Well, I would agree that statistical edge exists, but I am just referring to pure options position and disregarding fundamental and technical factors. I tried to mean when picking a random option at a random time without any mispricing throughout the strike range, there isn't a statistical edge of one position versus another --or that is what I assume. There's a whole field of statistical arbitrage, but I think this is usually taking into account of analysis of extraneous factors like volatility, fundamentals, technicals, etc. I suppose you might be able to construct an option position to exploit pairs trading, but I'm not sure if that would be worthwhile either. So anyhoo, from the replies that I got from dmo to spin, I gather my hunches are not far off from the mathematical truth. This will prevent me from making overly complicated positions with big commission penalty.
EDGE 1=exchanges=fees 2=market makers =bid ask 3=brokers=comissions 10=retail trader or investor = has to pay for all above
I am simple minded to just compare historical volatility with implied volatility to tell me if an option is mispriced or not? Other than that, if I were to analyze the stock then I would just consider the stock mispriced and not the option...
No strategies give you any edge. Excluding commission, any single option gives you zero expectancy. Expectancy of a combination of options = sum (expectancy of each individual option) = sum (0) = 0 Theoretically, if you pick any option randomly, you can't find any option that gives you 50% of winning 100% and 50% of losing 50%. If such option (or option strategies) exist, that means there is an option that gives you a positive expectancy. Your edge should come from something else. It might come from Mark's or Coach's advices. It might come from predicting future IV. Bottom line: You need to have a strong risk management to make it work for you. [edit] I never read about the implied statistical edge concerning options. Can you give me a link or a pointer? I like to read about such edge that should not exist.
Just comparing historical to implied is not sufficient, although it does help somewhat with decision making. That is, if the implied is high, you need to ask why, but if you are working with an index, you may have a modest edge if the implied is very high since it does tend to revert to the mean eventually. What this means is that you have a slightly better than average chance of declining volatility during the holding period. A mild edge, perhaps? You also need to look at the volatility "smile". Sometimes OTM calls are really cheap relative to OTM puts or vice versa. This obviously affects your strategy selection. Significant skew between months can be a useful factor in strategy selection, as well. Using this information wisely can improve your edge. It is a type of mispricing. For example, during January and February and March, index call options (SPX) were priced at very low values. This meant that Iron Condor strategies were not particularly wise during this time frame, and those who saw the market rise sharply paid a nasty price for selling cheap calls.
I think you guys talk about one and the same thing. While a lot of the successful institutional options traders (sell-side, hedge funds...) make consistent money by buying options they believe are trading below fair value and selling those higher than they perceived fair value the way they identify those are through their models and strategies. So, the price you pay, the timing of entry, how you manage your hedges is a function of the strategy you employ. Simple as that. So, yes, there are is/are some index option models out there that exhibit a statistical edge, meaning they statistically outperform the identification of fair value vs what other traders perceive as fair value. I agree that there is no out-of-the box solution that works. I was taught and learned a model or two that looked like they worked very well but I employed a lot of changes and tweaked here and there to make them work for my trading style.
that is nonsense. The same set ups that some of the top prop desks have in trading index options, for example, can be easily employed at home. The biggest perceived difference is transaction costs and the time to set up those systems. Sometimes extensive programming and development skills are needed to code a platform that easily extracts the implied vols from traded prices, to calibrate my models and then identify other "mispriced" options. But it can all be done.
Well, I was speaking in a relative sense. Here is an example for WFC options: June 17 put: 0.38 June 20 put: 0.90 June 23 put: 1.90 If I bought a June 20 put, I would assume there would be a 50 percent chance that the stock would go down $3 and a 50 percent chance that it would go up $3. It isn't exactly a 50% loss and 100% percent gain, but it's roughly close. I suppose this isn't very accurate since the chance of a $3 move either way isn't really 50-50, but I still assume the chance of a $3 move up is about the same as a $3 move down.
Continuing with the dice analogy here is an excellent explanation (courtesy of Maverick74 on 04-19-06): "Let's try this for a minute. Let's remove the whole idea of the greeks. Let's simply look at options as a bet on fair value. When I interviewed in Chicago for the first time to work for a market making firm on the floor, all the companies asked this question. They said you are going to play a game with one six sided die. You roll the die, whatever number comes up, you get that much money. So if you roll a 3, you get $3. If you roll a 6, you get $6. You are going to play the game over and over again. There are two players in this game. The roller and the house. The house is selling the bet, or selling premium if you will. The roller is buying the juice. The question is, as the roller, how much would you pay for the right to roll the die. Then, when you are the banker, how much would you sell the bet for? For most of you you can figure out this is a very simple probability game that you probably played in your stats class in college. The fair value of the bet is 3.5. You get this number by summing the outcomes and dividing by the total. (6+5+4+3+2+1)/6=3.5 That means the fair value of this bet over a long series of throws is 3.5. So if you are the roller, you want to pay less then 3.5 for the bet, say 3.4. If you are the banker, you would want to sell this bet for more then 3.5, say 3.6. What you just did is you made a market. You are 3.40 bid at 3.60 offer. Folks, this in a nutshell is what options are all about. It does not matter if you are the roller or the banker. The person buying the premium or selling it. If you buy the option below fair value, you will make money. If you sell it for more then fair value, you will make money. Market makers have been doing this since 1973 and nothing has changed since then except the technology. The whole idiocy of buying vs selling premium is as bad as the whole red state/blue state garbage. If you can grasp this very simple concept, trading options will become much more simpler for you. Yes, over the long run, the buyer of premium will generally outperform the premium seller for one reason and one reason only. Not because he/she is a better trader. But because of something called luck. That's right, luck. Luck, is very much like volatility in that it doesn't have a positive or negative bias. It simply is what it is. You will have both good and bad luck in your life. The difference here is that when you have good luck as the long premium trader, you might retire off of it. Bad luck to the option seller will bankrupt him/her. Good luck will do nothing for the option seller as there is very little upside in what they are doing." http://www.elitetrader.com/vb/showthread.php?s=&postid=1044299&highlight=dice#post1044299 Everyone here seems to have their own definition of 'statistical edge'. Placing any option spread is not a "set it and forget it" strategy. Anything can happen at any time and you need the "edge" (i.e., money management ability) to modify/adjust your position when and if needed.
Hi optionsgirl, This depends on the delta of the options. itm options has more delta (positive for call or negative for put) and otm options has less delta. The combination of delta and gamma of the options can estimate how fast you gain and lose money if you only based on movement of the stock excluding consideration of vega and theta. Chee Yong