Hello All, I've been a visitor/reader of ET for quite some time and finally became a member today. I find this and other forums (TSWorld, Wilmott, etc.) to be great "real world" resources where intelligent commenatators actually provide pragmatic ideas and advice. Thus this post. My colleague and I are thinking about building options strategy/position backtesting software. Not really for commercial ends but who knows in the end? (BTW, does anyone know of existing software that can aid in testing complicated strategies?) To produce realistic results, at least from our perspective, we want to build in multiple methods for dynamically hedging gamma. So I post here to solicit advice/ideas for those that are willing to share them. We are hoping to work in some basic approaches at first followed by more sophisticated methlods down the line. We've done fairly well over the years managing positions with "informed intuition" if you will, but that approach is impossible for us to build into backtesting. We are hoping that "systematic" based traders will be able to advise us accordingly. Thanks in advance for your time/consideration.
Dynamically hedging short gamma would be an arbitrage-like money machine...not possible (legally) in an efficient market. The schemes that I have seen that claim to do this actually have you buying high and selling low, which doesn't sound too attractive to me. One takes gamma risk in exchange for + expectation positions. The only "hedge" that has ever worked for me has two parts: 1) Set stops in advance 2) Honor thy stops. Then you live to trade another month. The options seller ONLY edge is timely, disciplined intervention. (Unless of course you are Riskarb who has a neural net to God.)
Hi Tempus, You wrote: Dynamically hedging short gamma would be an arbitrage-like money machine...not possible (legally) in an efficient market. The schemes that I have seen that claim to do this actually have you buying high and selling low, which doesn't sound too attractive to me. One takes gamma risk in exchange for + expectation positions. The only "hedge" that has ever worked for me has two parts: 1) Set stops in advance 2) Honor thy stops. Then you live to trade another month. The options seller ONLY edge is timely, disciplined intervention. (Unless of course you are Riskarb who has a neural net to God.) I HEAR you loud and clear on the timely, disiplined intervention. That's basically what I've attempted to do myself over the years. I think one of the problems is the nature of the problem itself and the fact that most people's default position is to try to "trade the breakout". What I mean is that most traders have researched ways to go long the break, time the breakout but most haven't tried to predict the duration of the consolidation! (possibly incredibly valuable information if done with accuracy) That's why I'm hoping to receive feedback from the few that may have done so. I've tried on a rudimentary level to do it myself but its has never come close to priority one. I wish it had as then my backtesting software project would have a potentially solid basis from which to operate. Thanks for contributing.
Will, Optionvue allows you to backtest data 3 years. It has all the bids, offers, last price, volatility numbers, just as if it was real time. You can backtest day by day, or even by the hour or half hour. You might want to check that out.
Will, another thing, I know I have said this before and I'll say it again in case you never read my previous post. But it is my experience that trading options successfully really requires you to be a good underlying trader. Options just adds edge to a good underlying trader. Backtesting strategies always sounds sexy on paper, but if you can't trade the underlying, options will not give you any edge. Short gamma traders tend to be great trend traders and long gamma traders tend to be good bottom and top pickers. My advice would be to backtest good stock trading strategies and then apply the proper gamma (long or short) to give you the desired edge you are looking for. Just my two cents.
Hello Maverick74, Thanks for responding and for your insight not only in answering my post but across the forums and threads. I agree with you BTW regarding your observations of long and short gamma traders. Thanks for the advice on trying out OptionVue. I didn't know that it was that powerful and will look into it. Maybe if I provide the forum with more specifics you can get a sense of where I am going (to what level) of backtesting. BTW, this would be just one of several examples as things are wide open at the moment. As I am sure that you know, "backers" from outside and some from inside the options community can become fixated on delta neutrality. It seems to me that it's a lot easier to convince a financier to lay capital on the line if you extol a delta neutral approach. This seems to resonate with them as does theta neutrality considering the contraction of volatility and premiums. Anyway, if I were to manage a large number of positions I'd need to handle in theory both short and long gamma positions based upon the way these backers would expect me to trade. They would also like to see delta neutrality at the end of the day. So even if I have a few clerks assisting me I'd need to describe some logical modus operandi to both them and the investors. Plus if I have a few sound methods I can see which ones function the best in what environments. I have theories but I don't have answers and thats where I am hoping this forum and backtesting can help me, subscribing to the many heads on one subject theory. I think I could do a decent job (based on past performance) of making some type of informed judgement call but that is not really scalable. The problem for me is that order flow may dictate many of the positions that I will acquire so I really can't dictate what I'll put on entirely thus I won't really be able to apply the options position to fit with a backtested stock trading strategy. That is good suggestion however. So thanks again and if anything else comes to mind please let me know if you wouldn't mind.