 17, your win/loss ratio is 39/17, or 2.29. Use that number as well as the win probability of 0.47 and see what your system does over the long run. Make sure you generate a couple of equity curves, but provided that your parameters will hold over the long run, you've got yourself a great edge.

 

[bangkoker]

It's all about expectancy. If (win prob * avg. win size) - (loss prob * avg. loss size) > 0, then you basically have an edge, and over the long run, a profitable system.

Check out this random equity curve simulator:

http://www.hquotes.com/tradehard/simulator.html

With an avg. win size of 39, and an avg. loss size of 17, your win/loss ratio is 39/17, or 2.29. Use that number as well as the win probability of 0.47 and see what your system does over the long run. Make sure you generate a couple of equity curves, but provided that your parameters will hold over the long run, you've got yourself a great edge.

Thank you, that is the kind of thought provoking answer I am after. I still have a lot to learn but with money management/patience/steady learning and a market that is favourable to my system and other styles I may trade I hope I will be in it for the long haul. and thanks again for the reply and link, interesting :clap: