So I've been thinking (that's dangerous all by itself) about CTAs (Commodity Trading Advisors). Based on what I know of CTAs, a lot of them use some type of trend following strategy (as opposed to a mean-reverting strategy). Some of them, at least conceptually, are pretty basic strategies. One example of such a trend following method is what the Turtles used...basically a breakout system. The strategy seems to have been pretty robust over the years (although maybe not such much lately with such low volatility in a lot of markets). In fact, over the long run, a lot of the trend followers have done pretty well. Sure, they have had down times. And yes, some of their drawdowns have been pretty deep (severe). But, a lot of them do pretty well with decent long term returns. That said, there are problems with most trend following strategies. One, the drawdowns. The drawdowns are usually a result of taking a position when you think there is a trend starting and you turn out to be wrong. In other words, the trend followers get chopped up a lot, taking a lot of small losses that, when added together, result in a large loss. The other major problem (that I can think of) is that the majority of a trend follower's profits may come from a very small percentage of their trades. What that percentage is I'm sure varies. But, let's say it is between 10 and 30%. Essentially, most trendfollowers need to catch (and are looking for) massive (2, 3, 4 standard deviation) moves to not only make up for their losses but to also have a decent return at the end of the year . Now, I know I have made some generalizations and some oversimplifications but I think what I have said so far generally holds. Okay, so here's what else I was thinking (ie., time to wake up. This where I think it could get interesting). What if a trendfollower were to use DOTM options instead of, say, futures? I think that DOTM options would do the following: 1) Minimize their losses during the chop. Since trendfollowers have to take every buy/sell signal (they never know which signal is going to lead to the "the big one") they are wrong a lot. When they're wrong, the losses would be minimized (in comparison to trading futures outright) since DOTMs don't cost much. 2) DOTMs, while minimizing the losses, still allow the trendfollower to benefit from extreme/outlier moves in the market (2+ standard deviation moves). And since the bulk (and maybe, in some cases, all) of their profits come from these outlier moves, they are maximizing their strength (so to speak). 3) DOTM options cap their losses in the event of a catastrophic move in the wrong direction (opposite of their position). If such an event occurred, the trendfollower might have a difficult time unloading their position without a significant loss. 4) Unlike ATM or slightly OTM options, this premium doesn't have any (at least minimal) time decay or vol effect (at least before it starts to move toward ITM). So if you're wrong about the market direction or the market just sits in a range, you're not losing money due to time or volatility (or you shouldn't be losing much). DOTM (at least initially) seem to be a pure directional (delta) play which is what a trendfollower would want. 5) Buying DOTM options are less capital intensive than the margin (capital) to trade the underlying market. If you've got anywhere from 100 million to 1B dollars (like some CTAs do) that is a LOT of DOTM premium. To sum it up, DOTM options seem like a great instrument for minimizing losses during choppy periods while still providing the opportunity to profit nicely from extremely large moves in the market (which is what I believe most trendfollowers want to do anyway). So, here's my questions. Why don't CTAs do this? Has any one looked at (possibly backtested) what the results would be if a trendfollower where to buy DOTM premium instead of the underlying? What would be the downside of such a strategy? What am I missing? -SG
Major issue is liquidity and slippage. If they are still in a trade.. they will also have to rollover to the next expiration month which adds to slippage... they will have also have to deal with IV and making sure they the options are priced somewhat in their favor.. because they can be right about direction and still end up losing money.
This is basically what Nassim Taleb's fund Empirica has been doing. Buying DOTM options. Of course, he and his team perform quite a bit of - apparently - quantitative research. Whenever he has spoken at a seminar - at question time he has made it very clear that he does not wish to discuss empirica's strategies (don't blame him - the cheek of those quants asking questions) As he puts it, he watches his capital slowly erode, until the so called black swans make a guest appearance... Having said that, I do not know how empirica has been performing lately. He is currently on sabbatical anyway. I gather he prefers epistemology and teaching (he runs a course at Columbia U if memory serves) to trading.
If I'm buying DOTMs and I'm skillful in my trading I don't think slippage would be that much of an issue. If these CTAs can get huge positions on in the futures while minimizing slippage, I think they can do the same in the options. I realize it isn't exactly the same (liquidity for futures vs liquidity for options), but I think this problem is solvable. Regarding rollover, if they are still in a trade and it's time to rollover then they can just buy another DOTM option like they did before. You keep doing this until the trend "runs out". Regarding liquidity...that (in my opinion) might be an issue. But people (institutions) do put on/take off large positions in the options markets so I think the issue of liquidity (while legitimate) certainly can be handled. You could, for example, scale into your positions (which I think a lot CTAs do with futures anyway).
I thought that Empirica was essentially buying DOTM strangles. I thought Nassim's whole premise was that traders underestimate the frequency of catastrophic events ("black swans"). To that point, I thought he bought DOTM strangels. That way he doesn't have to be good at picking the direction of the move because when the massive move occurs, he'll make money regardless of what direction it's in. Is this not correct? The above (what I think Empirica is doing) is not the same as what I'm talking about. I'm saying take the exact same systems, techniques, and strategies that these large CTAs have been using for years successfully. But, instead of buying the underlying buy DOTM options. So, if they want to get long the market, they would buy only DOTM calls, not DOTM calls and puts like Empirica.
Actually - and understandably so - Nassim is pretty tight lipped about the whole thing. Your guess is as good as mine! But I do recall reading that he buys DOTM puts and calls on the same underlying. About the "black swans" yes - correct. As Nassim would put it, these rare events appear to happen with greater regularity than standard theory would have it. No wonder his handle on the Wilmott forum is "Kurtosis" (and by this I gather he means leptokurtic as opposed to platykurtic).
Do you think that this strategy is something that would primarily apply to large cap, very liquid underlying equities (such as INTC, QQQQs, etc.)? One issue that I could potentially see with buying deep OTM options would be that the open interest might not be large enough to support large capital positions? For example, if one was looking to buy December 2005 Calls in, say, TASR that are a fair ways out of the money: $15 strike, ask is $0.20, open interest 7357 - $147,140 $17.50 strike, ask is $0.15, open interest 1130 - $16,950 ...and of course, I doubt it would be possible (and even wise) to buy up the entire open interest, or liquidate if doing so. I guess my point is that if you're running some sort of hedge fund with multiple millions or even billions of dollars, you would need to play with something extremely liquid if you were chasing after deep OTM options.
A couple comments about liquidity. Especially in back months, the DOTM contracts are thinly traded. If there is no volume, it doesn't matter how good of a trader you are. You will have to pay the MM the spread. The # of OI does not impact whether you can open a trade. The MMs will sell you as many contracts as you want to buy. The only liquidity problem you will have is if you want to sell when a contract is offered at .1 or .15. Then, you will likely have a hard time finding a buyer.
Concerning black swans...what do you traders think about sell dotm options where the Black Swans wil not occur. eg S&P calls, silver outs, oil puts etc. What can cause these black swans?
This doesn't make sense to me. My understanding of options is that someone initially sells (writes) a put/call, which causes open interest to increase by 1. The option is then offered on the market (or sold at the bid to a MM), someone else can then come along and purchase at the ask (or put their own bid up I suppose). So, that said, can you explain how it is that MMs would sell as many contracts as one wants when there is a buyer? Are the MMs then both liquidity providers and options writers in the market?