Pay no attention to that man behind the curtain… well maybe you should.
When trading options with the TVO System, you’ve often heard me say the total cost of the contracts should not exceed 10% of total capital. But why 10%? How did we come up with that number?
Well, the truth is, at first the number was arbitrary. 10% sounded like a good number and when applied to the backtests, the results of a 10% position size produced the smoothest and most profitable equity curve over time. The returns were consistent and drawdown was manageable. But that still didn’t explain how it was working. I reviewed several position sizing models and discovered that “the man behind the curtain” was actually named Kelly, as in the Kelly Criterion.
Options are highly leveraged and proper position sizing is crucial in risk management. When trading a system without stops, as we do, this becomes even more important as the position size is what determines your risk level. Too many contracts can quickly sink your account. At the same time, too little will not effectively build capital. The Kelly Criterion (aka “Fortune’s Formula”) was developed as way to determine just right amount to put at risk.
The theory is that the formula will allow you to build your account steadily based on the parameters of your own system’s past performance. The pure version of Kelly, however, puts way too much risk on the table for most traders and investors, so a fraction of the formula has often been substituted for the full equation (Larry Williams was famous for his “Half-Kelly” method). This still does not take into account the leverage involved in trading options.
That’s where the Options Position Sizer comes in. After entering in the data from your backtests, a modified Kelly formula is then applied with Lambda (or leverage) calculated by the data from the options chain. The result tells you exactly how many contracts to trade.
Let’s see how it works…
The first 3 fields in red are filled in from the backtesting results of your own system. The numbers shown are the win rate, average gain and average loss from our backtest of the Total Volume Oscillator (TVO) traded with the SPY ETF over a 16 year period. When filling in these parameters it’s recommended to use the results from a data sample of at least 40-50 trades.
Next you enter your total account capital, followed by the Delta, Strike and price of the options you’re trading. This options data example is from an actual trade we made. The field in green is a result of the equation. It’s not always a whole number and since you can’t trade partial contracts, it’s rounded to the nearest contract.
The yellow fields on the right are the nuts and bolts of what’s going on. As shown at the top, the Kelly position for this system example would be a whopping 53% of your capital, an amount way too high to risk on options (or anything else for that matter). After modifying Kelly we end up with a much more modest 7.28%.
For most stocks and other securities you could probably just stop there, but we go one step further by factoring in Lambda from the options data. The result is 8 contracts which is a position size of around 9% (pretty close to 10%, right?). Depending on how your system performs and the types of options used, the percentage will vary. We’ve found that with SPY options on all of our strategies (TVO, HG, and IO), the percentage falls in the range of 6 to 14%. -MD
Subscribers can access the calculator here.
Not a member? Try our Options Position Sizer today when you subscribe to our options signals. The first month is free.
To view past positions check out our Trade History.
The Options Position Sizer is an educational tool. It is not intended to provide investment advice, and users of the Options Position Sizer should use their own discretion when making investment decisions based upon values generated by it.