Maybe the only way to do the inverse is to be a market-maker and collect bid/ask spread every trade, while continually hedging market risk.
No reason to think that hedging market risk faster/slower changes your expectancy. And if spread is the issue, no reason you: a) can't sit on the ask/bid, b) trade instruments with narrow spreads. Not all options have wide spreads.
Let's forget that anyone "knows" anything about an option's underlying for a moment and let's look at the "gambling" aspect of options. Assume I throw a dart at an options listing and it hits: ABC Bid: $2.65 Ask: $2.75 Now it doesn't matter what the strike price is, what the term of the option is etc. If the option is priced correctly in the market, the value of this option should be about $2.70. However as a retail buyer or seller you will not be able to buy this option for less than $2.70 and you will not be able to sell this option for more than $2.70. Let's assume one person buys one contract for 2.75 and one person sells one contract for 2.65 with a commission cost of $2.5 per contract. If this experiment was repeated many times, the person that bought the contract would get an average sell price of $2.70 for a loss of $5 on the contract and the person who sold would have to buy back the contract at an average price of 2.70 for a loss of $5 on the contract. Each person would have a negative expectation of $10 per contract ($5 loss of value (spread) and $5 commissions (of course the seller may average less than $5 if the seller does not buy back all the contracts but that can be risky). Realize that on any one trade, one trader will more than likely make money and the other will lose money. But over time and many trades, the wins and loses will begin to average out so that the buyer will get an average expected return of $2.70 and the seller will have to buy back at an average expected price of $2.70. When opening any options trade, the trader is fighting the spread and commissions. To do better than our traders above, one could try for a better price (closer to the middle of bid/ask) and try to get a better commision rate (such as $1/contract). Even with these two improvements, our random buy or sell option strategy still has a negative expected value, just less negative than our traders above. Every time you add a leg or adjust your options trade, you add on the negative expectation created by the spread and commissions. Unless you can prove to yourself that for some reason the options you are about to trade are mispriced in some way, you will have a negative expectation no matter whether you buy or sell. Joe.
Joe-- Well put and largely correct. But to be fair, you must also give the naive readers a bit of a hope and tell them how one can still be profitable in trading options, albeit the negative expectancy.
I don't think hedging market risk changes expectancy. When trading large size, however, it must be considered in order to reduce the chance for spectacular blowup. Trading large size is the only way to make a living off B/A. Is it coincidence that MMs are hedging off the market movement aspect of options, the part of the equation that many retail traders believe they can master? As for parking orders at B/A or trading instruments with narrow spreads, I think the retail trader will not have equal footing there, in terms of seeing order flow. If you believe you can predict the movement of the underlying, it would probably be cheaper and more efficient to just trade the underlying.
I believe most options traders employ spreads and I also believe that it's possible to configure spreads with positive expectancy. e.g. MCD closed on friday at 63.85 The April 62.50 call closed ask at 2.31, the April 65.00 call closed bid at .91 Using a probility distribution the probabilities of MCD fallling in ranges are: Range.........Prob 55-60.........4.2% 60-65..........63% 65-70...........32% 70-75..........0.7% http://www.optionistics.com/f/probability_calculator This accounts for 99.9% of the distribution Computing expectation for buying the 62.50 call by taking the P/L at the mid-point of each range gives: Ex = -.042(231) -.63(231) +.32(269) +.007(769) = -63 minus 1 comission of 2.50 = -65.50 i.e. a negative expectation for buying the April 62.50 call We apply the same logic to a 62.50/65 bull call spread: Ex = -.042(150) - .63(0) + .32(110) +.007(110) = +29 - 5.00 = +24 i.e. a positive expectation for the spread.
If you're trying to replicate, the essential element of option trading is to buy at a vol less than realized or sell at a vol greater than realized. Then you wait for probability to smile in your favor.
I don't follow your calculations. It appears that you are using probabilities for March but are buying and/or selling April options. Please explain further. Joe.
I redid the calculations doing Sept because it's a lot of work and I need Sept numbers: The probabilities are: MCD September: Range.............Prob 45-50...............1.6% 50-55...............7.3% 55-60..............17.3% 60-65...............24.1% 65-70...............22.2% 70-75...............12.2% 75-80.................7.6% Then I did the expectations for the Sept. 62.50 call and three spreads: 62.50 call: -14.2 62.50/65 bull call spread: -8.5 55/50 bull put spread: +20.9 50/45 bull put spread: +15.4 Again I used the midpoint of the range to get P/L numbers for each range for each trade. The bull call spread still has a negative expectation but less negative than the straight call, the otm bull put spreads both have positive expectations. So I said I believed it is possible to construct spreads that have positive expectations.... here are two. I am sure there are more. I am pretty sure this is right as I rechecked all the numbers. If you have a few hours to spare feel free to check my numbers.