Curriculum·R403 Risk Management as the Actual Product·about 43 min
Expectancy and R-multiples
By the end of this lesson you can
- →Express every outcome as a multiple of the risk taken, and explain why that unit is the right one
- →Compute expectancy from win rate and average R, and compare two systems that look opposite
- →Show why a high win rate carries no information about whether a system makes money
- →Explain why expectancy cannot be computed at all for a position with undefined risk
Senior · enrolled learners
This lesson opens with OptionSellers.com, November 2018.
- What happened
- OptionSellers.com managed client accounts by selling naked options on commodity futures, with no protective long options and no portfolio hedge, a strategy that collects a premium on most positions and carries theoretically unlimited loss on short calls. Between 8 and 13 November 2018 natural gas futures rose roughly 60 percent in under a week, described at the time as the largest short-term move in eight years. On 13 November the clearing broker, INTL FCStone, liquidated the positions into that move. On 15 November clients received an email headed Catastrophic Loss Event informing them that their accounts were not merely emptied but carried debit balances they owed the broker, because margin had been extended against their accounts to cover the losses. Reported losses exceeded $150M.
- The decision point
- The strategy had a very high win rate and it always did, because most weeks natural gas does not move 60 percent. What it did not have was a defined risk per position, and without a defined risk there is no denominator, so the expectancy of the system could not be computed by anybody including the manager. A record of many small wins is a record of many small wins. It is not evidence about the average outcome unless the sample contains the losses.
- Recorded loss
- $150,000,000
What you will be able to answer
- →What is an R-multiple?
- →What is expectancy?
- →Which system makes money?
- →Why can naked selling not be measured in R?
Orientation and Year One are open: anyone can read them without an account. From Year Two onward the lessons are for enrolled learners, because progress through the later years only means anything if it is tracked against a record.
It is free. We do not sell the list and there is nothing to buy at the end of it.
Sources and review
- https://earlyretirementnow.com/2018/12/18/the-optionsellers-debacle/
- https://optionsellerslawsuits.com/what-happened/
- https://steadyoptions.com/articles/james-cordier-another-options-selling-firm-goes-bust-r429/
- https://www.powertrading.group/options-trading-blog/founder-of-optionsellerscom-promoted-naked-options-selling-now-bankrupt
Confidence medium·Volatility low·Reviewed 2026-08-07·Owner unassigned
Contested
Total losses at OptionSellers.com are reported as exceeding $150M and were the subject of subsequent arbitration and litigation, with client counts and per-account outcomes varying. The mechanism, being naked short options liquidated into a volatility spike leaving debit balances, is consistent across sources and is what this lesson uses.
The 95 percent win rate and the plus 0.2R and minus 20R figures in part two are an illustrative profile of a naked premium-selling system, chosen to be recognisable rather than fitted to this firm's actual record, which was not published in that form. Any realistic parameters in that family produce a negative expectancy and the lesson states the parameters so they can be varied.
R405 owns backtesting, statistical honesty and the sample-size problem in full. This lesson establishes the unit and the expectancy calculation, and states the detection problem only far enough to motivate that course. Keep the split.