I never thought I'd put a wine collecting metaphor together
with the topic of writing options. But I've
been provoked. The occasion was provided
to me by my reading, for some unknown reason, all of the several hundred
comments that followed two or three different web-posts on the practice of selling
options. First of all let me say -- and
this is mostly unrelated to the main point I'll try to make -- that my capacity
for surprise at the uncouth commentariat that derives its strength from
internet anonymity and at what passes for their attempt at "dialogue"
is born again each time I read this kind of stuff. Sifting through all this dreck, though, I found at least one consistent, grating, repetitive theme emerging, voiced more often
than not with either passive or active hostility: the simplistic and black-and-white
opinion that option selling has infinite risk and limited gain and therefore is
a waste of time; a single trade can waste a whole year's worth of gains --
gains that are too small to be worthwhile in the first place -- so don't bother. Hmmm. My mind wanders to my wine metaphor.
It goes like this. Imagine
there is a person who: has spent 30 years tasting and collecting wine, has read
through vast swaths of what has been written on the subject if not all of it, has traveled to or
lived in a number of the world's best growing regions, has tasted through
thousands or tens of thousands of different wines and varieties from places
both grand and modest, has met and corresponded with no small number of working
winemakers, has contemplated and tested and validated at least small parts of
the large zone where food and wine come together. Now put that person, a wine pilgrim, (silent, so that we do
not take him or her as a wine-snob) at a table where the main dish is a something
like grilled fish made with a savory, maybe smoky, and highly textured sauce. It
is served, appropriately in my opinion, with a Pinot Noir from Oregon 's
Willamette Valley
I get a similar feeling sometimes when I see the infinite-risk-lmited-gain
argument laid out over and over and over as it relates to option selling. These are the guys (and they're usually guys,
right?) that have probably read page one, chapter one of "Introduction to
Options" and are now providing their opinion in a notch-too-loud
"voice" to everyone else at the table.
But let's take a closer look at all of this. This option selling thing is well worn and
known territory and it certainly doesn't take 30 years to master. Just a little more study or experience could tease
out an entirely different point of view once we can get past the first page of chapter
one. First, though, just for fun, let's
set up a fake little straw man for ourselves and say that there are two opposed
schools of thought on writing option premium: 1) it's a waste of time, one trade will blow out
a whole year, the risk is infinite and returns are boring, and this is not
"real" trading anyway, and 2) short option premium selling is a magic,
alchemic, utterly amazing insurance-selling-like cash machine for the ages…and
anyway Warren Buffet does it[2].
I realize this is a fake and exaggerated setup but both happen,
if they were real, to be wrong. And both
are wrong because option selling[3] like everything else in finance can't
escape the gravity of financial theory, trading math, and common sense. I am thinking of two things in particular here. First, consistently selling options in a
systematic way to get, say, exposure to the volatility risk premium (vs. looking
at only a single directional or hedging trade) is a capital allocation decision
whether you like it or not. A single unhedged or unmanaged trade might have
vast risk but many trades (well managed, let's hope) over time are a system and
as such the system acts like an asset to which one could allocate capital. And here, I'll hypothesize briefly that it's
highly doubtful that one could get a 50% or 70% or 100% CAGR on total committed
capital (we're not talking about a return on margin here) with super low volatility
consistently over many years from this "asset." If you could do that you'd
be running a hedge fund making billions. But it's not magic after all. The system will
behave more or less like any other asset class or strategy if it works. On the
other hand the mean-variance characteristics of the strategy might, in fact, really
be accretive to one's portfolio. The
decision to engage in option selling (as a system) then should be based less on
what the "commentary pundits" might say about infinite risk of one
trade and more on normal stuff like: is it (the system) well managed? What is
its expected return and is it positive and more than my required return
threshold? What is its expected variance? How does it co-vary with my other
assets? Does it, in the end, make my portfolio any more efficient? If you can answer those questions to your
benefit then do it, other wise don't. This is where the commenters make their
first error.
Second, option selling, as a system or strategy, can't
escape "the formula." By formula
I mean the trading success formula that an ex-trading partner/mentor -- and a
guy that ran a billion dollars for a while -- gave me my first day under his
tutelage: "Trading Success" he said "is a function of: 1)
risk/return, and 2) the probability of that return." To make it look kinda mathy let's say it's S
= f(R,P) where S is "trading success," R is the return/risk ratio and
P is the probability of that return (or maybe the ex-post win rate). Alexander Elder in his well known book calls
it "expectancy" and calculates it, if I remember correctly, as (average
win size * win rate) minus (average loss size x (1-win rate)) which happens to
conform to the "success formula."
