When I first started down this yellow brick road called algorithmic trading, I didn’t put a lot of thought into how to allocate a trading account amongst each position. That was until I learned about an advanced technique I could incorporate into QiT. Once it was explained to me, I was hooked.
It is a 2-tier type of position sizing we have christened QiT’s Advanced Slot Technology, and today I would like to disclose what it is and how it works.
Why it’s so important
As you know, only the best traders are principled enough to:
- have a plan before entering a trade – the algorithm
- take a trade when their rules tell them to even if every fiber of their body tells them no – follow the rules
- have a well-tested and well thought out money management regime
QiT’s Advanced Slot Technology is our money management regime. It quantifies the optimum percentage apportioned to each trade including the number of trades (slots) assigned to a portfolio.
Assigning an absolute percentage to each position is the most rudimentary money management system. For example, 4 positions, 25% per position. But what if there was a better way that would ramp up our returns but not the risk? Turns out there was.
Slot Technology answers the questions if we have 7 open positions with 10 maximum allowed, only 3 slots are left. Seems simple enough, doesn’t it? Nothing advanced about that. But what if, of those 7 open positions, 5 signaled to “Sell at Open?”. Do you have 8 slots to fill or 3?
How about if the signal is Sell at Limit? Do you wait until those slots exit and then fill them? Our algorithms don’t use intraday data so that wouldn’t be possible. How do the Slots get allocated in a CASH account as opposed to a MARGIN account?
I could blather on about the innumerable questions that came up but I think you get the idea.
It took us over a year to get it unwrinkled.
However, that is the tip of the iceberg. When you start digging deeper you’ll find this technology does a great deal more than just apportion slots.
It takes the maximum number of slots, breaks it into small and large (this is where 2-tier comes in) then assigns a factor to the large. This means the slots in each portfolio will be allotted a different percentage (where the percentage on the Signals Tab comes from.)
Let me demonstrate
I ran the optimization for these three parameters using the Momentum Retirement algorithm starting at January 1, 2016. It took 13 hours – (I’m thinking I need to rent a Cray).
The results were very illuminating. The Compounded Annual Return ranged from 60.84% to 34.37%. The Max Drawdown ranged from -2.47% to -5.47% (remember this is for 2016 only).
Allow me to digress for a moment. Most would assume I would default to the highest CAR and lowest MDD combination but, like everything in the Quant Trading, it’s not that simple. Over the years of doing these optimizations and the thousands of backtests I have run (yes Virginia, thousands), I have developed an uncanny aptitude to spot the combinations that are sustainable.
Sounds like intuition, doesn’t it? Well yes, it is and I’m going to admit it, I have developed an intuition for eyeing the best combination of parameters and use it daily. (explaining my evaluation process will be a very lengthy newsletter.)
The optimizations I run for each of the strategies will alwincludeudes these parameters so you will see them changing over time. I have been using the Slot Technology long enough now to know what works and what doesn’t.
For example, the smaller number of Max Slots will always result in a higher CAR. But that is not always the better combination because it will give you greater volatility. And, as I’ve heard from various members, heartburn in allocating too high a percentage of your account to any one trade.
We’re the Plan in “Plan your Trade and Trade your Plan”
Sign up today.