Sticky Strike or Sticky Delta?

[ironchef]

So that's sticky delta behaviour - for example, you buy an SPX call at 6 vols and when the market rallies vol actually goes up.

For my equities, in general, the ATM IV stayed about the same, so when the underlying moved the same IV appeared at the new ATM unless earning date approached, and unless expiry is near. On occasions, when the underlying gapped up/down, IV went crazy.

Since I exclusively trade directional, I like stable IV. I tried to model IV over time and vs underlying movement vs historical volatility to find some additional edges. No luck so far.

Do you have any suggestions how I can get a better handle on IV?

And I won't say thank you for your help this time.

Regards,

 

[ironchef]

Here is a good doc from Derman on the subject:

http://www.math.columbia.edu/~smirnov/Derman.pdf

Your link is very helpful. Thanks.

 

[TradingDemystified]

I have seen it explained elsewhere for a move in the underlying (for example, an up move) as:

1.- A shift to the right of the whole IV curve so the ATM stays at the same level (horizontal shift) <--This means that this explanation starts from a sticky-delta assumption
2.- A move up or down of the whole curve. For an up move in the underlying, a push down of all strike's IV's (vertical shift).
3.- A change in the curve's slope (skew). For an up move a steepening (low strikes pushed up and high strikes pushed down) is expected. The opposite (flattening) for a down move.

So essentially a move in the underlying can have its IV curve for a certain maturity modeled as:

Horizontal Shift + Vertical Shift + Skew Adjustment

How close to reality is this assumption? apparently this (or something like this) occurs in around 75% of moves.

BTW, I liked a quote from the Derman paper referenced above:

"Note - you don’t own at-the-money volatility, you own a fixed strike"

Which is why I think it is important to have a good understanding of Volatility Surface Dynamics and not just ATM behaviour, which most (retail) traders seem to lack (myself included).

 

[ironchef]

I have seen it explained elsewhere for a move in the underlying (for example, an up move) as:

1.- A shift to the right of the whole IV curve so the ATM stays at the same level (horizontal shift) <--This means that this explanation starts from a sticky-delta assumption
2.- A move up or down of the whole curve. For an up move in the underlying, a push down of all strike's IV's (vertical shift).
3.- A change in the curve's slope (skew). For an up move a steepening (low strikes pushed up and high strikes pushed down) is expected. The opposite (flattening) for a down move.

So essentially a move in the underlying can have its IV curve for a certain maturity modeled as:

Horizontal Shift + Vertical Shift + Skew Adjustment

How close to reality is this assumption? apparently this (or something like this) occurs in around 75% of moves.

BTW, I liked a quote from the Derman paper referenced above:

"Note - you don’t own at-the-money volatility, you own a fixed strike"

Which is why I think it is important to have a good understanding of Volatility Surface Dynamics and not just ATM behaviour, which most (retail) traders seem to lack (myself included).

I appreciate you trying to help me out.

I read Derman's paper a few times. It was hard for me to understand the concepts, what he was trying to say let alone how to use that to trade. It is beyond my capability.

So what is a small trader to do? Well, my counter parties are professionals and smart as hell. I let them model the outcome and tell me: They establish bid/ask for a reason, if they buy at bid or sell at ask, they win. Somewhere in between is a neutral position where neither buyer nor seller will have an edge. To trade if I can buy/sell somewhere mid, I have a 50/50 chance.

Indeed after 5 years, my win rate is about 50/50. But why am I profitable? I don't know, maybe because the market since 2009 has a tremendous up bias?