Why Sell Iron Butterfly trades? Iron butterflies are direction neutral and primarily a bet on the spread between the implied volatility versus the underlying stock volatility. Iron Butterfly trade ideas are based on selling ATM butterfly options because historical probability distributions are easier to model (given enough analytical / computing horsepower) with a closer expiry date. Selling very far OTM Puts and calls are harder to model because of fat tail distributions (See Taleb commentary on selling tail options vs volatility in the body) For Iron Butterfly sellers, even if the underlying asset moves against you, the losses are limited and the premium collected could be justifiable from an expected value perspective. Expected values are generated using past historical stock volatility tables and computed using a cumulative probability density function at various potential stock changes. Executing an entire basket of Iron Butterfly trades may have a higher chance of success than executing any one single trade Company Background and Key Events: Company Background: FedEx is an American based multinational courier delivery service company, with its headquarters in Memphis, Tennessee. The company’s market cap is USD 42.94 Billion and has an annual revenue for the year 2019 being USD 69.69 Billion. Key Events:The company is expected to release its first quarter estimated earnings for 2020, on 17th September, 2019. TRADE DETAILS OF - FEDEX - IRON BUTTERFLY - JUNE 27 2019 Expiry Date - August 9 2019 (31 Days to Expiry) Stock Price as of June 27 2019 - $163.31 Trade Details - Sell $162.5 ATM Calls & Puts and Buy $140 OTM Puts and $185 OTM Calls Net Option Premium / Max Profit to Seller - $11.01 Expected Value to Seller - $1.36 Max Loss to Seller - $11.50 ROI Based on Expected Value to Seller - 11.86% There is a 40% probability (1 minus 59.9% - the green zone in the chart) for the option seller to lose money and move outside the break-even zone (yellow & pink in the chart). However, the net premium justifies the potential losses on an expected value basis. Commentary on IV & Straddle Premium Trends The Implied Volatility (IV) of the company for the 30 day ATM has declined from 35% to 27%. The ATM Straddle premium is 6.12% while the put-call ratio (based on open interest) is 1.5.
Market is pricing in something around 7% terminal move from ATM as 1 Std. Dev in this series, and you think there's less than 2.5% chance it moves half of that? Close to Close range for the 5 day period up to 6/27 was already over a 5% move, and you think that there's only a 2.53% chance for a move similar to that to occur again in the next whole month? More power to you if you can model for mis-pricings at this order of magnitude but for the time being it seems like your model fails a sanity check. Either that or you haven't clearly communicated what you're trying to convey here.
Hi Magic, Sorry for the confusion in the communication. It's actually a 40% probability (1- 59.9% - the green zone in the chart)) for the option seller to lose money and move outside the break-even zone (yellow & pink in the chart). However, the net premium justifies the potential losses on an expected value basis. We also edited the original post to make it clear.
Alright, that sounds more realistic. You've basically got a stat vol forecast a few % under current IV. But you haven't really told us how you're modeling this. And the spread is pretty thin to begin with. Fly profile like this expected to come in fractionally under implied is available very often just under spot in the indicies. Not seeing a very compelling case for a trade here unless you've got more info to lend a high confidence level in your forecast, or have something correlated to spread this against at a better price.
Hi Magic, I am not sure we follow all your comments but here are some details on how we got to our expected value calculations: # We essentially model all past historical patterns for every stock. We have detailed tables for every period and stock move probability for a given ticker (e.g. FDX). For e.g. What is the probability of FDX moving 10%/9%/8% etc for every period 7/8/15/30/31/32 day etc. This is a very computationally intensive process but far superior to an assumption of the curve of the stock patterns and then approximating that through standard deviation. Here is some background from Taleb on the challenges of using SDev for modeling: https://www.edge.org/response-detail/25401 # Based on the probability tables for every single combination of stock move and period- we compute a cumulative probability density function for the various stock changes. That is in the 2nd chart you see above. # Once we compute the density function- we then run the expected value for each point in the stock price and then sum up the expected value. Hope this is clear. Happy to share more details if you have more questions.