You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Premium Sellers vs. Option Buyers
- Thread starter Squilly_D
- Start date
It wouldn't go to $5. That is his fundamental argument or point.
I think you are missing the point he is making.
Maverick74, could you please elaborate more? Are those numbers made up, just an exaggeration, or what was Taleb talking about? Was he talking about 67000 times in terms of theta instead of premiums?
I also think that the Federal Reserve meddling will eventually end really bad (like it has many times, since 1929), so something very dramatic could happen in the markets. Just like Spitznagel says too. But I have been calculating the expected value of SP500 far OTM options, and I don't see how could they be profitable unless both the price crashes and implied volatility jumps to very high levels (so you'd have to sell them before expiration to profit from that implied volatility change, not just from the change in intrinsic value).
Last edited:
how much did 1 week 10percent otm options rally in late August?
When events like that or 1987 or 2008 happen convexity pays huge.
But it doesn't happen very often. The philosophical debate is that if you can't predict when the extreme convexity is going to happen, does it make sense to be consistently long convexity losing money 99percent of the time or is it worth being short it and suffering an extreme drawdown?
What taleb doesn't tell you is that universa isn't always long convexity. They spend enormous amount of energy trying to predict convexity and even then they lose money 4 out of 5 years and hopefull the 5th year pays for 10 losing years.
When events like that or 1987 or 2008 happen convexity pays huge.
But it doesn't happen very often. The philosophical debate is that if you can't predict when the extreme convexity is going to happen, does it make sense to be consistently long convexity losing money 99percent of the time or is it worth being short it and suffering an extreme drawdown?
What taleb doesn't tell you is that universa isn't always long convexity. They spend enormous amount of energy trying to predict convexity and even then they lose money 4 out of 5 years and hopefull the 5th year pays for 10 losing years.
how much did 1 week 10percent otm options rally in late August?
When events like that or 1987 or 2008 happen convexity pays huge.
But it doesn't happen very often. The philosophical debate is that if you can't predict when the extreme convexity is going to happen, does it make sense to be consistently long convexity losing money 99percent of the time or is it worth being short it and suffering an extreme drawdown?
What taleb doesn't tell you is that universa isn't always long convexity. They spend enormous amount of energy trying to predict convexity and even then they lose money 4 out of 5 years and hopefull the 5th year pays for 10 losing years.
I don't know how much they rallied, but I guess that much less than multiplying for anything remotely near to 67000. I agree with Taleb in many aspects and I love his books, but I still don't know which OTM options he's talking about. I can't see how an OTM ES option could ever jump by that much, even in a Black Monday scenario.
Well, from what I understand from that interview, it seems like they make money regularly by selling ATM options and then rarely by buying far OTM options:
"If you care about performance, you should short at-themoney options, which expire and have very unstable deltas."
"We invest in traders who sell at-the-money options, and we concentrate on just buying the “wings” (the out-of-themoney puts and calls of the butterfly position)."
67000percent is 670x. So your spy example would be 1 cent going to 6.7. That's very possible.
I know they trade some strategies that sell ATM options and buy otm options but by in large the way universa works is you tell them you want to hedge/invest 1bn dollars. They ask you for a 40mm check. They will likely lose that 40mm but in a convexity scenario that 40 can become 250mm: earning you 25percent on the "1bn" you invested with them. They are net buyers of premium.
I know they trade some strategies that sell ATM options and buy otm options but by in large the way universa works is you tell them you want to hedge/invest 1bn dollars. They ask you for a 40mm check. They will likely lose that 40mm but in a convexity scenario that 40 can become 250mm: earning you 25percent on the "1bn" you invested with them. They are net buyers of premium.
Last edited:
67000percent is 670x. So your spy example would be 1 cent going to 6.7. That's very possible.
I know they trade some strategies that sell ATM options and buy otm options but by in large the way universa works is you tell them you want to hedge/invest 1bn dollars. They ask you for a 40mm check. They will likely lose that 40mm but in a convexity scenario that 40 can become 250mm: earning you 25percent on the "1bn" you invested with them. They are net buyers of premium.
No, no, no. Not 67000 percent. Ignore that, that's what the introduction says. In the interview Taleb talks about 67000 TIMES and 750000 TIMES.
"If you owned an option that was 20 standard deviations out of the money — and I had plenty of those — how many cumulative months of time decay could you sustain if it moved into the money?
AT: I don’t know. A few dozen?
NNT: I quizzed traders, and they were telling me two or three years. But it was 67,000 months of time decay. You get paid 67,000 times your bet."
"But if you have a 24-sigma event on an option that’s 24 standard deviations out of the money, your payoff is 750,000 times your bet, which is what happened in eurodollars."
I think some of their math is a little fuzzy. Another guy and I looked at a strategy that spitznagel mentioned on some site of buying 1 delta puts against a long only and we couldn't reproduce the results he claimed.
I think some of their math is a little fuzzy. Another guy and I looked at a strategy that spitznagel mentioned on some site of buying 1 delta puts against a long only and we couldn't reproduce the results he claimed.
Even if he's talking about 67000 times what you pay in theta, it still doesn't make sense. To get that much in terms of theta buying something like a $1770 ES put that expires on December 18 (premium $0.4, theta -0.009), you need the price to drop to $1770 and implied volatility to jump to 700%... In terms of premium it's just impossible, as it would mean going from $0.4 to $26800, which is much more than the maximum possible win (the strike, $1770).
67000% is extremely unlikely, but possible. 67000 times is just impossible. So 750000, even more absurd.
You guys are missing his main point. He is not talking about being long vol in things that have a history of large moves where that historical precedent is embedded to some degree in option prices. He is speaking about things that "don't" have that precedent.
Example, mortgage derivatives. The reason CDSs were so cheap was because there was no historical precedent that showed large waves of default. So the swaps (which were synthetic options) priced in a zero percent chance of that happening. When it did happen, you get those parabolic payoffs. Same was true in the Euro Dollar options.
See, once these events do happen, they start getting priced in going forward. Taleb is not talking about buying cheap vol in AAPL or SPY. He is talking about the fact that markets cannot and have NEVER been able to accurately price shocks that have never happened. Even shocks that have happened we have so little data to use that any pricing model is simply a guess and will almost always underprice the tails. If I asked you how much you would pay to insure a commercial building against a commercial airliner flying into it from a terrorist attack before 9/11 my guess is almost nothing. That price changed after 9/11.
Example, mortgage derivatives. The reason CDSs were so cheap was because there was no historical precedent that showed large waves of default. So the swaps (which were synthetic options) priced in a zero percent chance of that happening. When it did happen, you get those parabolic payoffs. Same was true in the Euro Dollar options.
See, once these events do happen, they start getting priced in going forward. Taleb is not talking about buying cheap vol in AAPL or SPY. He is talking about the fact that markets cannot and have NEVER been able to accurately price shocks that have never happened. Even shocks that have happened we have so little data to use that any pricing model is simply a guess and will almost always underprice the tails. If I asked you how much you would pay to insure a commercial building against a commercial airliner flying into it from a terrorist attack before 9/11 my guess is almost nothing. That price changed after 9/11.