Increasing Size During Drawdown

Quote from TheMagican:

Thanks for your clarificantion.I see what you mean.Honestly, i never thought about it,but i can say,that the EXPECTANCY of my model may vary and i can`t say beforehand what it would be.But in majority of the cases,i can say, it`s something around 100%,or higher.

But what is the use of it anyway?It`s not a good idea to artificially maintain the expectancy within the framework of a certain percentage.What if an expectancy for the next trade would be 99%(rough example)?You won`t take this trade?
More...

Well, the expectancy of EVERY model varies. You still seem to be missing the point. A trader doesn't artificially maintain the expectancy within a framework. The trader always tries for peak profitability (which is a function of expectancy, size, and turnover), but he should have some type of minimum expectancy rule. At the very least, the trader should know that a trade offers at least a slight positive expectancy, otherwise he is really just guessing.

I can almost guarantee that your gross expectancy (i.e. under all market conditions) isn't even close to 100% or you wouldn't be claiming to be under-capitalized. Another quick way to tell, is to calculate your beta and alpha. If your beta isn't zero with very high alpha, then your expectancy probably isn't even above 10%. But the point still is that you don't even have any idea what it is.
 
Quote from Epic:

Determining historical expectancy is very simple.

Expectancy = (Probability of Win * Average Win %) – (Probability of Loss * Average Loss %)

Example I

You are 50:50 winners and losers.
Average win = 2%
Average loss = 1%

(50*2)-(50*1) = 50

Expectancy +50%
IOW, for every dollar risked you expect to make 50 cents. In the real world, a 50% expectancy day trading system is VERY rare. Whomever owns it would be a very wealthy person.

Example II

You are 90:10 winners and losers
Average win 1%
Average loss 10%

Expectancy -10%
This is a losing system over time.

So the expectancy of your total history is easy to determine, just remember to figure it net of all costs. The harder part is to break it out into different types of trades and to determine what the expectancy is going to be for any potential trade. To do this, you must accurately determine the likelihood of all the various outcomes. I'm not willing to describe how to do that here, and it would take too much time anyway.

The point is that a huge majority of traders and managers really don't know whether they have a positive or negative expectancy over time. They just do what seems to work until it doesn't work any more and then they whine that the market changed. The reality is that their system had a long term negative expectancy and they didn't know it.
More...
thanks Epic, you really addressed the original question I was asking, namely position size. I'll research it more. It would be wonderful to have the whole thing reduced down to numbers.
 
Quote from oldtime:

thanks Epic, you really addressed the original question I was asking, namely position size. I'll research it more. It would be wonderful to have the whole thing reduced down to numbers.
More...

Yes, that's what I've been trying to explain this whole time. Whether to increase size depends completely on the effect that increase would have on the trade profitability.

Take it one step further. Profitability is essentially just a simple equation.

Expectancy X Size X Turnover

Size = % risk per trade
Turnover = number of round trips per day/week/month

Example I

Expectancy = +15%
Size = 1%
Turnover = 1 round trip per day (2 trades per day)

Expected annual profit = (0.15 * 0.01 * 1) * 250 = 37.5%

That is of course assuming the ability to compound daily.



Example II

Expectancy = +30%
Size = 1%
Turnover = 1 round trip per week

Expected annual profit = 15.5%


So now you see that profitability isn't just about expectancy either. In the above examples, the one with lower expectancy actually resulted in greater profitability. Your goal isn't to maximize expectancy, but to maximize profitability. This will end up being a function sort of like a heavily skewed bell curve. Typically, as expectancy increases, turnover decreases because there are just fewer really good trades. But also, as expectancy approaches 0, trading costs start to offset small profits.

You just need to figure out where the peak of that curve is for your strategy. The peak of that curve will tell you what expectancy to look for. From there, trade size is determined based on the maximum allowable drawdown. :D
 
Quote from Epic:

Determining historical expectancy is very simple.

