I was wondering if anyone out there is interesting in discussing some of John Bender's comments on option pricing from Schwager's 'Stock Market Wizards' interview. I found Bender's thoughts really refreshing in a world where people often repeat the mantra of buy low and sell high volatility without a real understanding of volatility's many causes, facets and quirks. I found myself nodding in approval to his statement that using the black scholes to price options in certain cases is quite limiting and narrow, because stocks don't always follow a random walk; just think of the effect of a high volume breakdown for ex, on the skew in probabilities for the calls and puts being in the money with x days to expiration. Anyone with similar ideas, contributions, criticisms to make? Would greatly enjoy your reactions!
------------------------------- Options just dont get much attention around here. Especially pricing models. Did you see this thread. http://www.elitetrader.com/vb/showthread.php?s=&threadid=5884&perpage=6&pagenumber=1 Its about randomness, position size, the influence of luck. Bender talked about probability distribution, and thats the first time it really clicked with me, as far as option pricing is concerned. I liked the way he explained how everyone tweaked the Black-Scholes to match actual prices. Maybe you could provide a specific trade to look at.
Redzuk: Thanks for your tip and link. I read the thread and found it quite interesting. Much could be said about Niederhoffer's and Taleb's diametrically opposite styles but overall, I agree with the guy who said that trying to pit short vol vs long vol trading is an exercise in futility- like the old and tired nurture vs nature argument. Also, I'd be really surprised if all Taleb did was buy up those out of the money options in the hopes of a black swan event, and I think he only lets us see what he wants us to see via his interviews and articles...As for someone who blows up twice in a row like VN, that just goes to show how hard it is for some people not to confuse unlikely with impossible, lol. Anyways, back to Bender. I think his comments could be extremely valuable for those who have ears to hear. First, on the fundamental assumption of the Black Scholes: that prices are random and hence can be modeled with the normal distribution. Talk about a weakest link! especially when you study price and volume action with a clear and open mind, not like our ivory tower friends in academia. Bender states the 3 assumptions about the random walk and then shows how weak those can be by way of the gold and the tech stock/fund politics examples. I guess the gist of his point is that, due to some common sense and very real issues in the market, such as stop gunning or herd behavior, some prices are much more likely to get hit than others, hence a skew in the probability distribution. Ex: BRL. Go back to last Friday. First breakdown took the stock down 2.46 points to 66.54. I checked the implied vols at 3.45p.m. The June 65 call was trading at 3 bid, 3.40 offered, vol around 40% implying that BRL had roughly 68% (1 standard deviation) chance of being between 71.8 and 61.3 at June expiry. Now here's the catch. Was it really likely that within the 1 s.d, BRL had a 50/50 chance of going up to 71! or down? absolutely not for technical reasons that the chart makes obvious. The skew in probabilities was clearly to the downside. For fun I tracked the June 65-80 bear call spread. Nearing close on Friday 31st, you could have sold the 65 for 3 points and bought the 80 for .35 as a hedge. (Er, I wonder what VN thinks of hedging naked sales...) Over the next 2 days, BRL traded as low as 63.05, with the June 65 calls trading as low as 1.40 ask. A gain of 1.25 net net per spread. one could have also just bought the 65 puts but with 2 weeks to go but time decay would eat away at those things too. One can always say, that in hindsight it always looks good. But to me the key is working with expectations, sure some mispricings could be hard to exploit, especially if the stock doesn't go your way immediately like it did with BRL, but if you have decent exit rules and good money management I think it's more than possible to trade these overvalued or undervalued options consistently and win. The edge and the risk control are the bottom line. If you have some ideas to share on the above or anything else, would love to hear, Maverick1
Maverick, Are you just saying that options pricing typically does not reflect chart patterns and that therefore if one can make decent directional bets he has an edge? Obviously, the pricing models assume one cannot predict prices. I wonder if there is any data showing how or if implied vol's are affected by obvious chart patterns?
AAA, You raise a good question that I wrestle with constantly in my analysis. Market makers are in general, on top of things when it comes to accomodating and anticipating big moves in their stocks-you can see how they jack up the vols weeks prior to events like a merger announcement etc. If rumors are floating around you can be pretty sure that they'll be the first to hear and act on them. Case in point, recently, the GSB takeover by Citi. Vols were pumped up weeks before the event as the mmakers made sure that anyone who tried the old event driven straddle trick paid a hefty price for it. So in most cases, they do their best to anticipate. (by the way, that deliberate pumping up of the vols resulted in an overpricing of the June 30 puts right before the gap up, another opportunity) When it comes to chart patterns, I personally believe that there are few very of them (less than 5) that offer you a real edge in the underlying asset over a decent sample of trades. Of those I think only one or two can offer you a consistent edge in the options if we're talking pure directional plays The mmakers are obviously aware of the many patterns that get traders all excited but they also know that many of those patterns fail too. From what I've observed so far, there's only one or 2 patterns that make mmakers nervous to the point of adjusting their implied vol levels. I track and stalk those setups with the help of ivolatility.com to see how the vols react. Very instructive. You've probably heard of ivolatility.com. Great website. In short, I'm saying that the option premiums typically reflect 98% of patterns because those usually carry a high failure rate and hence the BScholes 50/50 bet is the rational one. But...I do think that there are situations where inefficiencies repeat themselves. Note that, that doesn't necessarily mean that the market maker always loses on those, I mean those guys are hedged within seconds in the underlying and can manage their risk faster than anyone else. I forgot a major caveat in my earlier post: It seems to me that stock prices actually do follow a random walk...but not ALWAYS as some people would like us to think. It's the 80/20 rule right?, most stocks make 80% of their moves in 20% of the time. I think the mispricings tend to occur in that 20% window. (i'd be interested to hear what a turtle trader has to say about that...) I'm on a school break, so I'm probably writing too much here, so I'll turn back the discussion to you- what's your experience with option trading so far? maverick
First, a couple of problems with the trading aspects of your example: 1. Spreads between the bid and ask can be large. In the example you cited, the difference was 12%. If you decided to cover after a small stock decline, your profit may not be as big as expected because of this. 2. With three weeks to expiration, the 80 calls are not a hedge for the 65's. If the stock were to rally up to 70 or 75, you would be hurting. Why? The calls would be rapidly decaying as expiration approaches, especially, with a 40 vol. Volotility typically declines as a stock rises. You would need strong expectations for the stock to rise to 90 to justify such a position. Of course, your expectations are for a decline. Buying the 80's would be throwing money down the drain. It would be better to buy the july 70's as a hedge. I think you can look at the skew from a couple of different angles. I've got to go so I'll finish later.
