Analytical Toolbox: Consecutive Losses and Risk of Ruin
May 28, 2009
First the good news – unless you’ve specifically tested your system data for dependency from one trade to the next (i.e. wins leading to wins, wins leading to losses, etc.) consecutive losses can be viewed as random events and not an omen for the next trade. Assuming you have a properly tested system, remembering this may be just what you need to pull the trigger on your next valid signal – even if you cringe while doing it. Although more mechanical approaches reduce emotions while trading, it doesn’t mean there’s no trace of them.
The difficulty with taking the next signal after a string of losses is that there’s understandably a feeling of greater uncertainty for the trader. Is this a normal run of losses or one that will prove to be atypically long and deep? Worst yet, has the system stopped working? Edges in the market do come and go.
To help on the trader psychology front, consider viewing consecutive loss runs in your back-test data. This may simply help re-assure you that such consecutive losses have occurred in the past while the system has maintained profitability. Viewing these losses as a drawdown in the equity curve may also provide the perspective you need. A current drawdown that is much steeper than those occurring in the past may point to a system problem – something that is better assessed sooner than later.
In addition to viewing the system’s equity curve, traders can also gain confidence by calculating its risk of ruin prior to implementing it. It may very well be necessary to take a break from a system, but only you can decide that. Keep in mind, though, that missing subsequent winning strings will matter since it’s a system’s compounding of positive returns that leads to long-term profitability.
Reducing the discomfort of taking the next signal is only an appropriate goal if it has been properly tested under a variety of conditions prior to initiating it. Once implemented, if your system does not have a filter to prevent signals when conditions turn unfavorable, you may want to assess trade results that accumulate to confirm the original edge still exists.
Risk of Ruin
Figure 1 provides a view of the Risk of Ruin Spreadsheet from the Resource CD included with the Cornerstone Home Study Course. The approach uses Excel formulas from Perry Kaufman that modifies some of Ralph Vince’s work … I think that covers all references if you want to explore any further. The tool shown here provides expected results, which is the best we can hope for when trading the markets. The yellow shaded regions represent trader inputs.
Figure 1: System Risk of Ruin Probability
(Click here for larger view)
When there is a higher probability for wins versus losses the risk of ruin diminishes. Similarly, a system with an average winning trade size that is greater than the average losing trade size, risk of ruin decreases. This highlights one reason why it’s important to analyze system statistics – what if very few trades contributed to a significant portion of the profits and skews the average upward? It will give you a false sense security when using the risk ruin calculations. Stable results can go a long way towards realizing returns that are consistent with expectations, allowing you to implement your system as planned.
Consecutive Losses
System expectancies can also be used to view the portfolio impact due to consecutive losses, as well as dollars lost per trade. The Risk of Ruin Spreadsheet includes four different %Win views for consecutive loss analysis along with a tab that allows user-defined entries.
Figure 2 provides a view of the impact of consecutive losses for a system with an expected win percentage of 50%.
Figure 2: Impact of Consecutive Losses on System Portfolio Balance and Trade Losses
The higher the number of consecutive losses, the lower the curve appears on the chart (log scale on right axis) due to a portfolio balance that diminishes more quickly at first from the standpoint of absolute value.
Table 1 provides a progression of five trade losses as an example.
$10,000 Start: Consecutive Losses Assuming 50% Trade Size & 10% Loss | |||||
Trade # | 1 | 2 | 3 | 4 | 5 |
Loss | 500 | 475 | 451.25 | 428.69 | 407.25 |
Balance | 9500 | 9025 | 8573.75 | 8145.06 | 7737.81 |
Table 1: String of 5 Consecutive Losses
The Dollar Loss per Trade curve is a steeper on the left side of graph for all three consecutive lines because incremental bet size changes have a bigger relative impact on loss value. Table 2 provides a progression of two consecutive losses for four different trade sizes. There is a five-fold increase in the first two trade sizes (1% and 5%) and the second two trade sizes (9% and 45%), but this does not translate to a 1:1 relationship in losses when a second consecutive occurs.
$10,000 Start: Consecutive Losses Assuming 10% Loss | ||||||||
| 1% Trade Size | 5% Trade Size | 9% Trade Size | 45% Trade Size | ||||
Trade # | 1 | 2 | 1 | 2 | 1 | 2 | 1 | 2 |
Loss | $10 | $9.99 | $50 | $49.75 | $90 | $89.19 | $450 | $429.75 |
Table 2: Impact of Trade Size on Consecutive Loss Size
Moving Forward
Creating a system with an edge, back-testing it and assessing the results certainly promotes analysis to paralysis behavior. Add a few tweaks and you may lose the original, workable form. Unfortunately there are a few problems with not doing this work with some attention to detail. First and foremost is trading a system destined for ruin. The other problem with skimming through the process is that you could walk away from a profitable system that suits your style.
Initiating a system with smaller trade sizes can point out style or implementation problems that you may be able to resolve with a second or third version of the original form. This provides you with an opportunity to also improve the money management portion of the approach so you eventually go live with a system that allows you to realize compounded rewards.
To access other articles written by Clare White, please click here.
Clare White
Contributing Writer and Options Strategist
Optionetics.com ~ Your Options Education Site
Questions for Clare? Visit the Optionetics.com Discussion Board through Ask the Traders on the Optionetics.com home page.
Email This Article
View Full HTML
Digg This
Delicious
Subscribe© Copyright 1995-2009 Optionetics. All rights reserved. This material is for personal use only. Republication and re-dissemination, including posting to newsgroups, is expressly prohibited without the prior written consent of Optionetics. Optionetics is a registered trademark of Optionetics, Inc.