Algo & Quant traders

SimpleMeLike · Apr 24, 2024

p0box4 said:
I am far from mad, because I know your opinion is bullshit.
More...

Hello p0box4,

I respect that you think my opinion is bullshit.

I highly respect. It's an open forum think however you wants.

It's all good man.

Let's go click man and get some money.

See yall later man ES about to open

SimpleMeLike · Apr 24, 2024

globalarbtrader said:
No, you have misunderstood what type of algo trading I am doing. My average holding period is a few weeks. I probably trade fewer times than you. I probably pay less in fees than you (about 0.5% of account value in commissions annually).

Everyone can see in (almost) real time what I am doing: I publish my daily trading here every night:
https://github.com/robcarver17/reports/blob/master/Trade_report

You can see that yesterday I did 4 trades, a bit below average, but it’s rarely more than 10.

So to summarise, you guys are no longer saying that quant and algo trading is nonsense and there is nobody making money out of it. I’m glad we agree on that. Instead you’re saying:

you lost money using algos, therefore as far as you are concerned all algos are bad

you’re only interested in having fun and gambling whilst trading, and algo trading just isn’t fun

algo trading is too much like hard work

Fine – I guess we all trade for different reasons. I trade to make money, since trading profits form a significant part of my income (I haven’t had a full time job for over 10 years). And I can make money far more consistently with algos than I can without. And I believe this is true of most people.

And we have different ideas about working hard. Personally having to sit in front of a screen to trade strikes me personally as way too much like hard work, and also very dull compared to the more intellectual task of researching and programming. Once a system is set up, it’s a lot less work than manual trading – perhaps a few minutes per day. But horses for courses.

We’re not going to disagree on this. It sounds like algo trading isn’t an option for you. But – to repeat – that doesn’t mean an algo trading is automatically a scam.

You also say two things that are wrong:

a non algo trader can make a guaranteed monthly profit and therefore needs less capital

algos have bigger drawdowns, and therefore need more capital

To make money every month consistently over long periods of time is unrealistic. Only a tiny fraction of professional traders manage it (outside of the high frequency trading arena, and outliers like Rentech). If you have done that, it’s likely you have just been lucky. As you say, you’re just a gambler.

There is no reason why – in expectation – the profitability and drawdowns should be bigger for an algo system. Quite the opposite. The median performance for an automated trading strategy is significantly higher than the median performance of a discretionary retail trader using technica analysis (which is negative). This also means you need less capital to achieve a given level of profitbaility. More capital is better because it provides diversification, which is good no matter what sort of trader you are. But a less diversified algo trader is almost certainly better than a less diversified random point and clicker.

Given the figures you have used, you are taking way too much risk with your single contract. Fine if you are trading for fun:

.. not fun if you aren’t.

GAT
More...

Good Morning globalarbtrader,

Thank you for the respond. I will review it later after work.

wxytrader · Apr 24, 2024

themickey said:
"I have no idea what quantitative approach means."
Let me give an example.
Over many years a trader will learn many patterns which work in certain circumstances or certain unpredictable times.
These could be based on prices (eg round numbers) or S/R Levels, could be to do with volume, speed of moves, days of the week month or year, maybe moving averages, could be to do with correlations, topping or bottoming patterns, breakout patterns, etc.....

So trader accumulates lots of ideas and knowledge and (s)he trades several different instruments, maybe has a watchlist of a couple of hundred stocks.

One can code much of their ideas into an amazing brain/algo. The code recognizes instantly across a wide range of trading instuments/watchlist when such events occur and alerts you.

Just think, you store all your edges into code. No need to manually try and recall. Even when you are tired or distracted, the algo can keep track of several hundred instuments and faithfully reminds you when events occur instantly.
Some algos go further and fully or semi automate the trades.

For SML, it appears he is limited to trading ES & NQ, but others can be trading several positions simultaneously, the more money which is on the table working over more instruments, the greater the potential profits (and losses). If you have a profitable system then profits will take care of losses.
It's only a working algo which can do this efficiently.
More...

So like finziv?

wxytrader · Apr 24, 2024

globalarbtrader said:
No, you have misunderstood what type of algo trading I am doing. My average holding period is a few weeks. I probably trade fewer times than you. I probably pay less in fees than you (about 0.5% of account value in commissions annually).

