Option replication and exotics journal
- Thread starter riskarb
- Start date
- Status
Bigger size than normal?
Off the top of my head, isn't the 1218 inside this week's range? The 1210 sounds nice and safe.
Not so clear on the risk of 1245 for the next trade though -- like anyone could be.
Being somewhat conservative, I would have hoped to see some sort of pause in this last move up first. Big advantage is we are moving into an expiry week, so hopefully the market will lock into narrowing ranges.Thanks again for sharing the trades.
Quote from ssternlight:
Off the top of my head, isn't the 1218 inside this week's range? The 1210 sounds nice and safe.
Not so clear, on the upside for the next trade though -- like anyone could be.
Thanks again for sharing the trades.
The thought behind the 1218 barrier was to be flat PnL if hit, therein lies the "strong" synthetic designation. I'll reduce hedge if we don't get any weakness on Monday. I expect some of last week's gains to be unwound. I am short quite a few deltas here, but it begins to converge[bleed] with time.
I will be active in managing the hedge.
Yes, I saw the protection aspect. I just figured the likelihood of 1218 vs 1210 as being very different based on recent price action -- always nice to have your cake and eat it too. But I don't know the pricing for a 1210 so it's impossible to be intelligible about the tradeoffs of moving the trade down from 1218.
I guess what I was really asking was when you set up the initial trade did you consider the recent price action/range at all as a factor in setting your barrier, or were you just evaluating available premium, or were you engaged in some more complex/different evaluation process? I was alluding to this when we discussed the 1175 no touch trades awhile back.
Or to put it yet another way, does the premia available show a marked shift as you move the no-touch from just inside the previous week's range to just outside -- being simplistic for a moment about range evaluation.
Quote from ssternlight:
Yes, I saw the protection aspect. I just figured the likelihood of 1218 vs 1210 as being very different based on recent price action -- always nice to have your cake and eat it too. But I don't know the pricing for a 1210 so it's impossible to be intelligible about the tradeoffs of moving the trade down from 1218.
A rough approximation[very rough] is to double each debit req with each sigma otm on single no touches.
I guess what I was really asking was when you set up the initial trade did you consider the recent price action/range at all as a factor in setting your barrier, or were you just evaluating available premium, or were you engaged in some more complex/different evaluation process? I was alluding to this when we discussed the 1175 no touch trades awhile back.
Or to put it yet another way, does the premia available show a marked shift as you move the no-touch from just inside the previous week's range to just outside -- being simplistic for a moment about range evaluation.
Not directly; the model-output best represents the current statistical vol + the implied distro smile of the underlying market -- as gammas/vegas with respect to sensitivity. IOW, the model and the dealer are pricing the expansion in the range[>stat vol] not the artifact of stat vol, or where we were in terms of price. The distro is still assumed to be quasi-symmetrical. The curvature increases dramatically[inverse to vol//cheap debit] when the position is large, long delta. The vol smile reduces the debit based upon proximity/contamination via the put barrier. It becomes a bit more complex depending on delta.
Yes, I saw the protection aspect. I just figured the likelihood of 1218 vs 1210 as being very different based on recent price action -- always nice to have your cake and eat it too. But I don't know the pricing for a 1210 so it's impossible to be intelligible about the tradeoffs of moving the trade down from 1218.
A rough approximation[very rough] is to double each debit req with each sigma otm on single no touches.
I guess what I was really asking was when you set up the initial trade did you consider the recent price action/range at all as a factor in setting your barrier, or were you just evaluating available premium, or were you engaged in some more complex/different evaluation process? I was alluding to this when we discussed the 1175 no touch trades awhile back.
Or to put it yet another way, does the premia available show a marked shift as you move the no-touch from just inside the previous week's range to just outside -- being simplistic for a moment about range evaluation.
Not directly; the model-output best represents the current statistical vol + the implied distro smile of the underlying market -- as gammas/vegas with respect to sensitivity. IOW, the model and the dealer are pricing the expansion in the range[>stat vol] not the artifact of stat vol, or where we were in terms of price. The distro is still assumed to be quasi-symmetrical. The curvature increases dramatically[inverse to vol//cheap debit] when the position is large, long delta. The vol smile reduces the debit based upon proximity/contamination via the put barrier. It becomes a bit more complex depending on delta.
This is kind of obvious I am sure but...
If I followed the implications of what you were saying with respect to the expansion pricing + quasi symmetry, then wouldn't a no-touch trader be partial to setting positions immediately outside the recent price range as opposed to just inside it -- assuming one believes in ranges -- to improve their chances?
If the cost of moving up or down outside the range vs just being inside it is "adjusted proportionally by a model" without regarding to identifiable pools of liquidity then it would seem to provide a small edge...
It seems like pricing models don't recognize s/r in any meaningful way unless "quasi" in some way hides the "bumpy" nature of pricing at perceived s/r points.
I'm sure this is naive but I thought I would throw it out there in the hopes of learning something new.
Quote from ssternlight:
This is kind of obvious I am sure but...
If I followed the implications of what you were saying with respect to the expansion pricing + quasi symmetry, then wouldn't a no-touch trader be partial to setting positions immediately outside the recent price range as opposed to just inside it -- assuming one believes in ranges -- to improve their chances?
I was referring to the effects of pricing equidistant "put" and "call" no touches. Although there is little sensitivity to vol, the vol-smile effectively prices the put barrier closer to spot. As you price further otm the debits become very large[inverse condition]. You'll improve your chances, but pay for the privilege.
If the cost of moving up or down outside the range vs just being inside it is "adjusted proportionally by a model" without regarding to identifiable pools of liquidity then it would seem to provide a small edge...
Sure, all things being equal... if you believe that the preceeding range accurately reflected S/R values, which I do. Trader's edge/model-edge
It seems like pricing models don't recognize s/r in any meaningful way unless "quasi" in some way hides the "bumpy" nature of pricing at perceived s/r points.
Correct, but the dealer may very well price the position based upon directional bias. I have seen it many times.
This is kind of obvious I am sure but...
If I followed the implications of what you were saying with respect to the expansion pricing + quasi symmetry, then wouldn't a no-touch trader be partial to setting positions immediately outside the recent price range as opposed to just inside it -- assuming one believes in ranges -- to improve their chances?
I was referring to the effects of pricing equidistant "put" and "call" no touches. Although there is little sensitivity to vol, the vol-smile effectively prices the put barrier closer to spot. As you price further otm the debits become very large[inverse condition]. You'll improve your chances, but pay for the privilege.
If the cost of moving up or down outside the range vs just being inside it is "adjusted proportionally by a model" without regarding to identifiable pools of liquidity then it would seem to provide a small edge...
Sure, all things being equal... if you believe that the preceeding range accurately reflected S/R values, which I do. Trader's edge/model-edge
It seems like pricing models don't recognize s/r in any meaningful way unless "quasi" in some way hides the "bumpy" nature of pricing at perceived s/r points.
Correct, but the dealer may very well price the position based upon directional bias. I have seen it many times.
Riskarb.. thanks for this thread.. this is by far the best one in ET's history... i have never read a more informative thread..
I will be printing out each page.. tons to learn from your expert exotic options trading...
thanks man..
--MIKE
I will be printing out each page.. tons to learn from your expert exotic options trading...
thanks man..
--MIKE
- Status