When the Market Breaks: Predicting Fear... while Cherry Picking Chart Dates...
A longer text with a number of charts layered in between...
The "fear index" of course is the $VIX... a market innovation that debuted in 1993... which limits our use of it in viewing the market history of fear as a derivative, prior to that. Before 1990... you have to look at raw market prices to measure underlying behavior and its drivers, or count banker / trader suicides, or use some other metric. Wandering... I'd think it might be worthwhile for economic historians to try to reverse engineer the metric to apply historically... as that might illuminate how other changes... like the telephone, and then the internet... have altered market behavior as a function of change in information flows. So, any PhD wannabes lurking about... might take note...
Otherwise, history continues happening... Prior to 2007 - 2008 (as you might have noticed) "financial innovations" occurred... including some outside of banking, insurance and mortgage markets. From about 2007... people started paying attention. First, in 2004, derivatives traders decided it would be useful to place bets on volatility as $VIX... and then began noting problems with the time function in the way the math worked, and saw a need to fix it. Rather than me summarizing all of that... cover it with this simple to understand link to: Enhanced $VIX Methodology: The New $VIX Calculation
That also fleshes out the cast of characters... as there is no longer just "the Vix" ($SPX / 30 day) as an index based on order flow in $SPX options... but also Vix derivatives actively traded as futures contracts covering various time periods... and more... but that's enough for my purpose today. So, the $VIX and the futures contracts based on it: $VXST (9-day / $SPX weekly options), $VXV (3 month / intermediate-term / $SPX monthly ), and $VXMT (six-to-nine months / $SPX monthly). Throw in the $VVIX... the VIX of the VIX or the volatility of $VIX, and here's a current two hour chart showing relative performance of them:
What you see charted there... is a recent view of "change expected" in SPX over different time periods... from the short term (9 day) view using shorter term options... to the VIX short term / 30 day view... to the longer term views using both longer term monthly options, and longer term views of change expected over time in those longer term options. And, the VVIX... how much change in change you expect... So, basically, its a chart of "normalcy bias"... the expectation that significant change will generally not occur... paired with larger deviations from the expectation proven on shorter time lines versus longer timelines... or, its a chart of market myopia given the reality of uncertainty. Market up, $VIX down... market down, $VIX up... only, with variations imposed by the degrees of deviation from "market expectation" already priced in.
Awareness of risk changes slowly.... until suddenly realized... but may still average out over time ? That might sound like I'm saying its irrational... and some clearly is... but, uncertainty is real, and likely largest in specific risk exactly when you think you have it all figured out. And, there's another suppressive layer in the longer term that is buried in the "invisible" in this presentation... in the time value of the options... which applies equally to both sides in puts and calls... and grows very large as longer times are considered.
So, I'd say, a lot of the "real risk" is not apparent on the short term charts... its buried in the "constant" (shrinking) in the time value of the options prices... which has to be "scaled" as a time function vs $ relative to the risk and time functions inherent in the indices... so that it would be deviations from a linear progression / regression in the time value in the options prices relative to the... etc. But, traders expectations have skewing influence too... as, modified from Investopedia "volatility is a mean-reverting phenomenon, [so, $VIX related products] often trade higher than they otherwise should during periods of low present volatility (pricing in an expectation of increased volatility) and lower during periods of high present volatility (pricing a return to lower volatility)".
Anyway... its not just futures traders who can dabble in the trade based on expectations of directional trends or trend changes in SPX... You might just place your bets by buying options on SPX yourself and ignore the $VIX related derivatives based on those options... or you might note the ETFs related to SPX... including the SPXL (long) and SPXS (short) being 3X leveraged to SPX... before considering the higher risks in VIX related choices.
Here's a chart showing how SPX, SPXL, SPXS relate to the $VIX related choices, by comparing them to the implied leverage in $VVIX (lowest) to $VIX and $VXST (highest). As you can see, still using the recent two hour view, the 3X leveraged choices come in just slightly more leveraged to SPX than the least volatile of the VIX related futures.
On the more leveraged side, assuming you don't want to trade the options yourself... you can select from a range of ETF / ETN related products that seek to mimic the $VIX or $VIX futures related functions...
And, stop there, first... to consider [as noted above in: "a lot of the "real risk" is not apparent on the short term charts... its buried in the "constant" (shrinking) in the time value of the options prices] that the "holding periods" intended for these things... is "short term trades"... as otherwise the bias that is imposed by the expense in the leverage and the cost of time will kill you. A chart of $VIX over time will show variations around a center "relative" value. A chart of a $VIX related ETN over the same time will show it rapidly wasting your money away... faster if you bet wrong... but also faster if you bet right and wrongly hold too long. Here a one year chart of $VIX and its nearest intended and easily traded ETN proxy in UVXY... which says it intends to, and sometimes does, mimic the intra-day performance of $VIX. As you can see, it is a rapidly wasting "holding"... as It is not meant to be a "holding"... it is a short term "TRADE"... to capture intraday or shorter term moves.
