Fully Automated - LIVE Trading - $50k per month

frostengine said:
Today's results:
Daytrading Strategy: -$6,297
More...

You were up several thousand but ended up -6K. So you lost about 10K on your Long trades, losses which seem to be have closed near the market close.

eg. You were down 200 NQ points in MNQ positions i think, but not stopped out?

Did these Long trades have any sort of worst case stop loss in place?
 
Millionaire said:
You were up several thousand but ended up -6K. So you lost about 10K on your Long trades, losses which seem to be have closed near the market close.

eg. You were down 200 NQ points in MNQ positions i think, but not stopped out?

Did these Long trades have any sort of worst case stop loss in place?
More...

No, none of the strategies use any stop losses. A 10k swing is only 1.25% move in the account.
 
frostengine said:
No, none of the strategies use any stop losses. A 10k swing is only 1.25% move in the account.
More...


Maybe you're already aware of this, judging by your comments it seems you are, but I would just be extremely careful about sizing up. It looks like one of your biggest edges is your account size not your actual strategies, which is allowing you to trade smaller sizing relative to your account size, but still generate good profits (relative to baseline cost of living and etc).

For example smallest size on futures you can trade is a micro. So, someone with a 10k account even if they allowed 1.25% drawdown on some trades, would only give them 62.50 points of stop. Not even 1/3rd of the stop you took (and still didn't exit).

Not trying to say you're not smart enough to figure this out on your own, but there's been some people with large account sizes come through here lately and blow through it quickly making obvious errors. You have 20 years of experience, so less likely to happen. Still just thought I'd throw this out there.
 
Last edited:
frostengine said:
No, none of the strategies use any stop losses.
More...

When doing directional trading they say it is important to "Ride winners and cut losers short".

All my research suggests this rule is correct. It is very hard to maintain a trading edge over the long term edge without doing this.

But your systems might be the exception, good luck!
 
Millionaire said:
When doing directional trading they say it is important to "Ride winners and cut losers short".

All my research suggests this rule is correct. It is very hard to maintain a trading edge over the long term edge without doing this.

But your systems might be the exception, good luck!
More...

In all the systems I have designed, I have never seen an arbitrary "stop loss" improve the results I care about. All of my systems are designed to have a "reason" on why it gets out. This "reason" may be a black box as far as my understanding of it, but the models were trained to create that reason and the data supports the reason. This has shown to be far superior for my style of automated trading.

This goes back to an edge that has been pointed out on this thread a few times, trading with a size that allows me to not get worried or emotional over the outcome frees me up to not have to rely on arbitrary stops. I can trust the system and the billions of iterations that went into creating it.
 
frostengine said:
Background

My trading is 100% fully automated and run against a live (real money account). Primarily focused on micro future contracts MES, MNQ, MYM, and M2K. However, I am now starting to venture out into individual stocks as well. More on that in a moment as its partially the reason for this journal.

All of my strategies are generated via machine learning (self developed platform - not using off the shelf software). My execution engine is also self developed.

I currently have 448 independently trained strategies focused on the 4 micro contracts listed above running live. Half of those strategies can "swing" meaning hold the contract for up to 21 days, while the other half are pure day trading strategies and are flat by market close. The strategies are evenly divided between long and short.

Futures Swing Live Results:
2022-03-01 $30,036.00
2022-02-01 $46,064.75
2022-01-01 $-26,675.25
2021-12-01 $17,146.50
2021-11-01 $7,451.00

Futures Daytrade Live Results:
2022-03-01 $22,113.50
2022-02-01 $18,462.50

I have less live data on the daytrade strategies as I just started training/running those starting in late January.

Strategy Generation

I have 7 r5.8Xlarge boxes running at AWS constantly training new strategies. Generally speaking I can build about 28 new strategies per week. Which is what I normally target for a new contract. 14 long and 14 short specific strategies. This week I am training a daytrade strategy (flat by EOD) for AAPL.