Maverick, I guess the most basic question is, if you are pursuing a directional bet, why use options at all? Why not just trade the underlying? Let's say I am playing a reversal in an equity. I can go long stock, I can buy a call spread or sell a put spread or just buy a call. If leverage isn't a factor, aren't I better off just buyign the stock, or, in a few months , perhaps the SSF? That way I have only one spread to deal with, plus I can probably execute faster.
Nice input. I'll try to answer your points: 1) Wide bid-ask. Sure, that's a problem that cannot be overlooked. However, I look at it from a risk-reward standpoint, if the setup is excellent and i'm looking to make 1.00 or more off the short leg of the spread then it might be worth taking the trade if when you come back to cover you're faced with a .20 or .30 spread. But I can see how that would deter other traders. More liquid stocks could help avoid this but my question is- do liquid and widely followed stocks offer you as high a probability trade? In the BRL ex, your gain was net net 1.25, if spreads had been tighter, it could have been 1.45 or 1.50 at best. so the ratio is here almost 6:1. But this justification only holds if you have above average positive expectations on your win loss ratio in the long run. 2) Use of the 80 call as 'hedge'. I should have really put the hedge within commas. The idea with this setup is to see the stock go in your favor immediately. If it fails you get out immediately if the stock reverses and hits your stop. You absolutely need a price stop at which you liquidate your short option at a predetermined loss. You can use the delta to estimate that point. The idea behind the 80 call is not to use it as a typical hedge but rather as insurance against a gap up in the stock. You are seeking to maximize gains from the theta decay in the short option and of course from directional movement. Using a 70 call on the brl ex, even the july one as long hedge would have drastically reduced your profit potential. The 70 call at the time of the setting up of the trade had a delta of around .35. The 65 call had a delta of .65. Net delta on the BCS= -.30 so you only make 30 cents on every dollar Brl loses. Reduced risk, reduced reward. Law of nature. You mention that theta decay would increase rapidly with 3 weeks to go. You're absolutely right and that's the whole point of choosing the 65 atm call to sell. With the 80 'hedge' that you bought for... 35cents, what can it decay to? 0?. Not a concern if you expect to make more than a dollar on your short option. The Brl ex was a 1.60 gain on the short 65 call and defacto .35 loss on the 80 'hedge' What's the better trade, of the 65-80 (15 points) or 65-70 (5 points) bear spread really depends on two things for me. First, how much return you are willing to give up for less risk and secondly, how strong of a technical edge you have if you trade this strategy repeatedly. If you have a sample of 30 trades like that (wide hedge) and you win 60% of the time and cut your losses immediately on the other 40% then maybe it might be worth it. note- there are many ways to trade that type of situation, I'm certainly not advocating the above (wide bcs) as the only one or the best. Then you could outright sell the naked call but I think there's enough Niederhoffers out there we should know better for his losses. looking forward to your further comments.
AAA: Yup, interesting point. And as you say, why not trade cash for directional bets. I would say yes and no. I would trade the underlying for long stock trades. In fact, I trade breakouts that way. I happen to believe that buying calls or call spreads are inferior strategies when compared with simple buying in the underlying. Because options are a 3 dimensional game, not only do you have to be right on price, but also on vol and time. Those suckers melt in your hands at the slightest hint of hesitation on the part of the stock. Much better to trade cash unless you've got a sure thing trade, which as we all know, don't exist unless you're george angell. Bullish put spreads are a different animal. If you've got a real handle on statistical vol and premiums are juicy enough, could be good trades. Not so long ago I saw an AMZN 10-12.5 June put spread trading for .30 cents credit. If you're the kind of trader who can't sleep at night knowing that you're in danger if AMZN moving 2 standard deviations then maybe not a good idea. a 25:3 risk return might not appeal to you. but i believe many traders make those bets simply on the extreme unlikeliness of stocks moving 2 sds. When it comes to the short side, options may be worth it for multiple reasons. 1) the uptick rule. 2) vol expands dramatically with breakdowns. There is definitely an asymmetry in vol behavior between breakouts and breakdowns in stocks. Commodities are different, (gold, yen, oil etc) 3) when you sell options you earn time decay 4) when you play the short side you're on the other side of the public, the 90% folks who are long and are about to take it bad. SSFs will hopefully add some vol to the picture in the future though and you won't have the uptick rule to deal with though. hope that helps maverick