Sign up  |  Log in

Trading volatility

Have any of you fellows traded the VXX ETF (iPath S&P 500 VIX ST Futures ETN)? Or do you have a better way to make money on volatility that does not include confusing derivatives?
 
I’m not strong with derivatives, but I have a very good sense for fear and chaos. Even on the other side of the planet, I can feel the fear in New York. As an experiment I studied VIX, and the VXX ETF, during this oil freak-out. You can feel the fear and uncertainty peaking Tues before the Fed meeting, and it was my predication that it would, and it settled down post meeting with Fed confidence and prices holding. If you shorted VXX on Tues (betting on decreasing volatility), and released it on Friday, it was 8%, 4%, and 2% daily returns. In fact you could ride it up, and then back down. 
 
Why can’t I make money on this every time we get an ebola, oil, whatever freak out given my super-human sense of fear? Is there an aspect I’m not seeing where I get burned? And what the hell are these instruments anyhow, futures on what exactly??

Automate your Excel models with the industry's most accurate financial, market and economic data.

purealpha wrote:

Why can’t I make money on this every time we get an ebola, oil, whatever freak out given my super-human sense of fear?

There are tens of thousands of people who try to do this every day in the VIX or the underlying. Many of them end up losing money over time but a few will do very well.

Do your savings account a favor and don’t do this. 

Minerva wrote:

Do your savings account a favor and don’t do this. 

Second

^i traded the vxx and dived into levered inverse tbond etfs too (TLT,TBT etc) when i first started trading. Moral of the story, the tracking error is extremely large and the upside is usually 70% of the downside. Furthermore, and i think this is written in the vxx etf, it says it will approach 0. i third the comments, stay away and learn from my expenses

whatsyourgovt wrote:

^i traded the vxx and dived into levered inverse tbond etfs too (TLT,TBT etc) when i first started trading. Moral of the story, the tracking error is extremely large and the upside is usually 70% of the downside. Furthermore, and i think this is written in the vxx etf, it says it will approach 0. i third the comments, stay away and learn from my expenses

Yeah, that’s interesting. I noticed of course the chart approaches zero over time, and thus one would expect the upside to be less than the downside especially the longer you hold it (not long). Wasn’t aware the tracking sucked so bad. And I still don’t really understand what these underlying instruments are which violates the “invest in what you know” rule I follow.

That said, I am fairly confident in my ability to forecast short-term fear/volatility better than the average speculator. For example I totally nailed my S&P500 short these last weeks, hedged my equity portfolio near the top, released it Wed AM in pre market and let everything float back up. This is working off my sense of market fear, so I reason that in addition to hedging I can perhaps make a little extra on the side with VIX speculation…

…but seems somewhat dicey.

It is dicey.  I have won and lost in VXX and others like TBT.  If you’re going to do it, then hold if for short periods of time.  No more than a few days or you will get steadily ground down to zero.  Remember two-three years ago when the shiz was hitting the fan all over?  The VIX didn’t blink.  I fourth the comments above and warn to stay away.  I learned from my expenses too.

Sweep the Leg: "I’m tired."
KMeriwetherD: "Well, you were basically Legolas in the Battle of Water Cooler."

isnt vix based off  put/call contract volume? or am i thinking of something else?

"You want a quote? Haven’t I written enough already???"

RIP

igor555 wrote:

isnt vix based off  put/call contract volume? or am i thinking of something else?

something else. Vix is based on implied vol, calculated based on S&P options prices, not volume, over a specificed range (30 days out i think)

villnius wrote:

igor555 wrote:

isnt vix based off  put/call contract volume? or am i thinking of something else?

something else. Vix is based on implied vol, calculated based on S&P options prices, not volume, over a specificed range (30 days out i think)

Correct.  And because the VIX index is based on the spread between option prices, it can be subject to manipulation by those who place ridiculous bid/ask quotes in illiquid securities that never get executed.

The VIX index is based on options expiring <30 days.  The VXX ETN is priced on options expiring in the front month AND back months.  So this “tracking error” is actually due to the market structure of volatility, ie contango or backwardation, causing a negative roll yield.

Volatility has been in backwardation since 2009, so that is why the VXX has moved from 1000 to nearly 0 (split adjusted) since 2009. 

But on this last market correction starting October 2014, I saw a few days where the VXX outperformed the VIX.  Things are now back in backwardation, but to less of a degree imo just based on the price action.  Bottom line, expect volatility to increase during 2015 and transition into a state of contango.

