Over this weekend’s Stocktwits Brunch, I mentioned how the past week was the most statistically significant upmove we’ve seen in the past year, if not the entire bull market.
So how did I figure that out? It goes back to the options.
If you remember one thing about the options market, it’s that it is a risk exchange.
That means when you trade $AAPL options, you are not trading the company, you are not trading the stock, you are trading the risk within the stock.
This risk is seen in the implied volatility, priced in through the extrinsic value of the options.
This pricing assumes a certain kind of price action in the future, that the market will stay under a kind of bell curve.
What we can do then is go into the past and see where price has deviated from its “bell curve.”
Enter Historical Volatility
In the options market, Implied Volatility looks into the future. To look into the past, we use Historical Volatility:
If you’ve traded the markets at all, you know that it doesn’t trade cleanly underneath a bell curve or any smooth surface. So measuring the deviation from the expected movement can be helpful, especially when trading options.
For example, if a stock makes a 2-sigma move, that means it is 2 standard deviations from the expected movement– this should happen only about 2% of the time… it happens more often than that, which is what makes trading options so much fun!
This study was given to me by a fellow trader but the original source was created by Tom Utley, and it’s derived from one of Jeff Augen’s books on option trading. You can see the study here:
The study is simple: if the day was up it’s a green bar, if the day was down it’s a red bar. The time period is set by the user, and I prefer 20 on a daily chart because that looks back 1 month in time (20 trading days = 30 calendar days). The deviation from that period’s historical volatility gives the bar its magnitude.
How to Use It
What’s interesting about this is the adaptive nature of volatility and how a larger move on an absolute basis can have a reduced measurement on the chart– this is because context matters.
Can you use this to time the market? Probably not. If anything, it can show you a “shot off the bow” to the downside– but this becomes much more important when trading options and picking points in which to sell them.
Now that you know how to properly use it, here’s the code:
# Tom Utley 3-17-2009
# Thanks to Jeff Augen
# Price Spikes in Standard Deviations
input length = 20;
def closeLog = Log(close / close);
def SDev = stdev(closeLog,length) * Sqrt(length / (length-1));
def m= SDev * close;
plot spike = (close – close) / m;
spike.AssignValueColor(if close > close then Color.UPTICK else if close < close then Color.DOWNTICK else GetColor(1));
You can copy and paste this into thinkorswim and it should work nicely for you.
Comments? Questions? I’d love to hear them, let me know in the comments below.