Futures: Make $15,000 a day

Pekelo · Mar 21, 2009

Quote from TheCaymanIsland:

That is part of the risk that has to be decided so I think a starting point is determining what amount of return I should be able to generate on 1 contract and then extrapolate that out to hit the 15,000 goal.

Exactly. So you should have asked either a % (can I make X% a day?) or a per contract per day(can I make 3 ES per day?) question.

Throwing out a desired amount is meaningless because it really depends on the answer to the above questions and the aviable capital and the leverage used...

TheCaymanIsland · Mar 21, 2009

Quote from John Paterson:

Can you give an example of an "algorithmic futures platform"?

Do you mean something like moving averages and trend following or something like a high frequency scalping algorithm?

It's like going to a car forum and telling people you want to buy a good car.

Moving averages and trends

infolode · Mar 21, 2009

Quote from TheCaymanIsland:

Moving averages and trends

You might want to rethink that one.

TheCaymanIsland · Mar 21, 2009

Quote from Pekelo:

Exactly. So you should have asked either a % (can I make X% a day?) or a per contract per day(can I make 3 ES per day?) question.

Throwing out a desired amount is meaningless because it really depends on the answer to the above questions and the aviable capital and the leverage used...

Okay, so if I trade 1 future contract per day, what % can I expect to make using an algorithm that uses moving averages and trends? Available capital and leverage are basically unlimited.

ASusilovic · Mar 21, 2009

Quote from TheCaymanIsland:

Okay, so if I trade 1 future contract per day, what % can I expect to make using an algorithm that uses moving averages and trends? Available capital and leverage are basically unlimited.

Usually, traders / managers are paid 2/ 25 / 35 / 44 answering these questions...

rluser · Mar 22, 2009

Quote from TheCaymanIsland:

Okay, so if I trade 1 future contract per day, what % can I expect to make using an algorithm that uses moving averages and trends? Available capital and leverage are basically unlimited.

It sounds very much like the percentages are unlimited.

Pekelo · Mar 22, 2009

Quote from TheCaymanIsland:

Okay, so if I trade 1 future contract per day, what % can I expect to make using an algorithm that uses moving averages and trends? Available capital and leverage are basically unlimited.

The question is still unclear. My response was about the return per contract, not using 1 contract. It is really a simple math question, knowing:

1. Average daily return per contract.

2. Capital in the account.

3. Leverage used.

You change any of these 3, the answer changes to your question. Let's say you make average 3 ES per contract on a 10K account using 2 contracts, that is 3% daily return. If you start using 3 contracts, the daily return goes up 50%, so it will be 4.5%.

If you decrease leverage by either adding more money but not increasing the number of contract used, or decreasing the number of used contracts but not changing the capital the daily return in % will drop...

We CAN NOT know, what kind of average per contract return your algorythm can do. You are the only one who knows that, and if you have a desired % return in your mind (or a fixed dollar amount), you just adjust the leverage part of the equation and you will get how many contracts you need to use to reach that goal.

So let's say the algorythm does make 3 ES per contract daily. That is $150 but since you wanted 100 times that, you need to use 100 contracts to make 15K daily. The risk part will be decided by the leverage..

John Paterson · Mar 22, 2009

Quote from intradaybill:

15K is a realistic amount for a hedge fund. In my experience (2 years trading for a hedge fund in the past), depending on what you trade, you should assume a maximum annual return of 20%. This means you need to have a minimum capital to start with of about 20 Million. If you make more, that is fine.

Now, as far as contracts per day, you need to design a trading algorithm that will risk no more than 1% on each trade and generate enough trades to reach your goal. This is to keep your risk of ruin low.

For example, let us assume you are trading Crude futures and you have an objective of $0.5 per trade and a stop loss of $1 per trade. With 20 Million you can risk 0.01 x 20,000,000 or an amount of 200,000 on each trade and this means you need to buy or sell 200 contracts each time, since 200 x $1,000 = 200,000

Thus, each time you win you make 100,000 and each time you lose 200,000. If you can maintain that, then your expectancy per trade, assuming you can make a trade per day, will be 15,000 if

15,000 = w x 100,000 - (1-w) x 200,000

Solving for w we get: 71.6%

So you need to have a win rate of more than 71% to get your average daily return.

If your trade every 2 days, then you need to have a higher win rate so that your expectancy is 30,000 per two days,

w = (30,000 +200,000)/300,000 = 76.66%

As you can see, you need to be more profitable and this becomes a problem as the frequency of trades decreases.

In general, if your trading frequency is N, then

N x E = avg. win x w - (1-w) x avg. loss

I suggest you read this paper and also this one before you start doing anything. They can save you a lot of time and trouble.

Good post. The answer for the OP is that he must know the expectancy of his trading system per contract traded, assuming on a daily basis. He can then divide 15K by that to estimate the number of contracts needed to achieve that objective.

Leverage does not matter as long as his capital is large enough to cover margin plus maximum intraday drawdown. He claims to have unlimited capital so these issues are irrelevant.

ammo · Mar 22, 2009

i suggest in your tests you break the day into 3 parts and if you are up ,the nite session,the es performs differently at different times of the day,very low volume choppy from 11-12:30 cst,higher risk in this time zone,quicker 3-5 point moves 8-9:30,and at the end of the day we can see some 15-20 point moves, those 3 timeframes each give you a different perspective on your p/l and the amount of size to put on Are you saying they have a program that is drawing tl's for u and you are using agreeable ma's to get into and out of trades?

TheCaymanIsland · Mar 25, 2009

Quote from Pekelo:

The question is still unclear. My response was about the return per contract, not using 1 contract. It is really a simple math question, knowing:

1. Average daily return per contract.

2. Capital in the account.

3. Leverage used.

You change any of these 3, the answer changes to your question. Let's say you make average 3 ES per contract on a 10K account using 2 contracts, that is 3% daily return. If you start using 3 contracts, the daily return goes up 50%, so it will be 4.5%.

If you decrease leverage by either adding more money but not increasing the number of contract used, or decreasing the number of used contracts but not changing the capital the daily return in % will drop...

We CAN NOT know, what kind of average per contract return your algorythm can do. You are the only one who knows that, and if you have a desired % return in your mind (or a fixed dollar amount), you just adjust the leverage part of the equation and you will get how many contracts you need to use to reach that goal.

So let's say the algorythm does make 3 ES per contract daily. That is $150 but since you wanted 100 times that, you need to use 100 contracts to make 15K daily. The risk part will be decided by the leverage..

Thanks Pekelo. So let's say I'm trading ES and NQ Futures. I have 500,000 in the account and I expect to bring in an average of $50.00 a contract. Then what sort of risk parameters should I set up if I expect to make $15,000 a day. I should obviously expect to trade 300 contracts. Is the $50 a reasonable assumption though for an algorithm?