A value of S greater than zero always needs to be the goal for a new
trader trading anything. Later it becomes clear that S needs to be more than
zero. It more properly needs to follow the same capital allocation thinking as
above: does the strategy's realized success meet my various capital allocation
criteria. But for now lets just look at the formula for what it is because this
is where the commenters make their second error.
In traditional trading a common rule of thumb is to risk $1
to make $3. But that is unanchored. You also need to do that in a [preferably
continuous] series of trades where over time you get it right at least 1/4 of
the time in order to break even within a strategy or system. So using the trading
success or expectancy formula for a 3:1 risk return and a 25% probability a
winning trade we should expect to at least break even: S = ((3 x .25) - (1 x (1
- .25))) = 0. Let's call this first formula a Type 1 "system":
probability-thin with a "regular" risk-return. Now, let's switch it around and risk $3 to
make $1 but let's also try to get it right at least 75% of the time. S = ((1 x .75) - (3 x (1 - .75))) = 0 =
breakeven. Let's call that system Type
2: probability-rich with an inverted risk-return. Selling options is Type 2. For this type, you
just have to make damn sure that the risk doesn't go over 3 (in this example)
and that you get it right at least 3 out of 4 times but this is no different
that the risk control necessary to manage a Type1 system. So, therefore, I contend that if S is the output some
system that happens to have positive expectancy and it also consistently meets
your capital allocation criteria, it should be "in," otherwise it
should be out. It doesn't really matter if it is Type 1 or 2. This is painfully
simple stuff but the point is that selling options, despite the histrionics of
the web comments I read, is neither alchemy nor infinitely risky (under well
managed circumstances, anyway) because even though it can sometimes escape good
management it cannot escape the math. Use
the formula, Luke. For fun, here is a well
known type 1 guy on "the formula:"
Five to one means I’m risking one dollar to make five. What five to one does is allow you to have a hit ratio of 20%. I can actually be a complete imbecile. I can be wrong 80% of the time, and I’m still not going to lose. -- Paul Tudor Jones
There are plenty of other reasons, by the way, to contend that
option selling is a viable and supportable enterprise. Others have written well
and at length and better than me on this stuff so I won't regurgitate it all
here. The two things that I usually
point out to myself, especially since I have a history of having a hard time
finding the high probability trades that we can leave to Mr. Tudor Jones,
include:
- Capital isn't tied up so much in a directional wager. I no longer have to plunk down money and hope it goes in a particular direction. I just have to have some cash/margin capacity and a probability-based case for where the underlying price won't go in a particular time frame. This is much easier to do than predicting where it will go.
- Time and probability are in my favor. It is in the nature of options "decay" over time; they expire. That's bad if you own a long option that is not "in the money" and which will expire worthless but it is good news if you sold the option to the buyer and have already collected the premium. The combination of time decay and selecting where it won't go stacks probability in my direction. It's a little like having death and taxes on my side rather than against me.
A third reason, if I were inclined to add one, to support
the practice of option selling would be that the volatility risk premium, which
is what one is exposing oneself to by selling options, is a real live legitimate
risk premium -- just like the equity and credit risk premia we usually think of
as legit -- that has been validated both academically and empirically. That
means it is not so much of a stray risk artifact or the outcome of a lucky
backtest. It also means there is no
shame in finding ways to add it to the design of a portfolio. Here is a guy from AQR commenting on
volatility in "Embracing Downside Risk" in 2015:
Many investors focus a great deal on trying to maximize their upside participation while minimizing their downside risk. Options arguably provide the most direct downside hedge, but often at a significant cost, reflecting investor preferences. This cost, commonly referred to as the volatility risk premium and measured by the difference between the option’s implied volatility and its underlying asset’s realized volatility, is compensation paid by option buyers to sellers for bearing undesirable downside risk. The size of the volatility risk premium is related to investors’ asymmetric risk preferences. …Data and economic theory suggest that investors who attempt to deal with downside risk by being long options should expect to underperform. On the other hand, those who seek out downside risk exposure by selling options should expect to outperform for the same reason.
Which brings us, naturally I think, to a discussion of risk.
The kind of risk these guys on internet
pages are talking about, though, when they talk about the infinite risk of
short, naked options is what I consider to be normal risk; known risk; probabilistic
risk; the kind of risk that factors into the management of either individual
trades or the risk of a system using "the formula." This kind of risk
is easily managed and that allows selling options to work well as a type 2
system that I described above. In terms
of managing risk, any options book (or web site) can delineate risk management
techniques better than I can but here is the gist of it: have a pre-trade exit
plan, use stops or conditional stops, pay attention to the chart, use premium
based risk metrics, use spread/ratio/combo trades, stay as far out of the money
as makes sense, trade with a trend or seasonal tendencies, keep the time/tenure
short, sell after volatility spikes, diversify, manage position sizing risk,
etc. etc. The commenters I mentioned
either know this kind of thing and are being deliberately combative, which
wouldn't surprise me, or they don't know this in which case they are talking
out of their wrong end you know what. Either
way, all "normal" trading risk is part of the trading game and can usually
be managed pretty well by the wise.
