Gamma Scalping
- Thread starter skadinkus
- Start date
Here is an example that might help regarding gamma scalping.
Assume XYZ stock or soybeans or some other optionable contract has a historical vol of 40% and you are able to buy a straddle around 35% providing you with a theoretical edge. By buying an at the money straddle-long 1 call & long 1 put your delta is around 0 or flat.
You hold on to the straddle and if xyz moves up $2 you will find that the option model has made you delta long 100 shres, so you adjust by selling 100 shares of stock redering you flat again. 2 days later XYZ stock goes down $3, your model gets you short 400 shares-so you equalize by buying 400 shares. Gamma scalping is 'forcing' you to adjust by buying low-selling high.
At the end of the option life, if the vol indeed comes to 40%,all your adjustments to your delta should compensate for the premium you paid for the straddle. In the end you might have spent $5000 for the straddle but made $8000 worth of adjustments.
The real ability is in the predicting of the historical vol, the identification of the options below the hv and the 'art' of adjusting. Ex. If xyz moves $2 down and you're short 300 shares, do you adjust and buy 300 shares now or wait to see if major support will be violated thus giving you better adjustment levels.
Hope this helps.
Assume XYZ stock or soybeans or some other optionable contract has a historical vol of 40% and you are able to buy a straddle around 35% providing you with a theoretical edge. By buying an at the money straddle-long 1 call & long 1 put your delta is around 0 or flat.
You hold on to the straddle and if xyz moves up $2 you will find that the option model has made you delta long 100 shres, so you adjust by selling 100 shares of stock redering you flat again. 2 days later XYZ stock goes down $3, your model gets you short 400 shares-so you equalize by buying 400 shares. Gamma scalping is 'forcing' you to adjust by buying low-selling high.
At the end of the option life, if the vol indeed comes to 40%,all your adjustments to your delta should compensate for the premium you paid for the straddle. In the end you might have spent $5000 for the straddle but made $8000 worth of adjustments.
The real ability is in the predicting of the historical vol, the identification of the options below the hv and the 'art' of adjusting. Ex. If xyz moves $2 down and you're short 300 shares, do you adjust and buy 300 shares now or wait to see if major support will be violated thus giving you better adjustment levels.
Hope this helps.
Natenberg's book Options Volatility and Pricing has a lot of well-written material on this subject, although I don't recall him using the term gamma scalping. Suffice to say, there are a lot of subtleties to this subject.
Has anyone tried this with consistent success in stock options?
I've tried it with good results in the futures market 8 years ago and was wondering if anyone here does it with stock options. I would think there would be a lot of cheap(er) straddles out there since there are a lot of covered call writers in equities which could depress a close to the money strike price making it a good candidate for a straddle buy.
I've tried it with good results in the futures market 8 years ago and was wondering if anyone here does it with stock options. I would think there would be a lot of cheap(er) straddles out there since there are a lot of covered call writers in equities which could depress a close to the money strike price making it a good candidate for a straddle buy.
In practice, it's usually necessary to have a rather large capital base to make "trading the whip" worthwhile. Certainly you'd need options positions that need re-balancings > 500 shares or so...I'd think that transaction costs might be prohibitive anyway.
I trade derivatives and I ran a global derivatives book for a major securities book - for fifteen years a great deal of my group's yearly profits came from scalping gamma (both from the long and short side). You'll have to take into account the cost of gamma rent - more than just theta! - vol exposure with respect to strike and stock price, term structure of vol, slippage, etc...
Historical vol is just that - history. It's a fair place to start, but it's only one part of a trader's toolkit. I've found it more interesting to arb index prems vs. a subset of index members. The initial trade is usually long gamma, delta-neutral(ish), pos or neg vega based on market view and when possible earning theta. It's an interesting long / short portfolio
Cheers,
Marc
I trade derivatives and I ran a global derivatives book for a major securities book - for fifteen years a great deal of my group's yearly profits came from scalping gamma (both from the long and short side). You'll have to take into account the cost of gamma rent - more than just theta! - vol exposure with respect to strike and stock price, term structure of vol, slippage, etc...
