Dynamic Option Valuation Applied to the Russell 2000

"Lighting makes no sound until it strikes."

-Martin Luther King Jr.

The Russell 2000 is interesting right now as the historic post-Covid run-up in small cap stocks - most notably the meme stocks of AMC and GME - has left the index in uncharted territory.... literally.

This gives us a chance to explore some dynamic option valuation techniques that I've previously discussed here and in presentations to the Houston Investor's Association.

Let's begin with the monthly chart...

Here, we get a good look at the run small cap's have had going all the way back to 2009 but especially in the post-Covid rally.  From the context of 'variance' to the index's 20 period moving average, this period represents the most over-extended the Russell has ever been from the trailing mean.

Here's the distribution of the historical 20 period variance measurements...

In this chart, five of the top seven measurements are from this year... that means the upside variance from this rally has been greater than the downside variance the index experienced in the wake of the Financial Crisis... which is just bananas.

Mean Reversion Trade

So turning to the options valuation perspective with the expectation that AMC and GME will eventually get called up to the Russell 1000, let's see how we can capture a mean-reversion in the Russell 2000...

Russell will publish its list of reconstituted indices on June 28th and the benchmark market capitalization necessary for a move into the 1000 is roughly $7.5 billion which stocks currently satisfy...  this gives us plenty of time for these trades to play out.

For the monthly chart, let's give the strategy seven candles (seven months) for the mean reversion to play out which conveniently brings us to the end of the year.  As of this writing, the Russell is trading at 2,268.97 so if we were to sell short the December 31st 2,300 call we get an option with a -49 delta and a bid price of $126.20.

 The seven month population data distribution has a mean return of 2.08% and a historical volatility of 11.67%. However, the option is pricing in an implied volatility of 21.21% which produces an expected value of +$5,191.

Another way to value this trade is create a data set of expected market performance given the current variance of the index.  Using historical 7-month forward returns when the variance exceeds +2, gives us the following  distribution...

 This data set, while sparsely populated, produces an expected return of -2.77% and a historical volatility of just 7.13% which produces an expected value of +$11,687.

Finally, when looking at the historical population distribution of daily returns over a 140-day period, the mean return was quite different from our monthly population. The expected return registers as a +5.86% mean versus +2.07% from the monthly data. To incorporate this perspective, we can run a Monte-Carlo simulation using the first two moments of the daily returns: 5.86% expected return and 11.20% historical volatility

 

This simulation produces an expected return of ($1,532).

Weekly Chart

Now, let's look at the Russell's weekly chart to see how it's formed a distinctly different technical pattern...

Here, we can see the massive MACD (Moving Average Convergence/Divergence) bubble that has formed in the weekly chart.  Similar to the variance we saw in the monthly chart, this bubble represents the largest recorded divergence of the 20 and 50 period moving averages in the history of the index!  With a catalyst event in site, we can easily map out a convergence strategy.

While the bubble has already begun to converge, should the index fall below the 20-week SMA, there's a technical expectation that it could use the averages as support and resistance levels.  Currently, the 20-period SMA is at 2234 and the 50-period SMA is at 1879.

To capture this, we can use the moving averages as points in a put spread trade.  If we give this trade 9 weeks to play out, we'll use the July 30th expirations to structure our trade by going long the 2230 put and shorting the 1880 to produce a max profit-to-loss structure of 5.5x.

We'll value this trade using two sets of assumptions. The first will be the historical performance of the index when it's trading above it's 20 and 50 period SMAs.  This produces a 9 week projection of a 1.21% expected return and a 17.8% historical vol...

This set of assumptions produces an expected return of ($2,483) per a max risk of $5,560.

However, given the unique chart setup, let's look at what we can expect to happen should the Russell move into the range below the two averages as we discussed.

Historically, there are not enough data points to support our second set of valuation assumptions but we can use the first moment of the data we do have to produce a simulation.  When the index has a MACD bubble of +2 or greater AND the price is between the moving averages we project the index to return -3.83% over a 9 week period.

Using this projection with the same 17.8% volatility expectation gives us the following simulated performance...

 

The expected return from this simulation is +$4,301.

It's also important to note that using spreads introduces an added timing dimension into the trade (as I noted in my criticism of Michael Kuhow's call spread hedge using the VIX)... but, shorting out-of-the-money puts makes far more sense than calls because of vol skew.  In the trade discussed above, the short put has an implied vol of 32.3% versus 22.3% for the long put.