Expectancy = (Probability of Win * Average Win %) – (Probability of Loss * Average Loss %)

Example I

You are 50:50 winners and losers.
Average win = 2%
Average loss = 1%

(50*2)-(50*1) = 50

Expectancy +50%
IOW, for every dollar risked you expect to make 50 cents. In the real world, a 50% expectancy day trading system is VERY rare. Whomever owns it would be a very wealthy person.

Example II

You are 90:10 winners and losers
Average win 1%
Average loss 10%

Expectancy -10%
This is a losing system over time.

So the expectancy of your total history is easy to determine, just remember to figure it net of all costs. The harder part is to break it out into different types of trades and to determine what the expectancy is going to be for any potential trade. To do this, you must accurately determine the likelihood of all the various outcomes. I'm not willing to describe how to do that here, and it would take too much time anyway.

The point is that a huge majority of traders and managers really don't know whether they have a positive or negative expectancy over time. They just do what seems to work until it doesn't work any more and then they whine that the market changed. The reality is that their system had a long term negative expectancy and they didn't know it.
More...

Let`s try it.

Ave win = 0,18%
Ave loss = 0,12%

89:22 Winners and Losers

(89*0,18) - (22*0,12) = 16,02 - 2,64 = 13,38

Expectancy of the model is +13,38%

Am i correct?

Where is next?I still don`t get it whats in it for me,if i know how to deduct and multiply:confused:

Please share yor wisdom.
 

Attachments

Quote from TheMagican:

Let`s try it.

Ave win = 0,18%
Ave loss = 0,12%

89:22 Winners and Losers

(89*0,18) - (22*0,12) = 16,02 - 2,64 = 13,38

Expectancy of the model is +13,38%

Am i correct?

Where is next?I still don`t get it whats in it for me,if i know how to deduct and multiply:confused:

Please share yor wisdom.
More...

The biggest loss (-$4K)occured on Wed20

I added to the postition once and closed at the end of the session.If i kept adding,i`d blow my ass.
 

Attachments

But it seems that NT blotter produces the complete bull crap in the statement.For e.g.,how can it be 13,95% of the Cumulative Profit if the initial amount at the begining was $40K?Also,Ave winning trade of 0,18% is 72 for $40K.Or what are those numbers related to?:confused:

Someone apparently dumber then f.k.Either the platform developers or myself.:D
 
Quote from TheMagican:

Let`s try it.

Ave win = 0,18%
Ave loss = 0,12%

89:22 Winners and Losers

(89*0,18) - (22*0,12) = 16,02 - 2,64 = 13,38

Expectancy of the model is +13,38%

Am i correct?

Where is next?I still don`t get it whats in it for me,if i know how to deduct and multiply:confused:

Please share yor wisdom.
More...

Now you have a slightly more realistic picture of where you've been until now. You know that in aggregate you expect to make $13.38 for every $100 risked. Like I said before, that is the easy part, but at least now you know.

Let's take it step-by-step. Determine your expected annual ROR. Answer this question.

What was your turnover? IOW, how many round trips (open then closed trades) do you get in an average month? We aren't just looking for average holding period here. Average holding period is useless at this point.
 
Quote from Epic:

Now you have a slightly more realistic picture of where you've been until now. You know that in aggregate you expect to make $13.38 for every $100 risked. Like I said before, that is the easy part, but at least now you know.

Let's take it step-by-step. Determine your expected annual ROR. Answer this question.

What was your turnover? IOW, how many round trips (open then closed trades) do you get in an average month? We aren't just looking for average holding period here. Average holding period is useless at this point.
More...

111 trades over the month.Though,the numbers for Ave Win/Ave loss incorrect.That blotter delivers complete crap instead of the accurate statement.But let`s try it manually.
 
Quote from TheMagican:

111 trades over the month.Though,the numbers for Ave Win/Ave loss incorrect.That blotter delivers complete crap instead of the accurate statement.But let`s try it manually.
More...

111 trades? Or 111 round trips?
 
You must log in or register to reply here.
Back
Top