“They may do great things or not (they deal with poop), but that is besides the point.” - Greg Harmon, CFA CMT      in regards to $ASMB

JBrowntown wrote:

I fourth the comments above and warn to stay away.  I learned from my expenses too.

Yeah, I’ll probably stay away.

But here’s a silly question; given the nature of this security (declines to zero), when VIX hits 20 like last week, why don’t people just short it with the intention to “sell and hold” for the long term.

Looking at the last 12 months that would be a 45% return (assuming one didn’t pull it off when the VIX was high in Feb, which would be an 82% return YTD). Three year returns would have been astronomical! My broker charges 1.5% a year, I checked and the lending rates are not higher for this security. Of course there could be massive volatility over the year, yet is seems mandatory by the nature of the security that it eventually declines?

^ If you did this, I’d be sure to have some OTM calls somewhere before bankruptcy, just in case.

“I can no longer obey. I have tasted command, and I cannot give it up.”

Yeah, the above play seeks to profit from a piece-of-crap security that functions in a bizarre way over time. Yet placing any sorts of bets on a security that one acknowledges as functioning bizarrely is kinda crazy. ;)

Therefore insurance is mandatory!

Hmm, I’m going to think this thru some more. Volatility has been creeping up, and I think it will remain high next year. Yet for the last five months VXX is basically flat. If one were to execute this strategy, perhaps next year around Q2 when avg volatility is higher, during a particularly wild patch, would be a good time to go short and hold.

smhsmichael - I think you’re confusing contango/backwardation….contango is an upward sloping term structure (longer dated is higher than shorter dated) and this is what it looks like 75% of the time.

purealpha - The nature of the security is not designed to go to zero, there’s no guarantee of that.

There’s no true underlying to the VIX in the sense of a security or an asset you can hold or borrow.  That means that you can’t short it (or even hold it) without using futures indexed to a number produced by a formula.  And you have to roll those if you are going to hold it for a long time.  Over the long term, you get killed on the roll, which is why shorting a spiked VIX isn’t guaranteed to make you money just by holding it long enough.

Even if you could be reasonably sure that the VIX would eventually get to 0 or some low number, it could take so long to get there that the CAGR doesn’t make it worth the volatility in the interim.  If the VIX takes 5 years to go from 40 to 20, that’s (without transaction costs) around 8% gains from shorting per year.  Meanwhile the S&P has been giving you much better returns, even before adjusting for the levels of risk.  Add transaction costs, and the frequency of truly shortable spikes, and it just doesn’t seem like the right sandbox to play in.  And remember that you’re unlikely to be able short at the true top of the spike, and if you miss that top, you lose a lot of the gains.

I once had a system for trading the VIX that looked phenomenal in a backtest, but then I tried it with VXX as a proxy, and it was dubious at best and a shredder of capital at worst.  Tracking error with the ETF and paying the roll yield on the VIX itself ate all the alpha that appeared in the backtest.  It also didn’t help that the VIX tended to spike, so that (if I remember right), most of the gains came from a few short spikes, very unevenly distributed.

The lesson for me, the VIX is a true financial siren, with all the allure and the danger of the mythical creatures.

You want a quote?  Haven’t I written enough already???

purealpha wrote:

whatsyourgovt wrote:

^i traded the vxx and dived into levered inverse tbond etfs too (TLT,TBT etc) when i first started trading. Moral of the story, the tracking error is extremely large and the upside is usually 70% of the downside. Furthermore, and i think this is written in the vxx etf, it says it will approach 0. i third the comments, stay away and learn from my expenses

Yeah, that’s interesting. I noticed of course the chart approaches zero over time, and thus one would expect the upside to be less than the downside especially the longer you hold it (not long). Wasn’t aware the tracking sucked so bad. And I still don’t really understand what these underlying instruments are which violates the “invest in what you know” rule I follow.

That said, I am fairly confident in my ability to forecast short-term fear/volatility better than the average speculator. For example I totally nailed my S&P500 short these last weeks, hedged my equity portfolio near the top, released it Wed AM in pre market and let everything float back up. This is working off my sense of market fear, so I reason that in addition to hedging I can perhaps make a little extra on the side with VIX speculation…

…but seems somewhat dicey.

There is a very good technical reason why overtime double ETF value will approach zero. 