But I also think that they, the commentatiat, are missing or
at least undersell what I'll call real risk.
That means I think their comments fall somewhere between lightweight ephemera
and red herrings. Real risk isn't normal
trade risk, well managed or otherwise. It is either something that is outside
the system and can't (usually) be predicted or it is inside the system and is a
risk that wasn't easily seeable and when it hurts it hurts most when you
least want it to. This kind of risk
actually does scare me where the normal, manageable "infinite"
downside of short options doesn't. That
latter is trading risk and can be managed. Real risk looks and feels different. I couldn't find all the expositions on risk
that I had wanted to use in this post but here are some that I still had at
hand that capture some of the point I was going to make:
Cliff Asness, AQR:
"Now,
keep in mind, to be risky, that investment has to lose sometimes, particularly
when it really hurts to lose! This is something often lost as some investors
assume that risk is simply that something occasionally goes down…The bad news
is, of course, that risk is risky! Remember this shouldn’t be some mere label
called “risk” but actual pain coming at the times it’s hardest to bear (this
might be short sharp pain, or long periods of less-severe agony, but it has to
be pain when pain really hurts). In our opinion, those espousing risk-based
stories, sometimes, at the margin, don’t seem to want to uncover this real pain
but merely use the word “risk” as a proxy for “rational” so everyone can feel
good about investing. “Oh, you get paid for doing this on average, but
sometimes it ends your career or ruins your retirement” may or may not be an
exaggeration but it’s certainly not a great tag-line for an investment product!
But it is indeed a reason for the expected return premium associated with
bearing this risk to be real."
Peter Mladina:
Risk is multi-dimensional. Risk is
volatility. It is tail risk. It is the permanent loss of capital. Risk is a
failure to meet financial goals.
Dirk Cotton. The domain here is retirement risk, which may
not seem relevant, but I like the message on normal vs. real risk and the idea
of real risk coming from outside a planned system:
Losing your savings due to market volatility [i.e., sequence risk or "normal" probabilistic risk] is only one risk of retirement and it isn’t the worst outcome... About half of us won’t live long enough to be exposed to sequence of returns risk [which] will rarely be a big contributor to bankruptcy [i.e., real
risk] and it takes decades [for sequence risk] to erode savings. Bankruptcy
will strike like a bolt out of the blue as a result of spending shocks [and chaotic
positive feedback loops] and may cost much more than just your savings. (All
four of the families I wrote about in Positive Feedback Loops lost their homes,
too.) The crisis will be difficult to stop once it starts…. When a planner
tells you that you have a 5% probability of depleting your savings, she
typically means a 5% probability of going broke as a result of market
volatility [i.e., normal risk]. Alas, there are other ways to go broke. If
spending systems are chaotic, which I suspect but can't prove mathematically,
there are conditions under which their outcomes are unpredictable and
probabilities don't help.
A holder of a deposit account in a Cyprus
bank before their 2011-13 banking crisis might have originally worried about
things like fees or interest rates or maybe service hours but they were probably
not thinking about having a vast portion of their capital seized by the state.
That is real risk.
My ex business
partner:
My former partner raised nearly a
billion dollars for a currency hedge fund and ran it for something like 10
years or more. At any given time he might have had, by my best guess, maybe
$100 million or more of other people's money in open currency option positions, much of which had a strong tilt to short options (e.g., ratio spreads). Somewhere around 2008 or 2009, when
volatility was running itself up a 90% cliff, he started having explosive
losses which, since currency funds by the nature of the instrument are pretty
liquid, meant that clients abandoned ship like plague-infected rats. Losing money, losing clients, liquidating
complex positions under duress, he had the zen-presence of mind to also lean
into the volatility to recover his NAV back to about where it was before the
vol spike. The problem, though, in the end wasn't the intermediate losses and
recovery, it was that no one came back after the recovery. Me? I think that to watch a non-linear loss
explosion of that scale with money that is not yours, to have the rectitude to stay with the rules and
discipline that made the strategy work over time in the first place, to not lose money (time-weighted anyway and
after the smoke cleared) when all hell is breaking loose is either courage of a
very high order or a monkish kind of detachment that seems almost superhuman. This
kind of thing is real risk. I'm not sure I could have done it. I asked him once why he didn't hedge out what
seemed like obvious tail risk. He
replied as if to a child: "You don't understand. No one in my lifetime had
ever seen or believed in a 90 vol. No one knew it was coming. You can buy
homeowners insurance but it's hard to plan for a meteor hitting your house and
family."[4]
and me:
The risk of losing a bunch of money
on one poorly planned and managed short options trade doesn't scare me. It
might irritate me but doesn't scare. On
the other hand, having a "normal" portfolio, doing all the proper
asset allocation things, taking all the "normal" steps to mitigate
risk, not engaging in limited-upside-infinite-risk options selling…. that can
be scary, too. Really scary. In 2008 and
early 2009, I vividly recall losing more on a whole bunch of individual and separate
days -- each seeming to come one after the other in what felt like an unending
stream -- than I had made in some entire years from employment income before the crash. The best we can say about the final flush in early 2009 was that it was nauseating. And that was only 1/3 of the risk I felt at
the time; the other two don't belong here.