And, "guessing wrong" on which way the market will move, and time wasting are not the only risks... as the intra-day variations from the $VIX base can also be very large. These tools do not follow $VIX linearly as price quotes on a futures contract... but are easily spoofed away from that expected $VIX relationship by (fund management alterations or) traders bidding them up higher or pushing them down lower than the implied $VIX value during the day. Some of that may be a result of "anticipating $VIX moves" as noted above... but much is also likely to be "anticipating other traders moves"... spoofing... requiring you manage awareness of any deviations growing between "the price" and "the (assumed) basis". So, large price risks can grow quickly as prices move rapidly in a tool like UVXY... when the SPX price is not moving that way. Or, similarly, as price deviations are slowly built in over the course of trading in a day... and then suddenly resolved in a big move to "correct to trend".
That was clearly apparent in UVXY last year... when they changed the gearing from a claimed 3X to a putative 2X that clearly was NOT a 2X... they say 1.5X now... but a year ago leading UVXY to some wide variations from "expected" performance... likely due to management altering (their own) risks not in the leverage applied, but by shifting the time focus in contracts to a much higher than advertised percentage of longer / later dates...
Investopedia covers "some" of those risks in "trading dynamics" as: " Beware the Lag: Investors considering these ETFs and ETNs should realize that they are not great proxies for the performance of the spot VIX. These funds can be expected to perform very differently from the VIX. Some may rise or fall in tandem with VIX, but the rate at which they move and the lag time can make pinpointing entry and exit points challenging even for seasoned traders".
And, that is true... trading these things might rapidly make you "seasoned"... or make you broke... or both. I've deemed it "not worth the risk" to trade them for a living... but still well worth it as "seasoning" to try to learn what you can from how they do trade. IF you participate... expect a long learning curve... and keep participation deliberately small... set hard limits... to limit risks until you have a handle on "how they trade". Success requires you to have that awareness in how they trade, and how to trade them... rather than "having an opinion on what the market will do next"... which as "everyone has one" is not really a qualification for participation.
A range of products exist... which mimic different $VIX related futures...
The UVXY is a $VIX mimic (in theory, 1.5X leveraged to $VIX, but doesn't often appear to perform like it ). The VIXM and VXZ chart the same, but are structurally different... not overly close mimics of the "mid term" $VXMT which they track below... while VIXY is another "short term" $VIX focused mimic like UVXY... but also fails to track $VIX well.
And, not my focus today... but SVXY is a 0.5X leveraged short $VIX ETN... that by the inversion captures some of the benefit of the "shrinkage" penalties paid in UVXY... so does better than the gearing suggests it might IF you use it in a steadily rising market. [ as of Dec. 31, 2021: One month: 14.35%; Three months: 13.05%; One year: 48.66%;Three years: 13.27%] That might suggest trading strategies other than simply reversing field on long / short positioning at trend reversals ? Not getting into it today...
As that suggests, there's another level or two of of risk there still to consider, if not apply... as the $VIX intrinsically leverages "change" in SPX options flow with a high gearing ratio... and you can manage the level of that risk by selecting dates in futures contracts (or derivatives of them in specific ETFs) that amplify or limit the risks in relation to the $VIX spot price...
But, if the VIX is, as the charts suggest, about a 6X to 12X leverage applied to the SPX under active trading... it is also a dynamic application of that leverage the degree of which is determined in and by the trade... so that in moving away from the mean, it easily swings that leverage higher by a factor of 2X (or more) with sharp reversals or during accelerations into the extremes... while it clearly shrinks as mean reversion occurs... as you see happening at both ends of the two hour charts, above...
The SPX options are a leveraged derivative of the S&P500... the $VIX is the derivative of the volatility of the options in that trade... the futures that track VIX are derivatives of it... as the ETFs and ETN's that track the VIX futures are derivatives of them...
You know those commercials that feature some random person stepping into some situation to participate with the explanation... "No, I'm not an expert, but I'm an investor in QQQ... so..." an obvious rip off of the commercials about "I'm not a doctor, but I did stay in a Holiday Inn Express last night"... itself derivative of "I'm not a doctor, but I play one on TV"...
I'm stuck coming up with the right scenario in which to intrude with that derivative of the ploy as a counter to the QQQ commercial... "No, I'm not a central banker, but, I do day trade UVXY... and unlike central bankers... I'm not completely broke, yet" ?
Anyway, leveraging that thought... the last level of abstraction... is that UVXY (as also VXX and the "3X" ETFs like SPXS / SPXL, SQQQ / TQQQ, etc.) also have options that trade... so you can add another level of abstraction, leverage and derivativization by trading the options on UVXY instead of UVXY itself... and, paradoxically, both increase your risk exposure... and reduce your total capital at risk in the trade relative to the potential advantage in the trade... while reducing capital at risk more... by limiting WHEN you trade... to those periods when the time value in the options has already been largely spent...
Mean reversion, and deviations from it, are the core concept at work in my posting "correlation charts" here recently... and the time functions in them... as useful in deriving likely influences in the timing of other things...
including "inflection points" where reversals are likely to occur.
Those things are all still subject to "time drift"... and to error induced by ignoring correlation events and changes tied to them... as "performance alteration events"... being "the point" of the subject of "Cherry Picking Chart Dates"... ?
How things do correlate on the chart... depends on the chart point in time you pick to derive the correlations from... which... that horse seems dead... I'll just post a few charts with different start dates and leave it at that... to puzzle over why these things appear to vary with that time function as they do ?
|