Everyone can see in (almost) real time what I am doing: I publish my daily trading here every night:
https://github.com/robcarver17/reports/blob/master/Trade_report

You can see that yesterday I did 4 trades, a bit below average, but it’s rarely more than 10.

So to summarise, you guys are no longer saying that quant and algo trading is nonsense and there is nobody making money out of it. I’m glad we agree on that. Instead you’re saying:

you lost money using algos, therefore as far as you are concerned all algos are bad

you’re only interested in having fun and gambling whilst trading, and algo trading just isn’t fun

algo trading is too much like hard work

Fine – I guess we all trade for different reasons. I trade to make money, since trading profits form a significant part of my income (I haven’t had a full time job for over 10 years). And I can make money far more consistently with algos than I can without. And I believe this is true of most people.

And we have different ideas about working hard. Personally having to sit in front of a screen to trade strikes me personally as way too much like hard work, and also very dull compared to the more intellectual task of researching and programming. Once a system is set up, it’s a lot less work than manual trading – perhaps a few minutes per day. But horses for courses.

We’re not going to disagree on this. It sounds like algo trading isn’t an option for you. But – to repeat – that doesn’t mean an algo trading is automatically a scam.

You also say two things that are wrong:

a non algo trader can make a guaranteed monthly profit and therefore needs less capital

algos have bigger drawdowns, and therefore need more capital

To make money every month consistently over long periods of time is unrealistic. Only a tiny fraction of professional traders manage it (outside of the high frequency trading arena, and outliers like Rentech). If you have done that, it’s likely you have just been lucky. As you say, you’re just a gambler.

There is no reason why – in expectation – the profitability and drawdowns should be bigger for an algo system. Quite the opposite. The median performance for an automated trading strategy is significantly higher than the median performance of a discretionary retail trader using technica analysis (which is negative). This also means you need less capital to achieve a given level of profitbaility. More capital is better because it provides diversification, which is good no matter what sort of trader you are. But a less diversified algo trader is almost certainly better than a less diversified random point and clicker.

Given the figures you have used, you are taking way too much risk with your single contract. Fine if you are trading for fun:

.. not fun if you aren’t.

GAT
More...

So how does your algo handle a drawdown if you are holding for 3 weeks? Average down?

globalarbtrader · Apr 24, 2024

wxytrader said:
So how does your algo handle a drawdown if you are holding for 3 weeks? Average down?
More...

No it will most likely reduce positions

GAT

wxytrader · Apr 24, 2024

globalarbtrader said:
No it will most likely reduce positions

GAT
More...

Hmmm reduce a position...never thought of that. I usually average down but perhaps some situations call for reducing, and then increasing at an even lower price. I never add to a winning position only losing positions...or as I refer to them as future winners.

Frederick Foresight · Apr 24, 2024

wxytrader said:
Hmmm reduce a position...never thought of that. I usually average down but perhaps some situations call for reducing, and then increasing at an even lower price. I never add to a winning position only losing positions...or as I refer to them as future winners.
More...

That 2008 Lambo you keep talking about resides only in your head, doesn't it?

SimpleMeLike · Apr 24, 2024

globalarbtrader said:
No, you have misunderstood what type of algo trading I am doing. My average holding period is a few weeks. I probably trade fewer times than you. I probably pay less in fees than you (about 0.5% of account value in commissions annually).

Everyone can see in (almost) real time what I am doing: I publish my daily trading here every night:
https://github.com/robcarver17/reports/blob/master/Trade_report

You can see that yesterday I did 4 trades, a bit below average, but it’s rarely more than 10.

So to summarise, you guys are no longer saying that quant and algo trading is nonsense and there is nobody making money out of it. I’m glad we agree on that. Instead you’re saying:

you lost money using algos, therefore as far as you are concerned all algos are bad

you’re only interested in having fun and gambling whilst trading, and algo trading just isn’t fun

algo trading is too much like hard work

Fine – I guess we all trade for different reasons. I trade to make money, since trading profits form a significant part of my income (I haven’t had a full time job for over 10 years). And I can make money far more consistently with algos than I can without. And I believe this is true of most people.

And we have different ideas about working hard. Personally having to sit in front of a screen to trade strikes me personally as way too much like hard work, and also very dull compared to the more intellectual task of researching and programming. Once a system is set up, it’s a lot less work than manual trading – perhaps a few minutes per day. But horses for courses.

We’re not going to disagree on this. It sounds like algo trading isn’t an option for you. But – to repeat – that doesn’t mean an algo trading is automatically a scam.

