When new traders begin to use trading systems their first approach to the use of an automated trading strategy – regarding lot sizing – is often to make use of a fixed lot size for all of its trades. This approach certainly makes sense from an inexperienced point of view since you know how much your won/lost value per pip is for every trade and you can also make sure your system risks as little as possible – by using the minimum lot size – on every trade. Nonetheless – after some experience is acquired – it becomes obvious that this approach has MANY setbacks which make it a very poor choice to implement within any automated trading strategy. Within the following few paragraphs we will learn a little bit about why fixed lot sizing is so bad for trading systems and why statistical analysis and the achievement of long term profitability are greatly encumbered by this way to calculating lot sizes.
When you trade a fixed lot size on every position it means that you will trade a fixed amount of win/loss per pip on all of the trades you enter. The first consequence of this decision is that in order to maintain a fixed risk per trade you will need to maintain a fixed SL or TP. This is the first reason why a fixed lot size approach doesn’t work very well in the long term, the SL and TP need to be fixed in order to keep the risk per trade at the same level and therefore the system lacks any ability to adapt to changes in market volatility as it is always waiting for its signals to have a similar absolute expectancy something which makes no sense since the amplitude of market movements changes evidently as a function of volatility.
Now you might think that this depends on the nature of the strategy you’re trading but the fact is that the fixed lot size approach induces a lack of flexibility which is terrible for any type of trading system because – as I said above – it implies a lack of adaptability against the market or – even worse – a fluctuation in risk per trade when any adaptability is attempted. Strategies that rely on fixed lot sizes usually make absolute assumptions about the market – such as grid traders – which end up wiping accounts in the long run because these fundamental assumptions are always found to be wrong as market volatility fluctuates and the amplitude of market movements goes above or below the fixed level of the system.
Another very important fact which needs to be taken into account is the reliability and interpretation of simulations with a fixed lot size approach. People wrongly thing that simulations with fixed lot sizes are “more reliable” since there are simply no changes in the win/loss per pip along all trades but the fact is that the fixed lot size approach introduces a very dark dependency on the initial starting date that makes any system using this approach extremely difficult to evaluate. Suppose you start trading a strategy which has a fixed lot size that takes an account from 5000 USD to 15000 USD and then back to 10000 USD. The result of the test will reveal a 100% profit with a 33% draw down but the fact is that a person starting to trade the system with 5000 USD when the simulations reached 15000 USD would have reached a total account loss.
When lot sizes are fixed the relative draw down of a simulation depends tremendously on the starting point since risk is never adjusted to account sizes. If a trading system shows profits on a 10 or 12 year backtest using a fixed lot size it says nothing as there are potentially many account wipeouts along the way on draw downs which appear of little importance when looking at the “big picture”. In order to adequately evaluate the draw down potential of a strategy with a fixed lot size you would have to carry out separate simulations taking into account all possible starting points in order to ensure that there is no point in which starting out would imply a huge draw down.
The lack of adaptability against volatility and the lack of adequate statistical evaluation makes the use of a fixed lot size approach a very bad idea when coding or using an automated trading strategy. Strategies coded with fixed lot sizes are bound to very deceptive in back-tests (due to the hiding of potentially fatal draw downs) and therefore their rigorous statistical evaluation requires a very extensive testing which is most of the time neither worth it nor practical. The best thing is to have a system which dynamically adjusts to volatility, allowing the lot size to fluctuate in order to accommodate a fixed account balance risk per trade. This adjustment can be done in terms of many different aspects of the market, such as a percentage of the absolute value of a currency, the value of an indicator like the ATR or another type of volatility adjusted criteria such as a given price pattern size.
If you would like to learn more about my work in automated trading and how you to can learn how to build systems with adaptive money management mechanism please consider joining Asirikuy.com, a website filled with educational videos, trading systems, development and a sound, honest and transparent approach towards automated trading in general . I hope you enjoyed this article ! :o)