How do Market Makers determine the Option Price in binary Situations?

[Atikon]

Hello,

as far as I understand, Market Makers are bound to offer Bid's and Ask's for certain Securities no matter what to keep Liquidity flowing. Which must represents a challenge especially in Binary Event Situations. Let's says, for arguments sake, Boeing during the Corona Dip was a binary Event, how do MM price the Option?

From what I could gather, the IV Curve in Binary Events will look somewhat like a Hill, the ATM IV grows while the OTM IV remains somewhat stable. Do MM really use the BSM to determine the Price of the Options? Or do they use Binary Models and implement the Pricing into the BSM via e.g. IV? If the latter is true, how can one scan for those Pricings/identify those Pricings? What does the Demand Implied Probability Distribution would look like in these Situations (fat tails on both end/on the end that is more probable for Market Partifcipants)? Would Gamma look similar to the IV Curve? Will Gamma and IV contract the closer the Option is to Maturity? Or would a flatten Curve remain till the End because of the Binary Character of Boeings Fate?

Best

Atikon

 

[traider]

vol skew

 

[Atikon]

vol skew

Do you mean as in a reactionary approach? The Vol Skew would come to be if plenty of ppl would buy the ATM/all underpriced Options between the Tails and the IV would blow up, but in the meantime the MM would loose money on all those underpriced ATM Options. Wouldn't MM be aware of such binary Corporate Events and set the price in advance?

 

[MrMuppet]

Pricing is 10% math and 90% inventory control.
Once the MM gets hit by an order, he hedges and shifts his quotes. One sided flow is usually hedged by positions that are +theta so that the only job is just hanging on to your inventory without blowing up.

MM position is usually short ATM and long wings in a vega neutral fashion. As soon as IV explodes it profits from positive vega convexity.

During coinflips you just widen your quote and post the minimum size

 

[Atikon]

Pricing is 10% math and 90% inventory control.
Once the MM gets hit by an order, he hedges and shifts his quotes. One sided flow is usually hedged by positions that are +theta so that the only job is just hanging on to your inventory without blowing up.

MM position is usually short ATM and long wings in a vega neutral fashion. As soon as IV explodes it profits from positive vega convexity.

During coinflips you just widen your quote and post the minimum size

What order needs to hit the MM to make him shift quotes? That must be a material order right? So the MM will become aware by such Corporate Events through the buying behaviour of Market Participants? We are talking about a reactionary approach then?

 

[ironchef]

Do you mean as in a reactionary approach? The Vol Skew would come to be if plenty of ppl would buy the ATM/all underpriced Options between the Tails and the IV would blow up, but in the meantime the MM would loose money on all those underpriced ATM Options. Wouldn't MM be aware of such binary Corporate Events and set the price in advance?

vol skew

You are right, I lived through this on the CELG/BMY merger. The actual distribution was not lognormal but with fat tails on both end - vol skew?

I assumed MM hedged their positions to be delta, gamma, vega neutral? Since I am not in this business, someone in this business please help me out. Thanks.

 

[Atikon]

View attachment 229288


You are right, I lived through this on the CELG/BMY merger. The actual distribution was not lognormal but with fat tails on both end - vol skew?

I assumed MM hedged their positions to be delta, gamma, vega neutral? Since I am not in this business, someone in this business please help me out. Thanks.

Is this hull/colwell? Does it have a description for gamma/vega curves during such events?

I'm pretty sure they talk about the distribution not the volatility scew in that picture. Vol would blow up atm because the move is atm would be so giant, I think that what he meant by volatility scew. But I'm more interested in how market makers get to the scew, if they set it by themselve or let the buy action determine it.

 

[Atikon]

View attachment 229288


You are right, I lived through this on the CELG/BMY merger. The actual distribution was not lognormal but with fat tails on both end - vol skew?

I assumed MM hedged their positions to be delta, gamma, vega neutral? Since I am not in this business, someone in this business please help me out. Thanks.

 

[gaussian]

Is this hull/colwell? Does it have a description for gamma/vega curves during such events?

I'm pretty sure they talk about the distribution not the volatility scew in that picture. Vol would blow up atm because the move is atm would be so giant, I think that what he meant by volatility scew. But I'm more interested in how market makers get to the scew, if they set it by themselve or let the buy action determine it.

There are also volatility models that are robust to jump conditions. Look into the Yang-Zhang volatility estimator. I'd imagine MMs would use some form of either this or SABR to price their options.

 

[MrMuppet]

What order needs to hit the MM to make him shift quotes? That must be a material order right? So the MM will become aware by such Corporate Events through the buying behaviour of Market Participants? We are talking about a reactionary approach then?

No.
You need to understand that MM's do not have unlimited money. Therefore they cannot just quote the "right" price over and over when they get hit. Once they are out of margin, what are they gonna do?

As a MM you want do put your quotes where there is two sided flow. This basically means you want to buy and sell as often as possible, but you want your position to be as close to zero as possible.

So when your quote only attracts buyers and no sellers, you are quoting too low. Therefore you increase your offers and your bid until you get hit on both sides.
The models are NOT telling you the "correct" price of an option, since you would need to know the future distribution of the underlying asset.

Models tell you how an option is priced compared to other options so you hedge one with the other. Think of a simple DV01 calculation for bond futures. DV01 gives you a framework to compare the 2 year with the 10 year so you can establish a relationship and hedge one with the other.

Pricing models don't give you an edge, you still need to be able to trade.

When it comes to corporate events, you simply look at a calendar and widen/size down your quotes if you're flat. In case you already have positions, you just flatten your vega buckets and wait until it rains money.

If you want to learn more about options MM, try to beat this simulator: http://www.volcube.com/