Each week I plan to train a new stock

Goals/Journal Discussion

I have 3 primary goals for starting this journal:

#1 I am bored... everything is automated, even the strategy generation for the most part. Takes only a few minutes of manual work each day to sanity check new strategies. I am looking for this journal to provide something to do.

#2 To show what is possible with a fully automated system and hopefully inspire others.

#3 To get ideas on which stocks/markets I should train strategies on next - also to discuss best ways to trade those markets.

For example, I am now training AAPL. However, I am debating the best way to trade. Should I trade the shares directly or use options in liu of shares?

Historically, I found even with winning strategies where the option contract is held less than a day, a noticeable drag exists.

My concern with trading the shares directly is if it becomes hard to borrow during extreme moves. I need to be able to reliably count on my strategy executing on both the long and short side. For those who frequently trade AAPL on Interactive Brokers, have you ever seen it have an issue executing a short?

I already have the ground work in place to train SBUX as my next stock. After SBUX, which stock would you recommend next?
More...

Setup a live feed to capture your desktop on YouTube. Software is free.

https://obsproject.com/
 
Millionaire said:
When doing directional trading they say it is important to "Ride winners and cut losers short".

All my research suggests this rule is correct. It is very hard to maintain a trading edge over the long term edge without doing this.

But your systems might be the exception, good luck!
More...
What methods did you use to research this philosophy, as it lacks context and criteria?
 
frostengine said:
In all the systems I have designed, I have never seen an arbitrary "stop loss" improve the results I care about. All of my systems are designed to have a "reason" on why it gets out. This "reason" may be a black box as far as my understanding of it, but the models were trained to create that reason and the data supports the reason. This has shown to be far superior for my style of automated trading.

This goes back to an edge that has been pointed out on this thread a few times, trading with a size that allows me to not get worried or emotional over the outcome frees me up to not have to rely on arbitrary stops. I can trust the system and the billions of iterations that went into creating it.
More...
Overall I find this all very impressive, and I'd like nothing more in my trading approach than to be able to apply ML to it.

A couple questions. How much knowledge of coding is required when you submit your inputs into ML in order to run the iterations needed to develop a system, and is there a particular ML engine you recommend for this type development? Also, is your engine able to analyze criteria that applies to many time frames/periodicities simultaneously and thus yield an effective strategy that based on the collective analysis?

I noticed you close out positions at the end of regular hours trading. Is that limitation a product of the ML or the parameters you've given it so you eliminate overnight risk? I was wondering if that is costly to do, as some anticipated price improvements may simply need more time, or if it is accounted for in the expected results of the trading system.
 
Last edited:
Aisone said:
Overall I find this all very impressive, and I'd like nothing more in my trading approach than to be able to apply ML to it.

A couple questions. How much knowledge of coding is required when you submit your inputs into ML in order to run the iterations needed to develop a system, and is there a particular ML engine you recommend for this type development? Also, is your engine able to analyze criteria that applies to many time frames/periodicities simultaneously and thus yield an effective strategy that based on the collective analysis?

I noticed you close out positions at the end of regular hours trading. Is that limitation a product of the ML or the parameters you've given it so you eliminate overnight risk? I was wondering if that is costly to do, as some anticipated price improvements may simply need more time, or if it is accounted for in the expected results of the trading system.
More...

My platform is all "self developed" - not using any off the shelf ML software. Therefore, to do what I am doing would require a lot of coding experience. However, I think there are decent tools out there that can be used with less experience.

Its less about the "engine" training the strategies and is more about the data you feed it. Yes, my system is able to analyze multiple time frames simultaneously.

As far as closing trades out at end of day. I am running 2 systems where I am posting Twitter trades. My futures swing trade system and my futures daytrading system. The daytrading system is designed to be flat overnight, so that is what your seeing exit positions before close.
 
You must log in or register to reply here.
Back
Top