Watch HowardCohodas Trade Index Options Credit Spreads

[HowardCohodas]

I used to call this "hand-waving"... in that you wave goodbye at any and all factors over x^3 :cheesy:

From the movie, "The Day the Earth Stood Still." The Michael Rennie, Sam Jaffe version.

Barnhardt: Have you tested this theory?
Klaatu: I find it works well enough to get me from one planet to another.

 

[arabianights]

I dunno **** about that movie, but if you used newtonian models you could get to the moon, or probably jupiter, but certainly not to, say, HD156668b.

See what I'm saying?

 

[Martinghoul]

What I don't understand is how come this implied skew can be so far out as to offer consistent 10% per month. Does that mean pricing is buggered or that there's expectation of something coming or what?

Ah, but there's the rub... In general, as you'd imagine, skew is priced correctly by the mkt. That is to say fading skew works (offers outsize returns) until it doesn't (blows you up, like what happened, among other blowups, with LTCM). Hence the question with Howard's strategies is whether what he's doing is a blowup waiting to happen or whether he's found a way to filter the specific instances where the mkt is overvaluing skew. If it's the latter, it's genuine honest-to-god edge.

 

[Martinghoul]

When I did my calculations using Monte Carlo methods, of course a distribution must be chosen. I experimented with several as I read the academic papers on modeling option prices.

I studied mathematics in college and practiced as an engineer most of my career. The biggest difference in the two professions is their view of math. This can be summarized by the phrase, "however, for all practical purposes."

I wholeheartedly agree with you, Howard, in terms of striking the right balance between rigor and practicality. My background is also engineering, so I understand. I will read your FAQ, but Monte Carlo doesn't really tell me anything about the underlying model. Monte Carlo is just the method, whereas what matters is the setup of the simulation. For example, using Monte Carlo with a simple BSM framework (e.g. constant volatility, simple GBM) is unnecessary overkill. So maybe you can clarify what the MC thing is all about.

 

[HowardCohodas]

I dunno **** about that movie, but if you used newtonian models you could get to the moon, or probably jupiter, but certainly not to, say, HD156668b.

See what I'm saying?

Yes I do.

Movies require a willing suspension of disbelief, but there is often wisdom buried within.

 

[arabianights]

Yes I do.

Movies require a willing suspension of disbelief, but there is often wisdom buried within.

No, that's not what I'm saying at all... however you've been ever so eloquent in saying completely the wrong thing, so I suppose you deserve some credit

 

[HowardCohodas]

13 DEC 2010 Trading Summary

Iron Condor 8
Has only three days of life left. Since I have been trading weeklies, I don't detect much difference in the dynamics of the weeklies vs. the monthlies in their last week before expiration. None-the-less, I asked a premium credit to complete the IC. I was sure my order would expire worthless when I saw the fill within about a second of the closing bell. So I added a spread to IC 8.

21  10/27/10            DEC 10  RUT CALL   800   810   16%    55%   2.9%    8.C
16  10/18/10  11/23/10  DEC 10  RUT PUT    590   600    1%    86%   9.1%    8.P  Closed @ 86% - Rolled to reform IC
37  11/23/10  12/03/10  DEC 10  RUT PUT    650   660    0%    85%   8.6%    8.P  Closed @ 85% gain - Rolled to reform IC
42  12/03/10  12/10/10  DEC 10  RUT PUT    700   710    0%    87%   7.0%    8.P  Closed with intent to roll
57  12/13/10            DEC 10  RUT PUT    740   750   26%    19%   1.3%    8.P  Opened to complete IC

Iron Condor 13
A roll opportunity was presented and executed. An spread to complete the IC was also executed.

56  12/13/10            DEC5 10 SPX CALL  1300  1325   22%    9%    0.3%   13.P  Opened to complete IC
30  11/12/10  12/13/10  DEC5 10 SPX PUT   1075  1100    0%   89%    6.9%   13.P  Closed with intent to roll
54  12/13/10            DEC5 10 SPX PUT   1175  1200   39%  (19%)  (1.9%)  13.P  Open

Iron Condor 16
A roll opportunity was presented and executed. An spread to complete the IC was also executed.

50  12/08/10            JAN 11  RUT CALL   840   850   26%  (56%)  (2.3%)  16.C
34  11/22/10  12/13/10  JAN 11  RUT PUT    610   620    0%   81%    8.0%   16.P  Closed with intent to roll
55  12/13/10            JAN 11  RUT PUT    695   705   24%  (33%)  (2.7%)  16.P  Opened to complete IC

 

[HowardCohodas]

14 DEC 2010 Trading Plan derived from Dashboard

Opportunity
    Incomplete Iron Condors
        None
        
    Roll Candidates
        17.P - JAN 11  - PUT - SPX

    New Spreads
        None

Jeopardy
    Expiration - 2 days
        IC 6
        IC 8
        IC 12

 

[Hotch]

Re: Is Howard a Bumblebee?

