Last week, in one of my never-ending attempts to beat the market without trying, I presented y’all with a simple sector rotation approach. Each year, you’d automatically buy the sector that did the worst, punting it out at the end of the year for that year’s worst performing sector.

Long story short, it didn’t work, at least over the last decade. The Canadian version of the test was killed in the last couple of years when it went all-in on gold during the worst possible time, actually finishing negative over the decade. The U.S. version of the test did much better, but was still a slight loser compared to just buying the S&P 500.

Undaunted, I’m back trying something similar. I think one of the reasons why the test didn’t outperform is because it was forced to buy something at the end of each year whether there were attractive sectors or not. What if I was more selective, and only bought when sectors had dipped 30% from their highs? Admittedly, 30% is an arbitrary number, but hey, we gotta start somewhere.

I want to make another change too. Instead of holding a sector for a year, instead I want to expand it to a  five year holding period. It takes longer to climb out of a mess than a year. We’ll run the test from 2000 to 2012, depending on how long these ETFs have been trading. As a reminder, here are the sectors we’re following:

  • Healthcare (IYH)
  • REITs (ICF)
  • Basic materials (IYM)
  • Consumer products (IYK)
  • Energy (IYE)
  • Transportation (IYT)
  • Financials (IYF)
  • Industrials (IYJ)
  • Technology (IYW)
  • Utilities (IDU)
  • Telecom (IYZ)

I just spent a bunch of time doing my work, and I’m not exactly chomping at the bit to write it all down again. I’ll just give you guys the summary.

In total, there were 18 instances where a sector fell 30% off of its highs. Four times out of 18 the sector was lower five years later, and two of those were from choosing tech and telecom back in 2000 and 2001 as the bubble was bursting. Averaged all out, and an investor would enjoy a return of 43.75% each time he did something like this.

To compare to the S&P 500, allow me to look at it over the following three periods:

  • 2001 to 2006
  • 2002 to 2007
  • mid-2008 to mid-2013

I’m cherry picking a bit here, but for the most part when one sector was down 30%, there ended up being several down. In the test I was doing a lot of buying during these three periods.

From 2001 to 2006, the S&P 500 was down 5.1%. From 2002 to 2007, it was up 22.1%. And from mid-2008 to mid-2013 the S&P 500 was up 27%.

Well, how about that. It turns out we might be onto something.

But are we? What happens if you would have just bought and held the S&P 500 back in 2001 until now? You’d be up 47% on the price appreciation alone, as well as collecting dividends good for ~2% per year. And you wouldn’t have had to sit in years with cash quietly accumulating on the sidelines. At least a buy and hold investor would be collecting those sweet candy dividends.

It’s not that the strategy wouldn’t have outperformed, because it sure looks like it did if we ignore any potential cash drag. The issue is with all the waiting. The test didn’t buy much of anything since 2008, which is a long-ass time to be sitting on the sidelines.

Rather than trying to replicate the test, you’re probably better off to look for sectors that have declined hugely and cherry pick the best companies in the space.


The lesson in this

The more time I devote to trying to use backtests to beat the market, the stupider I feel the whole exercise becomes.

Getting good results from the market is helpful in getting rich. But so are things like savings rates, making an above average income, and not spending all your money on smokes and hookers. In fact, I’d argue those factors are far more important than the return you might get.

Look at it this way. Say the market returns 8% annually over your investing life. You work hard and get 9% on your investments. It makes a difference, but by the time 40 years passes, all the market outperform does is make you more rich than you would be if you just matched the market. A 1% out performance isn’t going to make much difference in the scheme of things, especially for someone who would rather do other things than study balance sheets all day.

It’s fun for me to try to create scenarios where it’s easy to beat the market using some approach. But ultimately, it’s a waste of my time. The market is pretty darn efficient. I’m not saying the market is totally efficient, but 98% of the time stocks are priced about how they should be.

The lesson is ultimately stick to the basics. If you’re an index investor, keep plowing cash into those index funds and ETFs. If you’re a value guy, go find value stocks. Using backtests to establish much of anything is a waste of time. The best investors look forward.

Tell everyone, yo!