SPX Credit Spread Trader

[yip1997]

Quote from uglyboy:

One of the things that interested me in this thread was the (semi) numerate discussions and the relative lack of boasting one one hand and abuse on the other. It's a bit sad that this thread is starting to degenerate. Perhaps now that there is a bit more volatility, it may be possible to get back to the OPs original aim- the discussion of credit spreads and related strategies.

In this spirit, I present a diagram that I've marked up, showing the theoretical distribution of prices assumed by the Black Scholes model and Merton's derivation (which is the basis of the theoretical value software you are all using - you are calculating theoretical values right?) Overlaid is a histogram of actual price distributions. This actual distribution is, as you all know, leptokurtotic. Notice that there is a significant difference in the frquency of events at 0, 1 and >2 sigma in real data as compared to that predicted by a logarithmic distro.

So here is a question for the newbies:
1) What does this imply about where you should be buying/selling options in your spreads?

2) Here's a question for the not-newbies: If this is true, why am I telling a bunch of strangers?

Ugly

As a newbie, I don't think we have enough info for making any trading decision based on the statistical distribution.

How are options priced? It never prices on hv. If it is priced based on statistical volatility, then obviously the atm is underpriced. However most of the time, IV is higher than HV, and so it is not easy to draw any trading conclusion from the graph.

I know the answer is naive, and wish you can provide a better answer for me.

 

[rdemyan]

Coach:

I actually get Mike Parnos' newsletter as part of a package deal. I have seen his discussion on the guts strangle before but it was quite some time ago and I didn't have time to really read it. Since he has chosen to openly publish this strategy, I'll see if I can find any additional info on the strategy. Besides it doesn't sound like it is proprietary.

Mike generally does not adjust his iron condors or spreads and that is probably why there is no discussion on this. He usually just gets the funds out as he puts it.

Quote from optioncoach:

Well the SPX is definitely out since ITM options widen the already "bend-me-over" spreads.

One advantage I like is that I can earn interest on the cash deposited in the prop using the ITM options for the Iron Condor but I still want to see what they look like on XEO. They seem appealing on XEO in some comparisons I made.

People have to remember that they are synthetically equivalent in the risk graph perhaps but not equivalent in real practice and that is what I am testing. After all why are some people here arguing that one should be used over the other or that naked puts might be better to use than covered calls. Cause in practice they have different margin requirements and potential adjustment steps. The article never touched upon these specifics except for the interest earned so I want to see what else they have different and in common.

 

[uglyboy]

Quote from cohenmichaela:

Perhaps you enjoy listening (reading) your own words?

This kind of comment is totally un-needed. You obviously have some insight. Why not elaborate on the statistical matter at hand rather than trying to kill the discussion?

 

[uglyboy]

Quote from yip1997:

As a newbie, I don't think we have enough info for making any trading decision based on the statistical distribution.

How are options priced? It never prices on hv. If it is priced based on statistical volatility, then obviously the atm is underpriced. However most of the time, IV is higher than HV, and so it is not easy to draw any trading conclusion from the graph.

I know the answer is naive, and wish you can provide a better answer for me.

I don't think your answer is naive, I think that you are clearly working hard to become proficient in a complex area. I hope that the same can be said of me.

I bring up this difference between the assumptions underlying the B-S formula (i.e. logrithmic price distribution without jumps) versus reality (leptokurtotic with jumps) to to try to stimulate discussion around whether there might be better strikes to sell and buy at.

There is a school of thought on this thread that the sum of probabilities times pay-offs is equal at all strikes. This is the fundamental underpinning of the BS equation, and is the argument against FOTM credit spreads (also the gamma risk). Based on the difference between logarithmic distribution and the leptokurtotic reality, it would APPEAR that buying ATM or FOTM and selling OTM would have an advantage. I'm not saying this is true, just that the model underlying many calculations (the B-S model) is only a model.

Ugly

 

[lloyd111]

Hi all, do any of you fine people have any experience or insight about artificial intelligence trading systems. I have received numerous calls attempting to recruit me into one such system being offered by "my trading futures.com" aka mind engines. Sounds intriguing but I cannot find any reviews either positive or negative about this. Any serious feedback would be much appreciated. Ty in advance.

Lloyd

 

[dagnyt]

Quote from uglyboy:

... it would APPEAR that buying ATM or FOTM and selling OTM would have an advantage.

Ugly

Just thinking out load...

It's not quite that simple. Buying ATM can be very costly when the underlying finishes near that ATM price more often than anticipated.

To compensate for that lost time premium, one could sell twice as many OTM options. Then one completes the butterfly by buying the undervalued FOTM. To gain even more edge, one buys an extra FOTM.

Thus, your data suggests that a good position might be: Buy one ATM straddle and sell 2 credit spreads - long the FOTM and short the OTM (1 or 2 sigma OTM) in both puts and calls.

I'd hate to attempt this, as the intiial debit would be too high to suit me.

I believe the best piece of information one can draw from the data you present is that selling FOTM is a losing proposition.

Mark

 

[dagnyt]

Quote from lloyd111:

Hi all, do any of you fine people have any experience or insight about artificial intelligence trading systems... Any serious feedback would be much appreciated. Ty in advance.

Lloyd

If you were able to develop such a system and it proved itself capable of beating the market, would you offer to 'recruit' others, or would you keep the secret to yourself and make a few billion using it?


Mark

 

[IV_Trader]

Quote from dagnyt:

Good question.

1) Commissions are so cheap (75 cents per option) that I don't need the lower costs of the prop shop.

2) Haircut would allow me to be more efficient with capital. But it also allows larger positions and hence, more risk. I don't want an excuse to carry larger positions.

There is one very good reason for having MM haircut, but I have not made any effort in that direction: I would own some black swan protection in the form put backspreads. I cannot afford to carry them in my retail account. Call backspreads are much cheaper to carry.

(To answer the inevitable question: Diagonal call backspreads can be done with strikes 30 points apart in RUT. That means a single 3 x 2 spread requires margin of 6k. For puts - because I want to collect a cash credit - the strikes are 60 points apart - and a simple 3 x 2 carries a margin requiremnt of 12k. That ties up too much cash to gain only one net put.)

Mark

Mark , if I understand correctly , haircut is also perfect for long gamma scalping. For an owner of the long combo , any buying ( or selling) of stock is "margins free" because prop co ( like Mav's) recognizing stock as a part of delta neutral position .

 

[uglyboy]

Quote from dagnyt:

If you were able to develop such a system and it proved itself capable of beating the market, would you offer to 'recruit' others, or would you keep the secret to yourself and make a few billion using it?


Mark

Unless you were trying to shill it on ET!!

 

[uglyboy]

Quote from dagnyt:


Thus, your data suggests that a good position might be: Buy one ATM straddle and sell 2 credit spreads - long the FOTM and short the OTM (1 or 2 sigma OTM) in both puts and calls.

I'd hate to attempt this, as the initial debit would be too high to suit me.

I believe the best piece of information one can draw from the data you present is that selling FOTM is a losing proposition.

Mark [/B]

Exactly, it would seem to imply you could buy butterflies on either side of the current underlying with the body at 1sigma, and expect.......I want to say positive expectancy, but I'm afraid........an advantage. Similarly, selling at 1 sigma and buying at >2 sigma to create a credit spread might take advantage of this. Again, I don't think this is actually possible.

I agree that the most obvious thing to take home is that unusual events - the ones that make CSs lose money occur more commonly than logarithmic distribution implies - thus the skew.