"The VIX is quoted in percentage points and translates, roughly, to the expected movement in the S&P 500 index over the next 30-day period, which is then annualized." http://en.wikipedia.org/wiki/VIX one could use VIX to price a straddle two weeks away instead of 30 days away by adjusting the formula mentioned in the wiki article. why should i price a straddle using historical volatility when i have implied volatility available? i guess one could use historical volatility as a proxy for implied volatility. or i could use both. i don't know almost anything about options but the statement that to price a straddle one must use historical volatility instead of implied volatility just does not make any sense to me.
I trade literally millions in option premium a month. I really don't know what the statement "straddles are historical volatility" means. You usually have good input bones, but that one really seems out of left field.
I guess it would be a directional trade, but on volatility? Personally, I use all these vol measures the way Barbosa used the Pirate Code: as guidelines, nothing else. I settled on a probability distribution that I think explains options very well, that's close to but not quite the normal distribution that all this math is based on. These measures are decent approximations, but don't really do justice to what's really happening. What I find really weird is that I find no mention of this probability distribution even over on Wilmott. Of course, as far as I'm concerned, that's a good thing. In the void between the normal distribution and actual, real life you can make money, if you're careful.
Yeah, twenty years ago that was my first option strategy. I ended up losing my (modest) account, and had to start over again.
Here is a track record of shorting SPY Straddles: http://www.arrowpublications.net/Calculations_Straddles.html Any comments please?
Whooo... finally someone reveal an edge in this site. I backtest the strategy and it work well (although somehow my broker refuse to supply me any data for 2008 when I share this strategy with them, and promise them I will trade with ton of contracts and fees .. :eek: ) Thanks buddy for sharing your edge!
But be careful with this strategy. I've read all Paul Kadavy's books and exchanged a few emails with him. He is a very nice person and always tries to answer your questions in details. That being said, I don't think a mechanical SPY Straddle Selling strategy can survive the 2008 market crash. Even the first 4 month of 2009 would have broken the system. Starting tracking record from May 2009 was really lucky because the market entered into a period with historical volatility (i.e. actual volatility) lowed than implied volatility (i.e. forecast volatility used in options pricing). Straddle selling generally makes money under this condition. A noticeable example of the potential risk of this model happened in May 2010 when May 6 flash crash caused the model down 13.96% in that month - a monthly draw-down unacceptable by any serious trader. On that note, Paul does provide a few tricks to reduce the risk of his model - One being stop loss (buy-back straddle when SPY moves outside of the predefined price/percentage range), and another being economic outlook (i.e. the model is ideal under current economic situation where it's slightly long-biased). I do think there are plenty of rooms to enhance the model. For example: Use strangle selling instead of straddle selling. Use more sophisticated position management models. Use probability cone model to forecast strangle strikes. So on and so forth. However, these more advanced models generally have more Art elements in them, and will be more difficult to back-test. All in all, this SPY Straddle selling model is one of the simplest, 100% mechanical models, that have a clear edge. One of the best under current market conditions. All thanks go to Paul!