Let me repeat the quote - Forget the risk and take the fall, if it's what you want, then it's worth it all.

So my 9 month dream run of making profits finally came to an end with Jan 2017. Not because there was something wrong with my algorithmic strategy, but I wanted to try few new things. I had spent almost 6 months preparing for this one month - collecting data, paying google compute engine cost, burning the midnight oil for iterations, etc.

Anyhow, here are the results. Out of my first set of tests against IDEA, HINDPETRO, ADANIPOWER, TATAELXSI, DIVISLAB and KSCL, I strangely got 6 of them right within two days of trade alerts. The strategy revolves around situation where the stock has run up quite a bit, has still not got anything structurally wrong or structurally extraordinary, revolves around mean reversion, money flow, historic patterns and peek levels, and some awesome machine learning to boil things down on probability. You heard it right - I managed to predict moves greater than 5% almost a day or two before. However, it is a totally different issue that I managed to make money by only one of them.

But I think I have got my hands on something awesome and I could have never figured it out if I hadn't gathered the courage to test it in 2017. It all looked pretty neat on paper but I chickened out in the very last moment. I think multiple things were at play -
  1. Playing with highly leveraged products like Futures without understanding the risk
  2. Wrong trade PAIRS hurt daily mark to market that further makes you doubt the trade
  3. Not taking probability into consideration and playing against huge delta risk was definitely going away from ground rules
  4. Hedged leg acting against you due to budget rally and increase in volatility which essentially ensured that you were playing naked with double edged sword
  5. Giving up on your research on popular and stupid public opinion that affected subconsciously
  6. Chickening out early because I was testing and was not having a detailed execution plan involving what if ... 
So I would say, this has been a great month. I learned a lot and feel pretty confident that I have got my hands on a beautiful filter and getting the executions right will help me come out with flying colors in months to come ... Amen!