You also say two things that are wrong:

a non algo trader can make a guaranteed monthly profit and therefore needs less capital

algos have bigger drawdowns, and therefore need more capital

To make money every month consistently over long periods of time is unrealistic. Only a tiny fraction of professional traders manage it (outside of the high frequency trading arena, and outliers like Rentech). If you have done that, it’s likely you have just been lucky. As you say, you’re just a gambler.

There is no reason why – in expectation – the profitability and drawdowns should be bigger for an algo system. Quite the opposite. The median performance for an automated trading strategy is significantly higher than the median performance of a discretionary retail trader using technica analysis (which is negative). This also means you need less capital to achieve a given level of profitbaility. More capital is better because it provides diversification, which is good no matter what sort of trader you are. But a less diversified algo trader is almost certainly better than a less diversified random point and clicker.

Given the figures you have used, you are taking way too much risk with your single contract. Fine if you are trading for fun:

.. not fun if you aren’t.

GAT
More...

Hello globalarbtrader,

How many total systems are trading?

Thank you,

globalarbtrader · Apr 24, 2024

SimpleMeLike said:
Hello globalarbtrader,

How many total systems are trading?

Thank you,
More...

About 50 trading rule variations on around 200 futures

GAT

SimpleMeLike · Apr 24, 2024

globalarbtrader said:
No, you have misunderstood what type of algo trading I am doing. My average holding period is a few weeks. I probably trade fewer times than you. I probably pay less in fees than you (about 0.5% of account value in commissions annually).

Everyone can see in (almost) real time what I am doing: I publish my daily trading here every night:
https://github.com/robcarver17/reports/blob/master/Trade_report

You can see that yesterday I did 4 trades, a bit below average, but it’s rarely more than 10.

So to summarise, you guys are no longer saying that quant and algo trading is nonsense and there is nobody making money out of it. I’m glad we agree on that. Instead you’re saying:

you lost money using algos, therefore as far as you are concerned all algos are bad

you’re only interested in having fun and gambling whilst trading, and algo trading just isn’t fun

algo trading is too much like hard work

Fine – I guess we all trade for different reasons. I trade to make money, since trading profits form a significant part of my income (I haven’t had a full time job for over 10 years). And I can make money far more consistently with algos than I can without. And I believe this is true of most people.

And we have different ideas about working hard. Personally having to sit in front of a screen to trade strikes me personally as way too much like hard work, and also very dull compared to the more intellectual task of researching and programming. Once a system is set up, it’s a lot less work than manual trading – perhaps a few minutes per day. But horses for courses.

We’re not going to disagree on this. It sounds like algo trading isn’t an option for you. But – to repeat – that doesn’t mean an algo trading is automatically a scam.

You also say two things that are wrong:

a non algo trader can make a guaranteed monthly profit and therefore needs less capital

algos have bigger drawdowns, and therefore need more capital

To make money every month consistently over long periods of time is unrealistic. Only a tiny fraction of professional traders manage it (outside of the high frequency trading arena, and outliers like Rentech). If you have done that, it’s likely you have just been lucky. As you say, you’re just a gambler.

There is no reason why – in expectation – the profitability and drawdowns should be bigger for an algo system. Quite the opposite. The median performance for an automated trading strategy is significantly higher than the median performance of a discretionary retail trader using technica analysis (which is negative). This also means you need less capital to achieve a given level of profitbaility. More capital is better because it provides diversification, which is good no matter what sort of trader you are. But a less diversified algo trader is almost certainly better than a less diversified random point and clicker.

Given the figures you have used, you are taking way too much risk with your single contract. Fine if you are trading for fun:

.. not fun if you aren’t.

GAT
More...

Hello globalarbtrader,

I understand everything you are saying indeed. I run algos before and built them.

But look below please for the day so far.

Even though I took big hit earlier, look the next hour, I was able to recover the loss in the same day. I control my own destined to make the money for the pocket.

If I have algo, I can go drawdown for weeks or month.

Last year I run one algo and lost like $13,000, it is still in drawdown last I check. That was a VERY long painful year last year running that algo.

Last year I waste the entire year make Zero dollars on one algo, that is alot of work for nothing just to lose money. One year down the drain.

Now with me clicking manual I am more in tuned with the market in real time.

I can click as much as I want to ALL day long too. At least the way, I can control may day to day wins and losses.

With algo you have no control, you have to sit and wait win or lose.