The Kelly Criterion is a method of management that helps you calculate how much money you might risk on a trade, given the level of volatility in the market. It emerged from statistical work done by John Kelly at Bell Laboratories in the 1950s. The goal was to figure out the best ways to manage signal-noise issues in long-distance telephone communications. Very quickly, the mathematicians who worked on it saw that there were applications to gambling, and in no time, the formula took off.

Using the Kelly Criterion for trading is only one method. There are other methods out there, and none is suitable to all markets all the time. Folks trading both options and stocks may want to use one system for option trades and another for stock trades.

To calculate the ideal percentage of your portfolio to put at risk, you need to know what percentage of your trades are expected to win as well as the return from a winning trade and the ratio performance of winning trades to losing trades. The shorthand that many traders use for the Kelly Criterion is edge divided by odds, and in practice, the formula looks like this:

Kelly % = W – [(1 – W) / R]
W is the percentage of winning trades, and R is the ratio of the average gain of the winning trades relative to the average loss of the losing trades.

For example, assume you have a system that loses 40% of the time with a loss of 1% and that wins 60 % of the time with a gain of 1.5%. Plugging that into the Kelly formula, the right percentage to trade is .60 – [(1 – .60)/(.015/.01)], or 33.3 percent.

As long as you limit your trades to no more than 33% of your capital, you should never run out of money. The problem, of course, is that if you have a long string of losses, you could find yourself with too little money to execute a trade. Many traders use a “half-Kelly” strategy, limiting each trade to half the amount indicated by the Kelly Criterion, as a way to keep the trading account from shrinking too quickly. They are especially likely to do this if the Kelly Criterion generates a number greater than about 20 percent, as in this example.