How Do Options Make Predictions?
- Thread starter kjb1891
- Start date
Hi guys,
I don't come to this site very often, but I found this thread fairly interesting and thought I would add my two cents and see what other people have to add to what I have to say.
There was something someone said a while back that I thought was particularly insightful:
This is a pretty good explanation for IV across strikes from my experience, and almost the sole reason for why skew gets so out of control when there is a great deal of fear in the market or when close to expiration.
I work for a major MM in Europe. I also think that, despite the fact we dictate the prices in the market, as a whole we aren't a very analytical bunch. There is plenty of money to be made as an MM without the need for the extra resources required for mathematical analysis.
As an MM we set price as a follower based on what other participants are doing in the market. We tend to have no opinion on what the absolute volatilities for each strike should be, and the thought has often occurred to me that mostly everyone else in the market has no clue either and are usually trading to hedge out their given risks, arbitrage against other markets or similar such non-speculative activity. I have no idea how much volumes are based on pure speculation from the 'smart money' players and to be honest I'd be very curious to know.
I found Maestro's 'poetic' descriptions of how option skews operate to be quite interesting, despite my reservations about some of his more outlandish statements. I just want to ask you this, when you describe 'option chains' as operating like a swarm i take that to mean that you feel that the vols in that chain should operate in an organised manner. If a strike vol does not follow the 'swarm' you can trade on that against the other more well behaved strikes. This makes sense to me.
Some other things you say also seem to suggest that you go further than this and that you can extrapolate from movements in the skew predictive power about what the absolute values of the vol or the underlying or should be. If this is the case, then that is beyond anything that I have personally witnessed and is an interesting idea, but I just wanted to clarify that this is actually what you meant.
I don't come to this site very often, but I found this thread fairly interesting and thought I would add my two cents and see what other people have to add to what I have to say.
There was something someone said a while back that I thought was particularly insightful:
This is a pretty good explanation for IV across strikes from my experience, and almost the sole reason for why skew gets so out of control when there is a great deal of fear in the market or when close to expiration.
I work for a major MM in Europe. I also think that, despite the fact we dictate the prices in the market, as a whole we aren't a very analytical bunch. There is plenty of money to be made as an MM without the need for the extra resources required for mathematical analysis.
As an MM we set price as a follower based on what other participants are doing in the market. We tend to have no opinion on what the absolute volatilities for each strike should be, and the thought has often occurred to me that mostly everyone else in the market has no clue either and are usually trading to hedge out their given risks, arbitrage against other markets or similar such non-speculative activity. I have no idea how much volumes are based on pure speculation from the 'smart money' players and to be honest I'd be very curious to know.
I found Maestro's 'poetic' descriptions of how option skews operate to be quite interesting, despite my reservations about some of his more outlandish statements. I just want to ask you this, when you describe 'option chains' as operating like a swarm i take that to mean that you feel that the vols in that chain should operate in an organised manner. If a strike vol does not follow the 'swarm' you can trade on that against the other more well behaved strikes. This makes sense to me.
Some other things you say also seem to suggest that you go further than this and that you can extrapolate from movements in the skew predictive power about what the absolute values of the vol or the underlying or should be. If this is the case, then that is beyond anything that I have personally witnessed and is an interesting idea, but I just wanted to clarify that this is actually what you meant.
Like you, the only way I can make sense of an option chain is by looking at the IV pattern strike to strike - the skew. Absolute prices mean nothing to me. Maestro however seems to be saying he is looking for patterns in the prices themselves, independent of IV. But I should probably just let Maestro answer for himself.
This simplicity of this method I developed several years ago might be helpful to the originator of this tread in answer to his original question: "How Do Options Make Predictions."
Here is an interesting method for predicting short term SPX
direction for the next trading day, using option volume and open interest on highly active and expensive March "SPX options."
Lets do a real example using Friday's (02/27) closing data:
Date...Option...Volume.....O. I........New O.I......% Move into O.I.
2-27....750C......28821.....33217......?????.............??%..............
2-27....750P......20111.....107467.....??????...........??%.............
On Monday morning before the opening bell, the CBOE will re-calculate the new Open Interest (O.I.) by adding or subtracting
any of Friday's option "Volume" that was held over the weekend.
If some or all of the option volume was "held," the O.I. number should be larger on Monday morning.
If the option volume represented "Selling" Volume, the O.I. number should be smaller on Monday morning.
As an example of what we are looking for, let's "pretend" its now Monday morning and volume has been reset into open interest (O.I.). Here is some hypothetical changes in O.I. that would generate a buy signal:
Date...Option...New O.I....% Move into O.I...Advantage...Edge
3-02....750C.....50550...........+60%................+53%.........Call
3-02....750P.....108880.........+07%
In the above hypothetical example (which hasn't really occurred yet), Calls have an Advantage of +53%, which I would consider a "Trading Edge" for SPX Call buyers on Monday.
