Gamma Scalping Part One - Pay The Day
I am taking the rest of the week off from writing live blogs and the AM Pit Report, I will be posting a few things here and there. I thought I would post a few of my more popular pieces for option traders to check out. This one is from last year on gamma scalping.
As a mentor and former market maker, I get asked all the time “how should I scalp gamma?” It is a surprisingly difficult question to answer. As a market maker I had very few constraints on my trading activity. I could hedge a back spread any way I wanted. If I wanted to hedge by selling stock, that was simple. I pulled up my stock execution system and unloaded the underlying. If I wanted to trade options against the gamma, I could trade just about any option in the spectrum. If I thought there was more edge in trading the January 2012 month to hedge a front month back spread, I would. If I wanted to hedge calls with puts, or puts with calls, that was also not a problem. Retail traders lack the ability to trade and execute many hedging strategies. I soon realized it really doesn’t matter what the trader uses to hedge. It matters how the trader hedges. While there are endless options to scalp gamma, there are two ways I teach students. One method is for very active traders that I call “pay the decay” the other is for less active traders I call “delta/gamma ratio hedging”.
Pay The Day
When I first started teaching gamma scalping, the only method I taught was “pay the day.” This is an intense and involved form of scalping and is probably more useful for full time traders than retail traders. All traders should understand the technique.
The trader uses theta of the position to calculate his daily “nut”. The trader uses the formula for change in slope to calculate how movement is needed in the underlying to “pay the decay.”
Looking at the blind straddle on 10/16/09, the position is long 8.25 gamma and short 6.09 theta. The trader would plug in 7/5*6.09 as the solution (7/5 to take the weekend decay somewhat into account), and use the gamma as the change in slope for the curve. The variable the trader is solving for is the change in price. The formula ends up looking like this
7/5 * Theta = 0.5 * Gamma * X ^ 2
Or solving for X:
X = SQRT ( 7/5 * Theta / 0.5 * Gamma )
Which reduces to:
X = SQRT ( 2.8 * Theta / Gamma )
Solving the equation In this case:
X = SQRT ( 2.8 * 6.09 / 8.25 ) = $1.43767
The SPY needs to move $1.44 in order to cover the theta decay for that day. This formula has to be re run every morning as decay and gamma change (both get larger every day). A major complaint is that anyone who has a ‘real job’ may not feel like doing this calculation daily. There is a much bigger problem with this formula though:
It doesn’t answer when to adjust.
Most of my students assume that they should set the scalp $1.43 from the previous day’s closing price. Not the case! Setting the scalp that far will only get hit about once every 3 days.
Personally, I set my scalps at 50% of the required move, and sell all of my deltas at that point. Then I buy them back when the stock moves back to unchanged. This creates two scalps that equal 72 cents of movement. I have to make this type of scalp twice, either round tripping the underlying moving 72 cents twice, or getting a combo of round trips to the upside and the down side.
At day’s end, I always zero out deltas.
If the underlying ‘gaps’ past the required move, I sell all of the position’s deltas (or at least 75% if I think the move could continue).
If the underlying hits a scalp and continues to run in that direction, I use the scalping point as my new starting point. This method requires a lot of work and can be very frustrating when the underlying really runs (I’ll be honest, if I sense momentum I will let the underlying run and set trailing stop orders instead).
This type of scalping will certainly cover a portion of my decay and surprisingly far more likely to cover my decay then setting the scalps $1.43 apart.
Why? Volatility predicts price movement not direction.
Scalping gamma at closer intervals I end up making far more scalps than setting the scalps further apart. When I was on the floor, trading Sun Micro systems, there were times where I would scalp 10 to 30 times in a day. By the end of the day I could have thousands of dollars in my pocket, even if SUNW (the old symbol) didn’t move. One of the neat things about long gamma is that the trader wants to get in as many scalps as possible. This method allows for that.
Stay tuned part 2 of gamma scalping which will be posted tomorrow
Don't forget to check out our events section the TradeMonster Event is approaching fast
选择“Disable on www.wenxuecity.com”
选择“don't run on pages on this domain”
