Search results

A lot of strategies discussed in previous threads are still used today. However, a lot of previous threads were using options for the executions - which had too much drag. With that said, the number of strategies have ballooned up significantly over the past 3 or 4 months. Really honed in a...

Looks like the twitter posting is working - might have some throttling issues - but appears to be posting.

No, a .25% expectancy means every trade taken has the expectancy to make a .25% gain. Some trades might make 1%.. some trades may lose .5% etc.. but overtime it averages out to a positive expectancy of .25% per trade. What you are referring to is win %. My win% is roughly 67%.

I am building up to minis. For the number of trades/contracts open at a time, I need a significantly larger account. As far as posting to Twitter. Figure with nearly 450 independent strategies, the signals will look like noise - so little risk in someone reverse engineering the strategies...

Might be some bugs with the Twitter integration, but the system should start posting live future trades tomorrow morning. https://twitter.com/AlphaStratLab

I'll do one better.. How about this, I set up a twitter feed that posts in real time every trade as it happens? If there is enough interest, i'll set it up.

Judge what?

Since I am trading predominately micro futures so far, each strategy is allowed to open 1 contract. I have an even number of long/short strategies so it typically doesn't get too extended in any one direction. However, there have been times where it has fairly large net long or short position...

When I first started training my strategies, I used a local desktop. Once I had the training process developed, I started to scale out to r5.8Xlarge boxes. These boxes provided the right mix of memory and compute power needed. I am running the machine learning on Linux boxes, and execution...

I am willing to discuss overall infrastructure... stock selection... lessons learned.. The strategy generation and data that goes into the strategy are off limits. The inputs that I feed into the ML algorithms are critical to the outcomes. About as much as I am willing to say is that it does...

I started out discretionary, however even from the earliest days of my trading career I was attempting to automate. I never was a really successful discretionary trader. Would have been better off keeping my money in a low cost index fund. I started this year with roughly $800k.

Unfortunately, this is an area I am not willing to discuss. The "secret" sauce to what makes my strategies work is in the data and algorithms. Garbage in = Garbage out.... I believe solving this issue was the most significant discovery I made in my 20 year trading career. Now I can train new...

I actually train the strategies on SPY, IWM, QQQ, and DIA - but execute them against the futures. My data goes back to 2002.

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...

As stated, the strategy never had a losing year. The bottom 5 years had the following returns: +39% +71% +78% +94% +101%

The drawdown I cited is intratrade - while the trades are still open. The win % is on closed trades. On closed trades the drawdown is significantly lower. If I only look at drawdowns when the trade is actually closed vs looking at it open, the largest drawdown was only 21%

This strategy was designed to replace the "buy and hold" portion of my portfolio, so it trades index ETFs. The strategy will scale into the position and will utilize some leverage in certain circumstances. So, it's pretty easy to imagine if one was to buy and hold SPY for example, you would...

I have a new automated strategy I am planning to go fully live with soon. I have been running it live with minimal capital since Feb. Stats are holding up well to historical, so now I am looking for some feedback on the best position sizing to utilize as I allocate a portion of my portfolio...

MTM PL Since Inception: +$63,758 However, the last trade the system made was in March. Apparently the bot does not like markets that go up in a straight line with no pullbacks. Not a single trade placed so far in April.

The past month has been a lot of nothing as far as PL goes... the bot proved it can sometimes be wrong. Overall there were periods in the past 30 days where the market was "tough to trade". The bot got caught twice on the wrong side of big moves. Overall, the PL since inception hasn't changed...