Measuring Profitability of Strategies - AlgoJi

In this article we will discuss the basic and most important parameters for measuring profitability of a strategy.

Number of Trades

It is not a profitability estimator per se, but as a starting point, you need sufficient number of trades to arrive any conclusion at all. What will be the ideal number of trades to draw a statistically relevant conclusion? It could be trades generated by strategy in at least two trading regimes. For a strategy with average hold time 30 min, this could be 1000. For a swing trading strategy with average hold time 3 trading sessions, it may be 200. A large number of trades also smoothen the equity curve and may lessen percentage drawdown.

Net Profitability

The Net Profitability indicates the bottom line- how much money you may make. This can be primary important estimator if supported by sufficiently large number of trades.

Expectancy

Expectancy is the single most favored estimator by many professionals. Unfortunately, most of the retail customers are never introduced to this concept at all. Mathematically, expectancy is the amount you expect to gain (lose) on every trade.
Expectancy = Probability of Win * Average Win – Probability of Loss * Average Loss
Strategy1: assume the a strategy generates 40% winning trades, where average profit per trade is 1000 Rs while average loss per trade is Rs. 500.
Expectancy= 0.4*1000 – 0.6*500 = Rs. 100
On the other hand, a strategy may generate higher number of winning trades and have lower expectancy.
Strategy2: Assume that 55% trades are profitable where average profit per trade is Rs. 1000 while average loss per trade is also Rs. 1100
Expectancy= 0.55*1000-0.45*1100= Rs. 55
Combining the Number of trades can give us significant insight. Assume Strategy1 generates 1000 traders per year and Strategy2 generates 2000 trades per year, while both strategies have the same expectancy of Rs. 55. The straegy1 will make Rs. 55000 per year while Strategy2 will make 110,000 per year!!

Profit Factor

Profit Factor is defined as the sum of all winning trades divided by the sum of all losing trades. A Profit Factor above 1.50 is considered good enough and above 2.0 it is considered ideal.

A strategy with profit factor less than 1.20 suggests too bigger a risk taken for making money. A strategy with profit factor less than 1.0 will lose money.

The Profit Factor for Strategy1 above is (400/300)= 1.33 and for Strategy2 is (550/450)= 1.22

It is not used often because it may hide vital insights about strategy performance.

Payoff Ratio

The Payoff Ratio is calculated as the average win divided by average loss.

The Strategy1 has Payoff Ratio of 2.0 while Strategy2 has Payoff Ratio of 0.91

We see that a Strategy with Payoff ratio less than 1 may still make money. The Payoff ratio cannot be used in isolation to judge the performance of a strategy.