Skip to main content

Don't Override Your System; stay FAT

Deepak Shenoy2 min read
Quick Facts
What is "Don't Override Your System; stay FAT" about?
What category is this article in?
Premium Portfolio
Who wrote this article?
Deepak Shenoy, Capitalmind's research team.
When was this article published?
Published 2 September 2008.
How long does this article take to read?
2 min read

So today was a crazy day. We were long four stocks and short one - Cairn - as our system threw out the trades. Within a few minutes we noticed that there was a strong move up - and after a 2% upmove in an hour, in which Cairn tanked about 3%, we decided to take a “discretionary” call - to exit. After all, the market was up 2.5% - how much more could it go?

As it turns out, a lot more. It ended up 3.57% up.

But that was not the worst of it. If we had not used discretion, we’d have been up 5%. Lesson learnt: don’t override the system.

We generally never do - in fact if either of us think of it, the other person stops him. And earlier there are instances when our discretion has proved to be beneficial. But net-net it seems to do nothing - and the discretion adds emotional baggage we don’t particularly want. The best, therefore, is for computers to take over and do the trading - no discretion, no emotion, just a P&L at the end of the day.

Fully Automated Trading - the only reason I want to stay FAT.

On another front, my Nifty straddle went nuts. From a net price of Rs. 126 yesterday it was up to Rs. 170 today - a Rs. 44 profit, which is 30% plus. Of course I can’t claim it’s a single day, and since I wouldn’t put all my capital on it it wouldn’t even be 30% of total capital. But say I were to allocate some 1/5th of capital it would return about 6%. That’s not bad at all, even if this were the only deal to happen the whole month.

Having said that, I went long another strangle near the money - the idea being that such moves will result in either an extension of the move, or a retracement, again helping a long strangle. If I were to suggest a new position - as of end of day today - it would be a 4700 call and a 4300 put - for a net cost of Rs. 120. A profit target of 20% would apply - and then yet another strangle. My preliminary back tests show that this is only successful when the index has been range bound for a while - but when it does set up, it’s largely successful.

Note: to program this into a trading system is trivial, so it’s not quite discretionary. But I haven’t quite done the full array of tests I would usually do with a system, so I consider it a discretionary trade.