Standard Deviation: The Tails Are Fatter
Standard deviation measures how widely returns are spread around their own average, and it is the number behind Bollinger Bands, volatility readings and most risk models. Financial returns do not follow the bell curve it assumes, which is why extreme moves arrive more often than the maths predicts.
How it works
Standard deviation measures spread. Two markets can have the same average return and be completely different to hold; the deviation is what separates them.
The calculation squares each difference from the mean, averages those squares, and takes the square root. Squaring is what makes large deviations dominate the result — a move twice as far contributes four times as much.
Say a stock’s daily returns have a deviation of 1.5%. A normal model then treats a 4.5% day as a three-deviation event, the kind it expects only about three times in a thousand sessions.
Where the bell curve breaks
The number is usually paired with a normal distribution. That model says 68.3% of observations fall within one deviation of the mean, 95.4% within two and 99.7% within three.
Measured on real returns, the shares usually come out different. More observations sit inside one deviation than predicted, and more land beyond three.
Both at once is the signature. A distribution can be more peaked in the center and heavier in the extremes than a bell curve, with the shortfall coming from the shoulders in between. That is the standard shape of financial returns, usually described as fat tails.
Across 1,000 daily returns, a normal distribution expects about three beyond three deviations. If an instrument delivers twelve, extreme moves are arriving four times more often than the model allows, which is the entire practical content of this page.
In practice
It is a price statistic. A wide range on no volume and one on heavy volume produce the same reading and mean different things.
Scaling across timeframes uses the square root of time. A daily figure multiplied by the root of 252 gives an annual one — a convention that assumes independence between periods, which markets do not entirely have.
Because differences are squared, one gap can move the whole figure. A single outlier in a hundred observations can raise the deviation noticeably, which makes short windows unstable.
This is the practical consequence for a stop. A level set at an extreme the bell curve calls rare will be reached more often than the curve says, so a stop placed on the model’s odds is hit more often than the plan assumed.
Costs do not shrink when the reading does. A round trip still pays the spread and commission when the deviation is low, and a quiet market leaves less movement to pay them from.
Every deviation figure describes a window that has already closed. Using it as a forecast assumes the next period resembles the last, which is a reasonable working assumption and not a fact.
What to do about the fat tails
The fix is not a better distribution, it is a smaller position. Any model that treats a three-deviation move as impossible will be surprised several times a year, and the only defense that works regardless of which model is right is a size that survives being surprised.
The second adjustment is to stop reading probability into band touches. A price outside two deviations is not reliably a 4.6% event; the true frequency depends on the instrument’s own history, and with news risk it can be higher. Treat deviation bands as a description of recent range and not as odds, and most of the trouble the measure causes disappears.
One more property is worth knowing because it explains a lot of odd behavior: the measure is not robust. Because differences are squared before averaging, a single unusual bar contributes far more than its share, and a twenty-bar window containing one outlier can report a figure double the one it reported the day before.
That instability is why short windows produce bands that lurch. The alternative measures — average true range, or the average absolute deviation — do not square anything and are far steadier for the same window length. Neither is better in general; they simply weight extremes differently, and knowing which one your indicator uses explains most of the disagreements between two volatility readings on the same chart.
What standard deviation is not
It is not a probability. The odds come from the distribution you assume.
It is not direction. A high reading says nothing about which way.
It is not risk. It measures variability, and losing money is a different thing.
And it is not stable. One outlier moves it, because the differences are squared.
How this connects
Keltner channels use average true range where Bollinger Bands use this measure, which is why the two envelopes react so differently to a single outlier bar.
The squeeze setups compare the two directly. A Bollinger squeeze is the low reading described in the failure section, and the TTM squeeze formalizes it by flagging when the deviation bands sit inside the range-based channel.
Implied volatility is the forward-looking version. Options prices embed an annualized deviation the market expects, while the historical figure measures only what already happened.
Volatility clustering is why a recent window still tells you something. Turbulent periods tend to follow turbulent periods, so the last reading is a reasonable short-term guide even though it is not a forecast.
Risk-adjusted measures inherit its flaws. The Sharpe ratio divides by it directly and the Sortino ratio swaps in the downside version, so the fat-tail problem on this page carries into both.
Other envelopes use the same arithmetic. VWAP bands and standard error bands both place their lines at multiples of a deviation, around the volume-weighted price and around a linear regression line respectively.
And positions that are short the extremes carry tail risk that any model built on this figure prices as rarer than it is.
When it fails
In a range the reading falls to its lowest just before the break. Quiet conditions produce a small deviation, a small deviation produces narrow bands and confident sizing, and the move that follows is measured against a figure computed when nothing was happening.
The second failure is assuming normality. The bell curve is a convenience, and the tail counts on any real instrument show what it costs.
A third is a window too short to be stable. Twenty observations produce a figure one outlier can double.
A fourth is comparing readings across instruments. A percentage deviation is comparable; a price one is not.
And a fifth is confusing it with risk. An asset that rises steadily has a deviation, and no drawdown.
The original data
The S&P 500’s worst decline shows why the tails matter more than the average day. From 1,565.15 on 9 Oct 2007 the index fell 56.8% to 676.53 by 9 Mar 2009, and it took until 28 Mar 2013 to close above the old peak. Across 19,299 trading days since 1950, 53.1% closed up: the ordinary day is close to a coin toss, and the damage is concentrated in rare stretches that a deviation figure fitted to calm years treats as nearly impossible.
The question: A three-deviation bar just printed. Fade it?
The count of extreme days is the number to carry. A deviation calculated from quiet years describes those years, and the bad stretches arrive as clusters the average never anticipated. Before using any deviation-based level, count how often your own instrument exceeded it, and size the position against the count rather than the theory.
Related
Bollinger Bands are the most common trading application. Historical volatility is the same figure plotted over time. And risk management is where the sizing consequence lives.
I do not trust the bell curve attached to standard deviation, though the measure itself is fine. Once you accept that the extremes arrive more often than the model says, position sizing stops being about the typical day and starts being about surviving the atypical one.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.