All of that seemed like risk to me. Real risk. Options aren't that scary
compared to that.
So when it comes to talking about risk and when I see what
looks like the uninformed and inexperienced throwing out entry-level platitudes
about a kind of risk they have not much bothered to explore or quantify and
about which they are likely to have limited experience and who have also likely
not seen or understood or lived through what I am calling "real risk,"
I get provoked (hey, it's a pretty small provocation but one does have to blog
something after all).
Since I am ripping on people that do not explore risk, let's
now take a look at a small managed short options strategy (system) to see if or
how the commenters' fears about infinite loss and crappy returns express
themselves in a live working system.
I've happened to run a short options strategy for about four years
now. I assume that that is long enough
to get a feel for how it might work in practice. On the other hand, the trading was relatively
infrequent or thin and the stakes were pretty small. So maybe it hasn't been tested by a downturn
or volatility explosions and maybe the small size and intensity meant I was a
little too insulated from fear and greed and behavioral factors but I have what
I have. I also know I am not the only
person on earth that does this and I know there are more sophisticated traders
that have fancier trade styles and more scale but I also don't have their data
in hand.
The basic strategy was to have a continuous net exposure to
short options to capture some of the volatility risk premium. By and large this
was done with naked -- and implicitly cash secured -- puts. The rules aren't
really the point here but know that there are rules and that rules are
important. They keep you out of deep trouble.
I believe any successful risk
management system has to be rules-based or it won’t work in practice because
trading on gut instincts is an impossible strategy that very few people can
pull off consistently. Ben Carlson
Also, note that there are some things to be a little teeny
tiny bit wary about here. For example, some months I did not trade at all so
volatility calculations are maybe suspect.
I shared the platform with other strategies so I had to back out some unrelated
trades from the analysis; I hope I didn't miss something or delete the wrong
things. There were some rule and strategy and intensity changes along the way
so there are maybe some consistency questions. I also didn't really "commit
capital" but piggybacked on an existing account and soaked up margin or
cash as needed rather than "investing" a specific amount. That means
the annualized rate of return is going to be a little subjective. But let's gloss over these kinds of things,
shall we?
What does it look like?
CAGR: 10.5%
St Dev 3.9%
Win Rate 82%
Average Win $1
(dollar normalized)
Average Loss $1.4
Max Win / Max Loss: 1.1
Expectancy $.55
What do I see here? First of all, I don't see infinite losses. I also don't see terrible returns. This looks
like a normal trading strategy and the lower-left-to-upper-right thing is reassuring.
Frankly speaking, it looks a helluva lot better than a bunch of other
strategies I have managed over the years. The max loss is not out of hand in relative terms
and again is not infinite. The probability is high. The risk return is inverted
but not obnoxiously so and was expected. The expectancy calc is solidly positive.
The mean variance stuff looks pretty good, too.
I would have calculated a Sharpe ratio but I no longer know what to use
for the risk free rate or what that even means anymore…but Sharpe would look
good if I did the math. In my opinion, there
is absolutely no reason not to consider this a viable strategy to add to the
design of a portfolio, my portfolio.
There is maybe one little tiny asterisk, though. In the literature, this kind of strategy has
a rap against it related to covariance.
Supposedly and logically, equity based short option (put) systems can correlate
a little too much in a slightly extreme way with an equity portfolio when one actually
might need and want something to not correlate at all or negatively correlate. But even that is manageable I think. Sell
calls, too. Or sell puts on treasuries at the same time as equity puts. Or sell
options on a diversified basket of commodity futures. Whatever the case or method, I still don’t
see the need to fear or contemn short option selling in such a loud and unreflective
voice.
--------------------------------------------------------------------
[1] Based on real people in real life…with less drama,
though.
[2] I find that the invocation of the Buffet name, like they
used to say of patriotism, can sometimes be a "last refuge" for
certain types people. But that is another story.
[3] In my head I am thinking primarily of selling options
naked though I realize there are many different ways of getting short exposure.
[4] Reconstructed dialogue, of course, but the sensibility
is correct.
No comments:
Post a Comment