Ever felt like your trading is a bumpier experience than it should be? The purpose of this post is to show one of the most likely reasons for volatility in traders’ equity curves and the risk of ruin based on your edge and money management skills.
Note: This post was initially published on my Slovenian site Trgovanje.eu.
Trading too big
Through talking to many (more or less) experienced retail traders and working with professional traders leads me to believe that traders generally trade too big. Some enjoy the thrill, some are over-confident and some are just not aware of the consequences of bad money management. But usually all of them have something in common – a very volatile and bumpy ride when it comes to their equity/P&L curve.
What is money management? In my view money management is a capital protection method with a combination (!) of loss limitation and trading edge. One of the money management approaches is to limit losses per trade as a % of your equity. Trading edge, should that be trade win ratio, risk reward ratio or both, enables you to select trades which have the potential to make the highest returns possible.
When I started trading I experienced a bumpy ride. I wondered what I was doing wrong, I questioned my analysis, my system, everything… but I needed some time to address over-sizing issue. Positions I opened were, relative to equity, way too big and I wouldn’t dare to trade that big anymore.
Let me help you build risk awareness through the results of the simulation I created. The simulation gives the probability of a draw down shown under assumptions below. I will also show an example of an equity curve one should expect to see having the same trade performance.
The goal of the simulation is to calculate probability of having a draw down bigger than 25%, 50% or 90%. Each draw down is measured from the most recent peak.
The assumptions are:
- Trade only one trade at a time.
- One can’t know if a trade will be successful or not in advance.
- Trades are independent of each other. This means that the latest trade result gives no information about success of the next one.
- Simulation was run for all combinations of the following assumptions.
- Trade success ratios are 40%, 45% and 55%.
- Risks per trade are 2%, 5% and 10% of equity curve.
- Risk reward ratios are 1:1.5, 1:2 and 1:3.
- Winning trades have a Log Normal distribution with an average determined by the RRR.
- 2/3 of the losing trades have a loss of max risk per trade and 1/3 of the losing trades have half of that.
I simulated 100,000 scenarios for each combination above. Each scenario consists of 100 trades with features above.
Let me show you an example of trades’ distribution for RRR 1:2, risk per trade 2% and win ratio 40%. An average return of winning trades (blue) is 4%, which is determined by risk per trade and RRR. Losing trades (red) are scaled on the right axis.
Before going to the results section, let me just write about one trade at a time assumption. It would perhaps be more realistic if I did a portfolio based simulation. But a) this is much harder to simulate and b) even though this seems less realistic, you have to know that on the other hand this simulation is always behaving exactly under chosen assumptions. When a trader starts losing money they usually change their behavioral patterns to some extent, and could start making more mistakes.
Let me first show you a table with probabilities of hitting a draw down depending on a combination of parameters. Note that a blank field means no scenario visited that draw down.
Results in a very straight forward way prove that a trader’s skill or an edge is important but not as important as the risk per trade the trader is taking to limiting draw downs.
For example, let us compare trader A with win ratio of 40% and risk per trade 2% of his equity to trader B with win ratio of 55% and risk per trade 5% of his equity. Clearly trader B has better skill than trader A, but probability that a certain draw down level is hit is significantly higher for trader B.
Not only that, but everyone can easily see what their probability of having a certain draw down is. Let me now show you simulated equity curves for 100 trades, 40% win ratio, 1:2 risk reward ratio and 2% risk per trade. Of 100,000 scenarios I show only 100 equity curves in grey color to distinguish between them but statistical metrics are calculated based on all scenarios.Results show that, under assumptions above, one could expect an average return of 80%, median return of 74%, minimum return of -46%, 98% probability of a positive return and 95% probability of 10% return or higher after 100 trades. A clear consequence is that the higher the trading skill the better the results one can expect.
Of course taking bigger risk per trade could lead to better returns, but in this case one is betting that a few bad trades in a row won’t happen. Should this happen it would very quickly lead to a game-over scenario. Usually taking bigger risks works until it doesn’t work anymore and numbers in the table above prove that.
It’s up to every individual himself to decide what risk is he willing to take. Good luck!
Should you be interested in any other statistics derived from the simulation or should you want to change the assumptions a bit to fit it better to your trading do not hesitate to contact me through the contact form.
If you like this post please subscribe to my blog here.