kevinkdog 

Look for small margin markets. There are a bunch of them, especially now with micros. Treat the beginning year or two as a great learning time, with limited risk. Can't lose 6 figures if you don't have it to start! Treat it as a marathon, not a sprint. Get returns slowly, not get rich quick. If you see things going well, look for alternative capital sources to increase funds. I used to take on programming free lance work to help fund my trading. Especially early on, when trading was a losing effort for me.

Thank you so much kevinkdog.

I like the "Can't lose 6 figures if you don't have it to start!". Correct.

Question:

1. With the profits earned from the small portfolio, would you encourage or recommend use those profits to increase contracts via position sizing methods in the current systems, or use profits to get more systems to build bigger portfolio of systems?

goodoboy 

Thank you so much kevinkdog. I like the "Can't lose 6 figures if you don't have it to start!". Correct. Question: 1. With the profits earned from the small portfolio, would you encourage or recommend use those profits to increase contracts via position sizing methods in the current systems, or use profits to get more systems to build bigger portfolio of systems?

Depends what your objective is. Aggressively adding size to 1 strategy can exponentially increase your account, but it also increases your chance of losing big. Adding more strategies can make overall overall curve smoother, at the expense of foregoing outsize gains.

kevinkdog 

Depends what your objective is. Aggressively adding size to 1 strategy can exponentially increase your account, but it also increases your chance of losing big. Adding more strategies can make overall overall curve smoother, at the expense of foregoing outsize gains.

Good Morning Kevinkdog,

Thank you for the response.

Your response is good and clear.

My only objective in trading is to build a portfolio of 10 Algos (5 on the ES, and 5 on the NQ), run the portfolio for the next X-XX years and using a position sizing method on each strategy to increase to max contracts over time as the cumulative net profits for each strategy grows. My objective is not for income needs, but for wealth building similar to Dollar Cost Average into 401K.

Is this obtainable in your opinion? Do you teach this type portfolio building where the objective is clear and precise?

Thank you

goodoboy 

Good Morning Kevinkdog, Thank you for the response. Your response is good and clear. My only objective is to build a portfolio of 10 Algos (5 on the ES, and 5 on the NQ), run the portfolio for the next X-XX years and using a position sizing method on each strategy to increase to max contracts over time as the cumulative net profits for each strategy grows. My objective is not for income needs, but for wealth building similar to Dollar Cost Average into 401K. Is this obtainable in your opinion? Do you teach this type portfolio building where the objective is clear and precise? Thank you

It is not appropriate for me to discuss teaching, since that would be self promotion.

So you want 5 ES strategies and 5 NQ strategies...

How many algos have you built so far?

How will you determine if these 5 ES and 5 NQ strats reduce your risk, as opposed to just amplifying it?

kevinkdog 

It is not appropriate for me to discuss teaching, since that would be self promotion. So you want 5 ES strategies and 5 NQ strategies... How many algos have you built so far? How will you determine if these 5 ES and 5 NQ strats reduce your risk, as opposed to just amplifying it?

Hello kevinkdog,

Thank you and understandable.

kevinkdog 

So you want 5 ES strategies and 5 NQ strategies...

This is correct. ES and NQ are liquid markets, my objective is one day all strategies is trading at XX contracts each. 2ES, 2NQ, 2GC, 2CL is an idea as well.



So far I have built 2 MES, and 1 MNQ. I am running micros because I do not have enough capital for full emini. Right now the MNQ is at 4 contracts, as profits is increasing I increase contracts on it and decrease as profits decrease, manual effort here weekly.

Both MES is in drawdown.

kevinkdog 

How will you determine if these 5 ES and 5 NQ strats reduce your risk, as opposed to just amplifying it?

I will confirm all 10 strategies are not correlated with each other. I am not sure I understand the question, sorry? Can you repeat it please?

goodoboy 

Hello kevinkdog, I will confirm all 10 strategies are not correlated with each other. I am not sure I understand the question, sorry? Can you repeat it please?

Correlation is one way to look at it, but there are other methods too.

Some questions to ask yourself about correlation:

What R^2 should I be under for "non correlation" ?

How do I look for correlation? Do I use trade results, daily performance results, weekly, monthly?

What recent history should I use to calculate correlation: the last week, last month, last year, whole backtest?

How do I account for the fact that 2 upward sloping equity curves will be fairly well correlated, simply because they are both making profits over time?


In other words, aiming for 5 ES and 5 NQ strategies brings up a lot of questions, beyond the actual building of the strategies themselves (which most people do incorrectly).

kevinkdog 

Correlation is one way to look at it, but there are other methods too. Some questions to ask yourself about correlation: What R^2 should I be under for "non correlation" ? How do I look for correlation? Do I use trade results, daily performance results, weekly, monthly? What recent history should I use to calculate correlation: the last week, last month, last year, whole backtest? How do I account for the fact that 2 upward sloping equity curves will be fairly well correlated, simply because they are both making profits over time? In other words, aiming for 5 ES and 5 NQ strategies brings up a lot of questions, beyond the actual building of the strategies themselves (which most people do incorrectly).

Hello kevinkdog;

Thank you for the response.

kevinkdog 

What R^2 should I be under for "non correlation" ?

I am not sure what this mean.

kevinkdog 

What recent history should I use to calculate correlation: the last week, last month, last year, whole backtest?

I use two softwares to tell me if the systems are correlated, this is good enough.

kevinkdog 

In other words, aiming for 5 ES and 5 NQ strategies brings up a lot of questions,

Yes, I would think the only questions is surround around correlation or similar strategies right?

goodoboy 

Hello kevinkdog; Thank you for the response. I am not sure what this mean. I use two softwares to tell me if the systems are correlated, this is good enough. Yes, I would think the only questions is surround around correlation or similar strategies right?

1. My point: Above what R^2 value is considered correlated?

2. If you say so. I see correlation as a lot more nuanced than just a correlation coefficient, but maybe that is just me.

3. If you want to just rely on the correlation results of your two software programs, without any further analysis, then I guess you have covered all your questions.


Something to think about: what if you have 2 ES strategies that historically show very low correlation? How much do you trust that low correlation to hold in the future, and how might that impact your position sizing decisions?

kevinkdog 

1. My point: Above what R^2 value is considered correlated? 2. If you say so. I see correlation as a lot more nuanced than just a correlation coefficient, but maybe that is just me. 3. If you want to just rely on the correlation results of your two software programs, without any further analysis, then I guess you have covered all your questions. Something to think about: what if you have 2 ES strategies that historically show very low correlation? How much do you trust that low correlation to hold in the future, and how might that impact your position sizing decisions?

Hello kevinkdog,

Thank you for the response. sorry I am late responding.

I do not know how low the correlation will be in the future. I just do not know that answer.

If in the future 2 ES strategies start to correlate with each other, they just correlate with each other, and I increase or decrease position size as time goes on.

Correlation is not what really concerns me, complete system failure is what I worry about.

sbrad39 It sounds like we tackling the same problem. I would love to hear the progress you make on this decision and what you are currently doing.

I have only practiced day trading the ES and NQ but I think I would rather swing trade over several days or even hold for longer. I am just wondering what timeframes I should base my trades on.

It would see to me that any chart timeframe other than the daily would require you to be watching the market all day long, correct?

My question to experienced futures traders: If I am holding a futures position overnight with a stop-loss set, is my fear of a black swan event that leads to a major loss beyond my stop-loss rational? Obviously many hold overnight but for some reason this fear is a mental block for me. Are some of you "no way in heck would I ever hold a futures position overnight" people?