-----------------------------------------------------------------------------
Here's the Simple Math for
[Percentage (%) Move into O.I.]:
Calls:
50550(new O.I.) minus 33217(old O.I) = 17333(contracts that move into O.I.),
17333 divide by 28821(volume) = (.60) +60% O.I. gain.
Puts:
108880(new O.I.) minus 107467(old O.I) = 1413(contracts that moved into O.I.)
1413 divide by 20111 (volume) = (.07) +7% O.I. gain.
--------------------------------------------------------------------------
[Math for Advantage]:
+60%(calls) minus +7%(puts) = +53% Advantage for Calls.
---------------------------------------------------------------------------
[Trading Edge]:
An Advantage less than +25%< doesn't have much predictive value.
An Advantage +25% to +49% has a Trading Edge with a reliability of +60 to +69% if traded in the direction of the Advantage that trading day.
An Advantage +50% to +74% has a Trading Edge with a reliability of 70% if traded in the direction of the Advantage that trading day.
An Advantage of +75% or greater has a Trading Edge with a reliability greater than >70%.
-----------------------------------------------------------------------------
This system is based on a very simple principle:
"If someone can afford to pay $2,000 per contract for a multitude of SPX option contracts with not much time left before expiration,
how long would they last as traders if they were frequently wrong?"
These guys are what I refer to as "smart money."
The question is:
"Do you want to follow smart money or the general crowd?"
Here is an interesting method for predicting short term SPX
direction for the next trading day, using option volume and open interest on highly active and expensive March "SPX options."
Lets do a real example using Friday's (02/27) closing data:
Date...Option...Volume.....O. I........New O.I......% Move into O.I.
2-27....750C......28821.....33217......?????.............??%..............
2-27....750P......20111.....107467.....??????...........??%.............
On Monday morning before the opening bell, the CBOE will re-calculate the new Open Interest (O.I.) by adding or subtracting
any of Friday's option "Volume" that was held over the weekend.
If some or all of the option volume was "held," the O.I. number should be larger on Monday morning.
If the option volume represented "Selling" Volume, the O.I. number should be smaller on Monday morning.
As an example of what we are looking for, let's "pretend" its now Monday morning and volume has been reset into open interest (O.I.). Here is some hypothetical changes in O.I. that would generate a buy signal:
Date...Option...New O.I....% Move into O.I...Advantage...Edge
3-02....750C.....50550...........+60%................+53%.........Call
3-02....750P.....108880.........+07%
In the above hypothetical example (which hasn't really occurred yet), Calls have an Advantage of +53%, which I would consider a "Trading Edge" for SPX Call buyers on Monday.
-----------------------------------------------------------------------------
Here's the Simple Math for
[Percentage (%) Move into O.I.]:
Calls:
50550(new O.I.) minus 33217(old O.I) = 17333(contracts that move into O.I.),
17333 divide by 28821(volume) = (.60) +60% O.I. gain.
Puts:
108880(new O.I.) minus 107467(old O.I) = 1413(contracts that moved into O.I.)
1413 divide by 20111 (volume) = (.07) +7% O.I. gain.
--------------------------------------------------------------------------
[Math for Advantage]:
+60%(calls) minus +7%(puts) = +53% Advantage for Calls.
---------------------------------------------------------------------------
[Trading Edge]:
An Advantage less than +25%< doesn't have much predictive value.
An Advantage +25% to +49% has a Trading Edge with a reliability of +60 to +69% if traded in the direction of the Advantage that trading day.
An Advantage +50% to +74% has a Trading Edge with a reliability of 70% if traded in the direction of the Advantage that trading day.
An Advantage of +75% or greater has a Trading Edge with a reliability greater than >70%.
-----------------------------------------------------------------------------
This system is based on a very simple principle:
"If someone can afford to pay $2,000 per contract for a multitude of SPX option contracts with not much time left before expiration,
how long would they last as traders if they were frequently wrong?"
These guys are what I refer to as "smart money."
The question is:
"Do you want to follow smart money or the general crowd?"
How do you know if these calls were buy to open or sell to open transactions?
No I don't know and don't care.
Even if these transactions were sell to open transactions, that means "somebody" is still buying them.
Who in their right mind would buy $2,000 option contracts in large numbers with very little time left before expiration, unless they are smarter than the average bear?
I can only say, that in using the Advantage and Reliability figures on my last post, this system worked for myself and friends of mine for a number of years.
Infrequency of trade signals caused me to develop other systems with more frequent trade signals that superceded the prior system.
Jeff had a thread some time back where he posted his calls in real time. He was phenomonally accurate, considering that his stated trade logic seemed to make little sense. Maybe he knows but isn't telling.
Have you posted similarly accurate market calls? If not then I ask who of you two really is clueless?