For the longest time aeronautical engineering analysis "proved" that a bumblebee could not fly.

This is a myth Howard. I find it unnerving that you'd believe such crap.

 

[HowardCohodas]

Re: Is Howard a Bumblebee?

This is a myth Howard. I find it unnerving that you'd believe such crap.

I suggest you do some research. I have. If your research is better than mine and I am wrong, I will gladly apologize.

 

[Martinghoul]

The third FAQ (link in signature) describes how I came to the concept of probability of touching, how I calculated it and why I now rely on the TOS model.

Howard, having read the FAQ, I am not much better off in terms of understanding your methodology for calculating the "probability of touching", which is a concept that's instrumental to your strategy. Basically, you have given no specific details on the methods you have used (and abandoned) and TOS calculation is proprietary. I would imagine that there's nothing proprietary about it, as it's probably just based on the Reiner-Rubinstein barrier option pricing paper, written in 1991. The formula for the risk-neutral probability of touching, based on BSM assumptions, is nothing fancy (you can find it on page 81 of Haug's "Guide to Option Pricing Formulas"), but unfortunately it has all the flaws of BSM.

 

[HowardCohodas]

Howard, having read the FAQ, I am not much better off in terms of understanding your methodology for calculating the "probability of touching", which is a concept that's instrumental to your strategy. Basically, you have given no specific details on the methods you have used (and abandoned) and TOS calculation is proprietary.

The concept of "Probability of Touching" is indeed critical to my methods, but it's precision is not. The sensitivity analysis I did indicated little change in result with several percentage points change in the probability of touching estimate. So when I started at 10% probability of touching, it was really 10%-ish.

I recently found an online calculator that seems similar to what I did. Monte Carlo Probability Calculator

 

[Martinghoul]

Right, thx... As I mentioned before, Monte Carlo doesn't matter, so this link doesn't really do anything for me. What matters is the assumptions regarding the dynamics of the underlying process. If you're using the good ol' GBM assumption, Monte Carlo, in spite of sounding marvelously clever, doesn't offer much more insight than closed-form BSM. If you told me that you were actually using Monte Carlo with a stochastic vol model, e.g. Heston or SABR, that would be a bit more interesting.

At any rate, this is all academic. What I am really trying to understand is what you believe to be your edge. You have said now that it's not the systematic mis-pricing of call/put spreads (although I thought that's what you alluded to in an earlier post #170). It appears as if it's not the filtering process by which you arrive at what spreads are "safe" to sell and which aren't, given that your strategy is not very sensitive to the parameter you use for filtering. So it must be either a) your true expertise is risk management, i.e. knowing when to cut and run; or b) something else entirely; or c) your strategy is a blowup waiting to happen. Assuming it's not c), pls help me understand whether it's a) or b).

 

[Hotch]

Re: Is Howard a Bumblebee?

I suggest you do some research. I have. If your research is better than mine and I am wrong, I will gladly apologize.

http://en.wikipedia.org/wiki/Bumblebee#Myths

Gee whiz, that was hard.

:clap:

P.S. YES EVERYTHING ON WIKIPEDIA IS TRUE

 

[HowardCohodas]

Re: Is Howard a Bumblebee?

http://en.wikipedia.org/wiki/Bumblebee#Myths

Gee whiz, that was hard.

:clap:

P.S. YES EVERYTHING ON WIKIPEDIA IS TRUE

Thank you for choosing a reference that proves my point. You even got to the correct section. Perhaps you did not read it completely because it essentially states what I said in my post.

If you need further convincing because of some prejudice against Wikipedia, there are other good ones as well.

 

[meanreversion]

You're skirting the issue again, Mr C.

If the options market is efficiently priced, what is the source of your edge? In other words, if you expect to make money in the long run and it doesn't come from any mis-pricing, where does it come from?

Are you using probability of touch as a proxy for the chance of finishing beyond the strike?

 

[HowardCohodas]

So it must be either
a) your true expertise is risk management, i.e. knowing when to cut and run; or
b) something else entirely; or
c) your strategy is a blowup waiting to happen.

Assuming it's not c), pls help me understand whether it's a) or b).

My thinking on these issues has evolved. My latest thinking on this is in post #174.

Notice my concluding remark.

 

[HowardCohodas]

You're skirting the issue again, Mr C.

If the options market is efficiently priced, what is the source of your edge? In other words, if you expect to make money in the long run and it doesn't come from any mis-pricing, where does it come from?

Are you using probability of touch as a proxy for the chance of finishing beyond the strike?

Have you read post #174?

 

[Hotch]

I don't know what you read howard, but what I read was that people either miscalculated, or simplified the problem so that it was no longer realistic, nothing about "proof".

 

[HowardCohodas]

I don't know what you read howard, but what I read was that people either miscalculated, or simplified the problem so that it was no longer realistic, nothing about "proof".

It was no longer realistic only because more modern modeling tools were developed. At the time, that was the limit of their ability.