I like the idea of going live after a big drawdown day......... I have noticed a type of seasonal correlation to the bot's performance... while looking at historical data and results it appears the bot performs MUCH better at the start of earnings season.... and then starts to fade out towards the end of earning season....... Does anyone have any ideas why this may be the case?
stock trader almanac by Yale Hirsch may have the answer http://www.amazon.com/Stock-Traders-Almanac-2007/dp/0471783773
The thing is, unless you used a hedge you could only do that once - the next time you will be CAUGHT in the drawdown. I would think that the bot performs well after earnings because the stocks have more direction and movement. After that as time passes more confusion = more chop. That's one theory anyway.
Your system is probably trend based and the market trends more during earnings season when the market is already trending to some degree. Turn your system off during fed days and also the days the fed minutes comes out. Also turn off on big news days. John
Most of the times, when you first go live with a strategy, there may be a continuation of the current profitability of the system, at least temporarily. Like for 2 months, the bot continues to work with almost the same reward and then gradually stop to produce gains and enter in a DD. It would be good to use this technique but only after the first initial run-up.
Any updates on how the bot's been doing since you shut it down? I've been following the thread and am curious. Thanks.
The past 3 weeks have been the best 3 weeks it has had in about the past year.... It has not been on my real money account unfortunately though.. but it has been doing incredibly well... Will probably wait few more months and go live with it again... still working on trying to put in more protections for now