XLF Volume Spikes 3.3 Standard Deviations Above Mean Yesterday

If I could pick only one ticker to watch in order to gauge the market’s health in the current environment, it would probably be XLF, the most popular of the financial sector ETFs. You could make an argument for RKH, the regional bank ETF, XBD, the broker dealer ETF, or any number of others, but XLF covers the entire financial sector, from Allstate (ALL) to Zions Bancorp (ZION).

With all the talk about the degree of a VIX spike needed to signal a bottom and other measures of capitulation, I am surprised I have not heard anyone else mention the volume in XLF yesterday. As shown in the graphic below, XLF traded over 469 million shares yesterday, eclipsing the previous volume record (set just last Friday), by over 150 million shares. The 469 million share turnover also represents 3.3 standard deviations above the mean, which translates into an extremely unlikely event. [Note that in the chart below, the Bollinger band settings for volume are for 3 standard deviations instead of the default 2 setting] This is capitulation-level volume in the sector that is most important to the stock market at the moment. If XLF can weather all the financial sector earnings due out tomorrow, I suspect that a bottom will be in for the financial sector.

On Measuring Volatility

Mike at HEDGEfolios.com has a good post up today with the title of Measuring Volatility. He touches a lot of bases, but it all starts with the following statement:

“When it comes to measuring or sensing stock market volatility, I do not follow the VIX.”

Now I may have invented that silly tagline, “Your one stop VIX-centric view of the universe,” but I am the first to argue that a defaultist mind set is the wrong way to approach the investment landscape. If you follow the same indicators with the same default settings as everyone else, you are setting yourself up not just to follow the crowd, but to be a half step behind it. In order beat the crowd, what is needed is a variant perception.

Back to HEDGEfolios for a moment:

“The key element of volatility using traditional methods like the VIX rests on the reversal at extremes in a contrarian indication such as buying when the VIX exceeds 30. This is a very dangerous concept and I do not advocate for its use… I never liked that approach so I do my own thing and look at each stock, the turnover in each and how the composite of all signal changes indicates the market volatility.”

Volatility is a wide-ranging concept. It can be defined, measured and applied to over 10,000 stocks and ETFs in many different ways. To think that best way to harness information about volatility is to buy when the VIX hits X is ludicrous.

Consider that the concept of volatility can be applied not just to price, but to volume, options prices, market breadth data, etc. Volatility is a characteristic of every slice of the almost infinite flow of data that is associated with the markets.

It’s not just what you measure, it’s how you measure it. Volatility can look forward when it is in the form of a forecast or a derivation, such as implied volatility. When volatility looks backward, the opportunities to get creative are even richer. There is historical volatility, average true range, Bollinger bands, Chaikin volatility, relative volatility, and a variety of ways in which to index volatility.

Go ahead and watch the VIX, but don’t think for a moment that you are going to have an advantage over the thousands of other people who are watching the same indicator. Sure, you might come up with the next great VIX permutation, but you are far more likely to get a leg up on the competition by revisiting some basic questions:

• Is volatility worth following?
• How can more knowledge about volatility make me a better investor?
• Which aspects of volatility should I pay attention to?
• How should I measure that type of volatility?
• How do I interpret those measurements for maximum ROI?

One of my favorite measures of volatility is the number of buy and sell signals my various systems generate each evening. It’s simple, but effective. And I can be sure that nobody is not going to show up on CNBC tomorrow touting the same approach.

Bollinger Bands and the Percent B Setting

It occurred to me, as I reviewed my last two posts on Bollinger Bands, that it would take quite a few posts to do anything more than scratch the surface of the Bollinger Bands indicator. For that reason, I am going to set aside my thinking on Bollinger Bands after today’s third post in the series, with a promise to pick up the subject again at a later date.

