Also, how did it compare to back tests during the same time period. From the sounds of it , very similar
Adventures in Automation
- Thread starter fan27
- Start date
My Risk component is configured so that I will never have more than 4 positions concurrently and at the time of the trades the max risk per trade was 2% (I have since lowered it to 1.75%). So when I had 4 concurrent positions in mid march my max account risk was 8% (there were stops on all of the positions).
Part of what fueled the massive gain was volatility dropped and all of strategies currently used a volatility based stop (i.e. 3 X ATR). So what first happens is my strategy code determines where the stop should be and how many points away from the entry price that is. Next, it determines how many contracts can be entered for a position based on max account risk per trade. For example, in ES if the stop is 20 points (20 x 50 = 1000 per contract) and my account is at 110,000 with max account risk per trade of 2%, the position size will be
110,000 x .02 = 2220 (max risk)
2200 / 1000 = 2.2 or 2 contracts for the position
I had an NQ and ES position that both had 3 contracts each. That one NQ trade with 3 contracts held for about a week resulting in a gain over $11,000.
Last edited:
I deployed this particular strategy portfolio at the beginning of March and the results are lining up almost exact to the backtest, accept for a couple of entries being off by a tick (sometimes in my favor, sometimes not).
As soon as I release my product within in the next few days, I am going back into research mode in between fielding product inquiries. I want to see if I can find some trade-able strategies in the major currency pairs. The main purpose of this is to add more capabilities to my product and gain a deeper understanding of AlgoTerminal...but of course I also expect this to add to my trading profits.
Baller
It has been a very quiet April...no trades yet.
One cool thing I added to my live trading tech stack is the ability to send pertinent application log data and alarms to different Slack channels. This sort of setup is common in the SAAS space. The objective is to proactively be alerted of problems and have the information necessary to diagnose and fix problems. So now, I will get alerts on my phone via the slack app and I can configure notification settings for different channels. The log data first goes to AWS Cloudwatch but I am going to investigate and app called Seq as it can process log data directly and post to Slack.
One cool thing I added to my live trading tech stack is the ability to send pertinent application log data and alarms to different Slack channels. This sort of setup is common in the SAAS space. The objective is to proactively be alerted of problems and have the information necessary to diagnose and fix problems. So now, I will get alerts on my phone via the slack app and I can configure notification settings for different channels. The log data first goes to AWS Cloudwatch but I am going to investigate and app called Seq as it can process log data directly and post to Slack.
30 bar exit?
Yes sir.