Also, the VIX is based on implied volatility (volatility that is already expected).  So your bet is to expect more volatility than everyone else.  In essence, VXX kind of acts like an insurance policy where the cost of insurance consistently erodes the value of the contract.  It’s good for short-term hedging, but bad for long-term holding.

Sweep the Leg: "I’m tired."
KMeriwetherD: "Well, you were basically Legolas in the Battle of Water Cooler."

Retail investors have gotten more involved with trading vol over the last year, especially with respect to going short vol. There was an article about it in the FT the other day describing the shift in open interest.

The gist I get from this whole thing is that retail people should not trade vol as it is a zero-sum game and very hard to make money while cometing against an honest to god vol desk. Also, the fact that apparently retail investors and others that that don’t normally actively trade vol are getting so fed up with paying for insurance for the last 5 years (with the memory of 2008, no doubt) with no real upside are actually switching to long equities and short vol is a sign that greed is getting too strong in the market.

Also, there is a conspiracy theory that Gross’s short vol positions that were unwound when he left PIMCO caused a lot of to heightened volatility in the market in October. Found that interesting.

It’s whatever, just make it count.

- kDot

brain_wash_your_face wrote:

Also, there is a conspiracy theory that Gross’s short vol positions that were unwound when he left PIMCO caused a lot of to heightened volatility in the market in October. Found that interesting.

This is right - well, not 100% but it’s close to the truth. 

brain_wash_your_face wrote:

Retail investors have gotten more involved with trading vol over the last year, especially with respect to going short vol

I sincerely hope they stopped doing that lazy short-only trade after first week of October - that’s when vol business came back to life last year. Volatility is a very nice product to trade (if you understand the theory right) and get that OTC volume drives whatever you guys see on screen. And there is no way for a retail guy to know what my broker did in the pit or my competitor did in OTC market. So I can completely understand and very much sympathise with retail folks - regulation on this front is definitely not uptodate. 

I would advise against any retail investors trading options. In general, retail investors do not understand options and they do not have the sophisticated vol fitting and other tools used by institutions. Banks and hedge funds see a different world compared to you. In addition, trading costs are high for options and this will increase your watermark. If you really want to bet your money on spurious and volatile things, then trade futures or FX. At least in these assets, your trading costs and technological advantage of larger institutions are minimized, so your odds of making money are higher at probably slightly less than 50%.

VXX (and similar ETFs like VXY) invest in VIX futures. VXX will buy the two nearest expiring contracts, and roll out of the front month as it nears expiry. Since the term structure of implied volatility is usually upward sloping, your position will generally decay over time. The same applies to being long regular options, but for slightly different reasons. Anyway, what I am trying to say is that you only make money on VXX if your timing is so great that you buy it right before a volatility spike. Otherwise, you will keep losing money.

VIX is an index of implied SPX volatility. It is not necessarily related to realized volatility. Furthermore, it is not investable, since the VIX measurement contracts change by strike; you cannot buy an option that is always ATM. It’s a neat thing to keep track of, but it is a misleading index, since it can change just by the different selection of ATM options, even if implied volatility doesn’t change.

If you really want to make money on options, sell out-of-the-money puts. These are priced at high implied volatility, since people’s utility curves are decreasing in slope. Of course, if you are otherwise long equities or other risk assets, this will magnify your potential losses (and it is why people pay a premium for those strikes in the first place).

“The aliens on Mars cloned Hillary Clinton.” - Turd Furgeson

ohai wrote:

I would advise against any retail investors trading options.

+1: Especially in volatile markets, retail investors have no way of knowing how/why the vol moved so much in such and such strike region. That is mostly because they don’t even get the same information we do. So yeah, be very careful.

bchad wrote:

Even if you could be reasonably sure that the VIX would eventually get to 0 or some low number, it could take so long to get there that the CAGR doesn’t make it worth the volatility in the interim.  If the VIX takes 5 years to go from 40 to 20, that’s (without transaction costs) around 8% gains from shorting per year.  

Hmm, I agree it’s necessary to take financing charges into account, and oportunity cost of tied up capital, but those time frames have never happened. VIX goes to 30 occassionally, 2010 flash crash to 45, 2011 selloff to 50, 2008 subprime meltdown to 100, and 1987 black monday 150. Futures and VXX don’t seem to spike as high as VIX though, and the lengh of time is usually very short, although it can in rare cases stay elevated for months. 