Before I wrap up this introductory look at Bollinger Bands, I want to be sure to mention the %b indicator. Quite simply, %b calculates the current price relative to the top and bottom Bollinger Bands. A %b reading of 0.5 means that the current price is exactly at the middle band (the dotted line in the gray line in the graph below that is a 20 day simple moving average in the default setting). A 1.0 reading puts the current price exactly at the level of the top band and a 0.0 reading establishes that the current price is at the level of the lower band. Unlike some oscillating indicators, %b is unbounded, so in the chart below it will exceed 1.0 when the VIX spikes above the upper band (note the most recent instances of this in January, March and the first week of June.) Each 0.1 increment above 1.0 translates to 10% of the band width above the upper band, so that a %b of 1.25 would mean 25% of the band width above the upper band. In addition to spikes above the upper band, when the VIX drops rapidly, it will periodically slide into negative territory, as it did most recently in February and May.

The %b indicator can also help to identify a change in the volatility climate. If one studies the %b peaks in late July and mid-October of last year, these %b readings of over 1.2 were nowhere the VIX tops. In each instance, however, the extreme %b readings did signal a dramatic change in volatility that ultimately ended with a volatility climax about three weeks after the 1.2+ readings in the %b indicator.

Finally, from a systems development perspective, consider that while converting Bollinger Band chart data into mechanical signals may seem like a daunting task, the %b numbers are often ideally suited for this purpose.

Bollinger Bands and the Standard Deviation Setting

On Friday, I used Bollinger Bands: Why 20 Days? as an excuse to begin my “evangelical crusade against rampant defaultism.” I did this mostly so I could make up a word (or so I thought, apparently others beat me to it), but also to warn against the tendency to favor the default settings in charting software in lieu of rigorously exploring the alternatives.

Using Bollinger Bands as my guinea pig indicator, on Friday I explored what happens when you tweak the default setting of 20 days. The other important setting in Bollinger Bands, of course, is the standard deviation setting. In Bollinger on Bollinger Bands, John Bollinger cites tests he conducted on various stocks, indices, currency pairs, and commodities, the results of which led him to conclude that a 2.0 standard deviation setting was quite robust across a variety of asset classes and time frames. Bollinger went on to recommend that the standard deviation setting should be decreased to 1.9 for a time window of 10 days and increased to 2.1 standard deviations when the time frame is extended to 50 days. In other words, modifying the number of days is more likely to provide additional insight than adjusting the standard deviation setting.

Undeterred, I have had a lot of fun experimenting with the standard deviation setting for VIX charts and a number of other charts. My takeaway: even if you do not discover any particularly revealing new information, it is important to experiment with the default settings to get a better understanding of how the indicator works in default mode and why the bands act the way they do.

In the VIX charts below, I have included the 2.5 standard deviation setting on the top, with the default 2.0 standard deviation setting in the middle, and the 1.5 standard deviation setting on the bottom. As the graphics show, the standard deviation settings can be fine tuned to adjust the frequency of signals generated by the indicator in a given time period. The exact settings should be largely a function of one’s preferred trading time horizon.

Bollinger Bands: Why 20 Days?

I am surprised by how many traders, bloggers, and hybrids (a category I seem to have morphed into) are content to stick with the default settings offered by their favorite charting package. On a related note, one reader had asked me why I stick to the 10/50/200 simple moving averages on most of my charts; another reader regularly decries the laziness of the default charting crowd, adding that they deserve the mediocre performance that their lack of curiosity and imagination undoubtedly lead to.

So consider this the first post in my evangelical crusade against rampant defaultism.

Let’s start with simple moving averages. Personally, I prefer my standard chart of daily bars to have SMA settings of 16, 38 and 155 days. That is not much different from the 20/50/200 setting that is favored by many charting packages, but it works for me and it gives me a slightly different perspective on the world, which often makes it easier to take a position against the crowd. Of course, I have different settings for looking at narrower or broader time frames.

Moving on to an example, Bollinger Bands is one of the indicators for which I rarely see anything other than the default settings of 20 days and 2 standard deviations. Why is this? Is it laziness? Is the indicator so perfectly constructed as to defy tweaking? Is it possible that 20 days and 2 standard deviations work that well over all time horizons? Hardly. Read Bollinger on Bollinger Bands to see what John Bollinger has to say on the subject or do some experimenting on your own without the safety net of Bollinger’s commentary (for more information on Bollinger Bands, check out John Bollinger’s Bollinger Bands web site.)

