More on the SMA Rule: Daily Analysis of Russell 1000 and Russell 2000

Following up on comments from my post on the SMA rule, I dug into applying the SMA rule using daily data on the Russell indexes. I only looked at the Russell 1000 and 2000 for this post.

Russell’s website provides daily index values since June 1995, so the earliest decision you could have made using a 200-day SMA would have been mid-March, 1996. In the charts below, the index starts at 100 on March 14. The “Buy and Hold” series simply invest in and earn the respective index returns, while the “SMA strategy” series adhere to the 200-day SMA rule: invest fully whenever the most recent index value is at or above its 200-day SMA, otherwise invest in cash. I assumed Cash earned 90-day T-bill returns. These charts assumed no slippage or transaction costs.

Clearly the SMA strategy as depicted above would have been preferable for the Russell 1000, (it certainly will by the time I’m done writing this post!) The SMA strategy is well behind the Russell 2000 Buy/Hold strategy – the Russell 2000 would have to fall at least another 30 percentage points for the SMA strategy to boast a higher return.

Implementation Costs are Probably Higher than You Think
The unexamined element in this analysis is implementation cost – and this is my only significant complaint about the book The Ivy Portfolio. The authors accurately note that commissions anymore are negligible, and that the ETFs they cite have small bid-ask spreads.

While all that is correct one of the problems with a lot of backtested strategy results, no matter how simple or complicated the strategy itself, is the assumption that you can transact at the close of the day (or the month) without any bid/ask spread or slippage. At the close without a bid-ask spread, and without any HFTs pushing the prices around.

I created a second set of SMA strategy series, assuming a 1-day lag before trading. The results for the Russell 1000 are even better, while the results for the Russell 2000 are even worse.

	Annualized Returns		Annualized St. Deviation		Worst Drawdown
	Russell 1000	Russell 2000	Russell 1000	Russell 2000	Russell 1000	Russell 2000
Buy and Hold	6.36%	6.73%	21.0%	24.5%	-55.4%	-58.9%
SMA Strategy	6.30%	3.96%	12.5%	15.2%	-26.9%	-44.2%
SMA (1-day lag)	7.69%	2.94%	12.5%	15.3%	-22.8%	-40.5%
Source: PortfolioWizards, Russell

That the SMA rule appears to reduce volatility seems well supported by the data examined here. Whether it results in greater accumulation of wealth depends. The Ivy Portfolio describes different rotation scenarios – if you’re going to try them yourself, make sure you understand how efficiently you could implement the strategy yourself and what to expect.