Rare Events: http://nastrading.com/vix-index-and-futures-performance-during-extreme-e...
Price history: http://en.wikipedia.org/wiki/VIX#mediaviewer/File:VIX.png
 
I hear what everyone is saying (most of it I get, though some of the technical is over my head). But I still don’t see why people with good market instincts can’t make money here consistently (via shorting peaks). So I modeled it out…
 
If you short VXX when VIX is performing a mildly-high peak in 22-25 range, you need around 3.5X capital to survive a VIX=100 event. For example if you short $10K, you need $35K cash total to hold thru a possible disaster event. Really that should be more than sufficient to survive a subprime-type collapse since the ETF is not as volatile as VIX. Return as calculated on the full 3.5X capital would be around 3.5% and take around a week (in the vast majority of cases these peaks settle in a week or so, storms don’t blow forever). So if you bang this out 4X a year on “OMG Ebola!” type peaks that’s a 14% return assuming VIX=22 peaks, more if you time a 30 peak, and massive money if you time a 50 peak. Also this capital IS NOT tied up for the full year, it is only tied up during VIX peaks, you can use it for other short-term moves when VIX is in a normal range.
 
However if you did this for years you would eventually short too early on what seemed irrational fear (ebola) or freak-event (flash crash), but it would turn out to be a true crisis. Since you have the capital you are fine. But 1) you would need to hold for maybe 6 months max which would cost you say 1% on the 10K, plus opportunity cost on the 25K, 2) when asset prices bottomed out you could not use this capital to go on a buying spree, but 3) having missed the peak the first time, you could use additional capital to double up and short a second time. Let’s say this freak event happens every 2-3 years.
 
Well guys, that’s still pretty good money, especially if you are in a sideways volatile market and have cash to play. What did I miss here?
 
As an experiment I shorted VXX @ 36.82 on Friday (nailed the interday peak). S&P500 returns are going to suck, and I think the market is schizo torn between irrational optimism and reality. But soon that settles down, perhaps good news on a few earnings next week, and I make a fat return by covering @ 31.85, OR this starts the end of the bull market and my position goes hella red. :)
 
Update: closed it out in pre-market for an easy 3% gain. Didn’t feel like pushing my luck in this schizo-market.

You missed this part:

1) you would need to hold for maybe 6 months max which would cost you say 1% on the 10K, plus opportunity cost on the 25K

You’ll lose more than 1% over six months from the roll-yield.  More like 40% would be my guess.

Sweep the Leg: "I’m tired."
KMeriwetherD: "Well, you were basically Legolas in the Battle of Water Cooler."

I’ve had this link for a few years, https://www.collective2.com/system65806998

smarter than VXX

New and improved products from Accushares:

VXUP

VXDN

Lose less to roll yield.  These are brand new, so arbitrators need to step in and set the NAV to actual market value.

Sweep the Leg: "I’m tired."
KMeriwetherD: "Well, you were basically Legolas in the Battle of Water Cooler."

VXUP/VXDN, interesting concept. But seems like the prices will deviate a lot from NAV when buy/sell is not balanced.

Edit: Wait, I guess there is no NAV, since the fund does not actually hold underlying securities.

“The aliens on Mars cloned Hillary Clinton.” - Turd Furgeson

VXX just keeps going down, down, and more down. Would have made a fortune if I had the balls to hold my short…

nice topic cheeky

Be aware of huge rolling costs of VIX future!

VIX is US (SPX) at the money 30 day volatility, very restrictive, why do you want to trade exactly it? 

Sorry For My Bad English

 

Have a look at 6 months to 12 months vol, curve looks better

(esp in Europe now -  DAX, Eurostoxx vol, single stocks, 

look at euro vol coz of Greece, IMHO you better short it (as spread to VIX) instead of just shorting VIX lol)

I personally like more options in most cases - have a look across all strikes, also nice play on delta possible.

 

Sorry For My Bad English

negatives rates wrote:

…why do you want to trade exactly it? 

Well for one reason, you would have made crazy returns if you shorted it back when this thread started. That seems like a good reason to me?

sorry for late response, what time did you want to short?

give me levels, and returns %

See, if vol spikes, you have no proper protection, although VIX is not soooooo dangerous as single stocks, where vol can jump overnight (like in 2008), when dispertion trades (bet on correlation) killed lots of volatility traders.


Sorry For My Bad English