In the three charts below, I have modified the Bollinger Bands setting to reflect a 25, 20, and 15 day setting – while keeping (at least for today) the 2 standard deviation parameter intact. Without providing too much commentary, note that in the top chart, the 25 day setting casts what I call a longer ‘time shadow’ to the right of any major moves. This time shadow is muted somewhat in the second chart (20 days) and is even less noticeable in the bottom chart (15 days). Note also how the 25 day setting almost resembles Donchian Channels (aka price channels), while the 15 day setting hugs the price action closely enough so that even a gentle trend can exceed the standard deviation bands. Consider, for a minute, the possible implications of customizing Bollinger Bands for securities that have a greater tendency to trend (i.e. commodities) vs. those that have more pronounced mean-reverting characteristics, like the VIX.

NASDAQ Chart with Hourly Bars

On Friday I posted a weekly chart of the NASDAQ going back six years. Today I am swapping the wide angle lens for a telephoto one and focusing on hourly bars for the past month. I am emphasizing the NASDAQ Composite (and the NDX) because that is where a good deal of the speculative activity has been as of late and where the correction was most severe last week.

The graph below shows four semi-arbitrary simple moving averages for the NASDAQ Composite Index: 13 bars; 20 bars (the faint dotted gray line in the middle of the Bollinger Bands); 30 bars (the same as the Williams %R time frame); and 100 bars. All this spans roughly a period of 1 ½ days to about 3 weeks. With the aforementioned Bollinger Bands and Williams %R data, as well as the Fibonacci retracement lines, there are many ways to keep score. As much as anything, however, I will be looking at the intensity and duration of the bull rallies off of the bottom. Given that many indicators point to the current situation as significantly oversold, the absence of a compelling bull rally may provide as much information as what is actually happening. In other words, a draw should favor the bears.

Finally, I still haven’t seen much in the way of fear yet…

Charting the VIX with 10 Day SMA Envelopes

While the VIX is currently in the process of what looks like a mini-implosion, I thought this might be a good time to offer up what may be the best single chart for keeping track of the VIX.

To understand where I am coming from, I should say up front that I consider the 10 and 20 day simple moving average to be two of the most important pieces of VIX data to watch – and the keystone for short-term mean reversion analysis. Of these two SMAs, I favor the 10 day SMA slightly over its older sibling.

Of course you can easily put the 10 day SMA and the 20 day SMA on the same chart, but what is often more useful is to know just how far the VIX has strayed from these moving averages. For a quick visual check of this, I recommend using moving average envelopes, such as the ones shown in the chart below. You can follow the link in the previous sentence to see how StockCharts describes these envelopes or just consider that they work in a similar fashion to Bollinger Bands, except that instead of using standard deviations as a measuring device for the width of the bands above and below a mean, the moving average envelopes used a fixed percentage of the simple moving average.

An example helps tell the story. In the chart below, the 10 day SMA (mean) is represented by a solid blue line, with a current value of 14.62. The 10% upper band (currently 16.08) is represented by green dotted line while the upper 20% band (currently 17.55) is shown with a purple dotted line. A similar pattern is evident in the lower bands.

There are a couple of important things to remember about moving average envelopes. First, unlike Bollinger Bands or any other standard deviation-based calculation, they do not reflect any sort of volatility. Instead, their width is entirely a function of the value of the underlying SMA, regardless of volatility. Second, the 20% bands are rarely violated and usually represent excellent mean reversion setups; the 10% bands are violated more frequently, yet also offer many good mean reversion setups.

While there are many ways to watch and analyze the VIX, in my opinion moving average envelopes are among the best tools out there for the chartist trader.

The VIX and Bollinger Bands

I’m not sure how I managed to fritter away five months on this blog and spend so little time talking about Bollinger Bands (BBs) and the VIX, but today seems like as good a time as any to dive right in.

First, for anyone who needs a brief refresher on what Bollinger Bands are or how they are calculated, I refer you to the StockCharts.com overview; for some thoughts on how to apply BBs to trading, OnlineTradingConcepts.com has a good summary. The best detailed source of information on Bollinger Bands comes, not surprisingly, from John Bollinger’s own site, BollingerBands.com. Unfortunately, John does not provide a lot of his thoughts on BBs for free, but you can get a good idea of his thinking from a 2001 list of his 15 basic rules.

