Which Moving Average Strategy Crushes MA Crossovers? This...

22 Mar 2023

If you read last week’s article, you saw results for the famous (or infamous!) moving average crossover.

It bombed vs buy and hold over the last 10 years, even when using take-profit and stop-loss levels.

So, how do moving average bounces perform with the same exit levels?

That’s what we’re testing today…


The Trading Truth Test Setup

Our setup is the same as last time, except we just need one moving average.

• Market: the S&P 500 index (using SPY to trade it)
  
  
• Timeframe: Jan 1, 2013 to January 31, 2023
  
  
• Bar interval: 30 minutes.
  
  
• Moving averages: 50 bars (simple moving averages, meaning every bar gets equal weight, unlike with exponential)
  
  
• Starting Equity: $25,000
  
  
• Max % of Equity Per Trade: 3%
  
  
• Commissions, fees and taxes. To keep things super simple, we’re again assuming these are all zero.
  

How Is “Bounce” Defined?

Traders look for ricochets off moving averages in a bajillion ways.

Let’s be real: most just eyeball charts without real rules.

That won’t work for rigorous renegades like us, though. :-)

So, for Test A, here’s how we defined a bounce when price is above the MA:

• A price bar’s low is above the moving average
  
  
• The next price bar’s low touches or pierces the MA for one bar. (A close below the MA is ok.)
  
  
• The bar touching or piercing the MA can’t go more than 0.5% below the MA
  
  
• The next bar’s low must again be above the MA.
  

Flip these rules for a bounce when price is below the MA.

I know, I know… many bounces don’t happen until 2 bars (sometimes more) hover on or below the MA. We're keeping the criteria here simple.

We define Test B the same way, except with an extra filter:

The 50-bar MA needs to be sloping up at least 0.5% over the last 25 bars for long trades (and -0.5% or less for short trades).

Alrighty, now, let’s check out the results…


The Test Results

Like with the MA crossovers test before, let’s first look at how plain ol’ buy and hold did over the same period.

If you parked your cash in the S&P 500, your money would be worth 2.9 times as much by the end. Pretty good.

So how did Test A do?

The ending equity after 10 years was $40,112.74, 1.6 times your money with 24.27% max drawdown.

Test B, in which we only took trades if the MA was sloping (trending) enough, we ended with $43,149.00, for a 1.7x return with only 17.75% max drawdown.

That makes the return on risk much better than Test A.

Yet, while these returns beat the pants off of our MA crossover tests, boring old buy and hold still whooped them both.


What Would You Change About This Bounce Strategy?

Trade a different market?

Longer or shorter time frame?

Tweak how a bounce is defined?

Comment below to share. Also, please let me know: what else do you want to see tested?