WilliamsVixFix Explained – Does It Work?

Last Updated on June 11, 2021 by Oddmund Groette

Back in 2007, the well-known trader and “indicator innovator” Larry Williams wrote an article in Active Trader about VIX and how you can create your synthetic VIX for any security you like. The aim of Williams was to create a “synthetic” VIX reading for other instruments than the S&P 500, the Nasdaq, and the Dow Jones 30.

In this article, we explain what the WilliamsVixFix is, how you can use it, and whether is a good trading tool or not.

What is the VIX?

The VIX is derived from the implied volatility of stock index options on the Chicago Board Of Trade (CBOE). It was introduced in 1993 and is intended to represent the “fear” or “complacency” of the market. Unfortunately, the VIX is only calculated for the S&P 500, Nasdaq, and the Dow Jones 30. Implied volatility is one of the major components in the price of an option: the higher the implied volatility, the more expensive the premium. Think of it as insurance. The more risk, the more you need to pay for insurance. Opposite, when market participants see few clouds on the horizon, premiums go down as the perceived risk is smaller. The more volatility in the markets, the more you need to pay for insurance. In the Black and Scholes option pricing model, implied volatility is the only factor that is “unknown” and subjective.

How does the VIX work?

The VIX oscillates up and down. A high reading means investor sentiment is one of increased fear, while low readings are associated with low-volatility conditions (and market tops). Thus, the VIX has a negative correlation to the S&P 500.

Is a high VIX good or bad?

When the VIX is high, it shows fear is high and vice versa. A high VIX normally means the market has fallen, at least in the short-term, and the risk premium for owning stocks increase. This is normally a good time to buy for short-term traders. The stock market has turned out to be mean revertive, and we all know the long-term tailwind from earnings growth and monetary inflation. A rule of thumb is that low volatility is often associated with market peaks, while high volatility is associated with market lows.

However, the VIX measures the implied volatility for one month. A high reading today might be a good short-term opportunity to buy, but not necessarily a good entry for the long term. As always, make sure you make quantified tests before you do any trades.

Below is the daily VIX readings for the S&P 500 since 2014:

Clearly, the VIX has been very high since the COVID-19 struck. Because of this, we can’t say that one level is good and one level is bad. It all varies and you need to use moving indicators.

How is the WilliamsVixFix calculated?

Larry Williams wanted to make a synthetic VIX for other products and not just the main stock indices. The formula for WilliamsVixFix is as follows:

VIX Fix = (Highest (Close,22) – Low) / (Highest (Close,22)) * 100

What does this mean? In plain English it means the following:

  1. Find the highest close over the last 22 days and subtract the low of today (or the current bar).
  2. Divide by the highest close of the past 22 days.
  3. The result is multiplied by 100 to “normalize” the indicator.

Why 22 days? 22 days represent the normal number of trading days in a month.

The formula is, as you can see, pretty easy. What it measures is the price volatility of the last 22 trading days and is, of course, a lagging indicator.

How does the WilliamsVixFix compare to the real VIX?

This is a synthetic VIX. How does it compare to the original VIX? Below is the chart of the S&P 500, WilliamsVixFix, and the VIX:

 

 

The blue line in the middle is the WilliamsVixFix. Although the readings on the right axis are a little different, we can clearly see the resemblance. The levels, swings, magnitude, and timing are pretty similar.

Because of the resemblance, Larry Williams reasoned that the synthetic VIX fix can be used in other markets. Here’s how it looks at the ETF TLT:

How to use the WilliamsVixFix?

In the Active Trader article, Larry Williams didn’t reveal any specific trading strategies. On the contrary, he suggested the strategy can be used for further use and experimentation.

We tried these strategies:

WilliamsVixFix on Bollinger Bands:

We tried the following strategy on the S&P 500 (you can, of course, test on other ETFs/futures):

  • Go long if the WilliamsVixFix closed higher than the upper Bollinger Band using x days and y standard deviations. Enter at the close.
  • The exit is when today’s close is higher than yesterday’s close. Exit on the close.

The best result is when using a length of around 10 days and a standard deviation of 2. This gives 212 trades since 2000, an average of 0.5% per trade, and a profit factor of 2.05.

WiiliamsVixFix using PercentRank:

A while back we read an article about using PercentRank as a variable. By using a longer lookback period the following strategy seems to work reasonably well:

  • Go long if the WilliamsVixFix closed high on the x-day PercentRank. We used 98% as threshold and entry on the close.
  • The exit is when today’s close is higher than yesterday’s close. Exit on the close.

The strategy seems to work for a lookback period of up to 50 days. Using 10 days, ie. that the S&P 500 must close higher than 98% of the observations in the lookback period, produces 366 trades, 0.44% per trade, and a profit factor of 1.78.

This is the code for Amibroker:

HighestClose = HHV(close,22);
WilliamsVixFix = (((HighestClose – Low) / HighestClose)*100);
Plot(WilliamsVixFix,”WilliamsVixFix”,colorBlue,styleLine);

Buy = PercentRank(WilliamsVixFix,10) > 98 ;
BuyPrice=Close;
Sell = C > Ref(H,-1);
SellPrice = Close;

If we turn it upside down and go short if it closes in the lower 2% of the range, use a longer lookback period of 50 days, and exit when the close is below the 5-day moving average, the strategy produces 113 trades, 0.32% per trade, and the profit factor is 1.92.

The Amibroker code is like this:

Short = PercentRank(WilliamsVixFix,50) < 2;
shortPrice = C;
Cover = C < MA(C,5);
coverPrice = C ;

Conclusion:

The WilliamsVixFix is a synthetic VIX, but in reality, it’s just a measure of the price volatility over the last 22 days. However, because implied volatility is mostly a result of short-term volatility, the indicator resembles the VIX. The indicator is mean revertive. Thus, it seems to work reasonably well on stock indices. Surprisingly, it works pretty well on the short side. We didn’t test on other instruments.

 

Disclosure: We are not financial advisors. Please do your own due diligence and investment research or consult a financial professional. All articles are our opinions – they are not suggestions to buy or sell any securities.