I have included a six month chart of the VIX below, with the default (20,2,0) BB settings in addition to a BB width indicator and an overlay of the SPX. For today, I am only going to offer a handful of observations (based on this chart and some research going back past the six month cutoff):

• When the VIX tags its Bollinger Bands, it is usually a good time to think about a VIX mean reversion play
• When the VIX tags its Bollinger Bands, it often signals a short-term change in the SPX trend
• VIX closes outside of the Bollinger Bands (which we are on target for today) are almost always associated with dramatic market moves and/or changes in the trend
• When the VIX BB width drops below 1.8, it frequently signals that consolidation is ending and a sharp move is just around the corner
• Looking back to late-March, a VIX close above 15 would definitely be significant in terms of support and resistance (we are currently trading at 14.76)

Valentine’s Day Massacre: The Curse of Not-So-Random Roger?

The Lauriston Letter (which can be counted on not just for high quality market commentary, but also for an unmatched sense of complementary images) pointed out an interesting tidbit on Tuesday:

“How often do you see volatility indexes move from the upper Bollinger band to the lower Bollinger band in one day?”

The answer, of course, is almost never, but I decided to dredge my archives anyway in search of a reversal that best resembled the two day swing from above the upper band to below the lower band during February 12-14.

I tried to do for Excel what datawink does for charts and caught very few fish – 11 over the past 17 years and only two during that past decade – that resembled the Valentine’s Day swing. The combination of trying to locate CCI and %R reversals of such large proportions in just 48 hours proved too daunting of a hurdle to come up with anything close to an exact match.

Ironically, the two closest matches turned out to be the ones from the last decade: the three days ending on 11/3/04 and the three days ending on 7/11/05. So I pulled up a bunch of charts to see if the VIX or SPX did anything of interest in the days or weeks that followed these two dates and came up empty. If anything, the markets were considerably more quiet than usual in and around this period, and for the other nine VIX swing periods as well, which is consistent with observations made by Brett Steenbarger about the predictive ability of low volatility readings. Perhaps these were just two random trees felled in a distant forest.

Undaunted, I decided to check the archives of some of the blogs that have archives going back to 2004 and made a stop at Random Roger’s Big Picture. This is where it starts to get interesting…

On Sunday, 10/31/04, Random Roger was posting about how Barron’s was “fixated on how low the VIX is.” In fact, the VIX was at about 15 at the time and Roger was quick to point out that the then current 15-16 range was not low by historical standards. The very next day, the VIX started an unprecedented high to low swing, moving from 16.76 to 13.79 in two days.

Eight months later, on July 6, 2005, Roger dusted off his old post and repeated his point that the VIX was still low by historical standards. What happened next? Even more dramatic fireworks, as the VIX pulled a stunning high-low swing from 13.92 to 10.53 in two days, resulting in the largest CCI (20) reversal since the inception of the VIX and confounding even John Bollinger himself.

So those are the two precedents. Is it merely a coincidence that not-so-Random Roger happened to warn about the VIX not being particularly low just before the largest VIX swings to the downside ever seen? And where was Roger before the Valentine’s Day Massacre of 2007? I have seen nothing on his blog that could lay blame at his feet, but would the VIX really have the temerity to make a move of this magnitude without his public approval beforehand?

I will watch this story closely and bring you any further developments…

Absolute vs. Relative Highs and Lows in the VIX

Price Headley, founder of BigTrends.com, is one of many who maintain that the best way to think about the VIX is not in terms of absolute numbers and a multi-year lookback period. He recommends that traders focus instead on relative highs and lows over the course of recent trading ranges. Specifically, Headley favors 20-bar Bollinger bands with 2 standard deviations and opines,

"...forget about where the VIX has been in years past, and instead focus on getting the most of out the VIX using relative bands like Bollinger Bands. You can see that when the market tagged these lower boundaries for the VIX, as in mid-December and mid-November, then the market is ready to correct or at least go sideways. This follow-up information will prove invaluable as a contrarian, being willing to sell the market when the VIX is at its lower band, while being able to get more aggressive when we see fear spikes to the upper band. "

For the record, I am a strong believer in applying Bollinger Bands to the VIX and am of the opinion that Bollinger bands can provide solid trading setups not just for the broader markets, but for the VIX as well.