Historical vol is just that - history. It's a fair place to start, but it's only one part of a trader's toolkit. I've found it more interesting to arb index prems vs. a subset of index members. The initial trade is usually long gamma, delta-neutral(ish), pos or neg vega based on market view and when possible earning theta. It's an interesting long / short portfolio
Cheers,
Marc
Thanks for your unique insight Marc. Can you give me some clarification about gamma rent? Does this mean vega?
I intend to put on straddles lower than historical vol as well as lower than its own vol. Example today is BA. If HV of BA is 30 and Jan ATM options today are around 27 which is less than HV AND is on the tenth percentile of its own IV since the range of BA's IV is 25 to 50 would it not make sense to buy straddle and scalp gamma? Do you see a flaw in that plan? I would appreciate it if you can offer more insight as to the validity of this. I would buy enough straddles to be able to gamma scalp about 200 shares every $1 move.
Thanks in advance
I intend to put on straddles lower than historical vol as well as lower than its own vol. Example today is BA. If HV of BA is 30 and Jan ATM options today are around 27 which is less than HV AND is on the tenth percentile of its own IV since the range of BA's IV is 25 to 50 would it not make sense to buy straddle and scalp gamma? Do you see a flaw in that plan? I would appreciate it if you can offer more insight as to the validity of this. I would buy enough straddles to be able to gamma scalp about 200 shares every $1 move.
Thanks in advance
When you mean theor straddle do you mean not actually buying the straddle and pretending to scalp it? That sounds dangerous since you won't have the 'insurance policy' to protect you if the underlying keeps going up or down. Regarding option software, there a very few out there that is reasonable. Try optionstar around $250 else everything else is over $700 but it has more features like portfolio tracking, vol charting,.etc
Good luck
Good luck
There are a few calculations to be done:
Rebalancing frequency: How often will the delta hedge be rebalanced over the life of the position? This is very important, especially with respect to the anticipated profit taken at each point. If the expected p/l from rebalancing < costs (commish, slippage, opportunity, theta, delta bleed, vol, etc) then obviously it won't work.
HV vs. IV: Historical vol is where we've been. Implied is the market's best guess as to where we're going. Actual vol is what the market gives you in p/l once you've established a position.
One quick, dirty little tool I use is to de-annualizing the implied vol to a more meaningful statistic (to me, at least). Assume there are 253 trading days in a year (or take whatever number you're most comfortable with). Since vol is the annual SD of the log returns process:
Implied Vol / sqrt(Trading Days in Year) = Implied % Range for the day
Or, if the implied is 32%, the market expects that the stock will move a little more than 2% / day. Look back over your time series and compare the H - L range with the implied range / day.
Has the stock given you an opportunity to capture enough gamma p/l to make it worthwhile?
Just two thoughts on selection and trading. Selecting the right maturity and strikes is another big topic. If anything is unclear feel free to send me a message
Cheers,
Marc
Rebalancing frequency: How often will the delta hedge be rebalanced over the life of the position? This is very important, especially with respect to the anticipated profit taken at each point. If the expected p/l from rebalancing < costs (commish, slippage, opportunity, theta, delta bleed, vol, etc) then obviously it won't work.
HV vs. IV: Historical vol is where we've been. Implied is the market's best guess as to where we're going. Actual vol is what the market gives you in p/l once you've established a position.
One quick, dirty little tool I use is to de-annualizing the implied vol to a more meaningful statistic (to me, at least). Assume there are 253 trading days in a year (or take whatever number you're most comfortable with). Since vol is the annual SD of the log returns process:
Implied Vol / sqrt(Trading Days in Year) = Implied % Range for the day
Or, if the implied is 32%, the market expects that the stock will move a little more than 2% / day. Look back over your time series and compare the H - L range with the implied range / day.
Has the stock given you an opportunity to capture enough gamma p/l to make it worthwhile?
Just two thoughts on selection and trading. Selecting the right maturity and strikes is another big topic. If anything is unclear feel free to send me a message
Cheers,
Marc
