A random number generator can beat your mutual fund. Given a choice between a random portfolio and a mutual fund, I’ll go with the randomizer every day of the week and twice on Sundays. You think I’m joking? I’m not joking.
Trashing the mutual fund industry is almost like beating a dead horse. Except of course that it’s a thriving, multi billion dollar dead horse. Still, pointing out that all mutual funds fail to do their job is like kicking in open doors. We all know that mutual funds underperform. The SPIVA Scorecards tracks and compares their performance to the respective benchmarks and the results are devastating. On any given 3 year period, around 80-90% of all mutual funds fail to beat their benchmark.
It’s not really the fault of the mutual fund managers. The entire construct of mutual funds prevent them from having a fair chance of matching the benchmark returns. Then again, according to Gordon Gekko, mutual fund managers are sheep, and sheep get slaughtered. Yes, I may have dissed trading quotes in the past, but movie quotes are always applicable.
Ok, so mutual funds are as useless as a… Well, they’re just useless. Leave it at that. But we could get the index return with a passive ETF index tracker. Buying a passive index tracker will give you an overwhelming probability of higher return than buying a mutual fund. If you buy the ETF, you will get the index. There will be a tiny fee, but else you will get exactly the index. But the real question is if you really want the index.
Most indexes are designed as a way to gauge the performance of a market or sector. They were not designed as investment strategies. Yet, that’s how they are used. As investment strategies, indexes are horribly bad and easy to beat.
There are several things wrong with indexes as investment strategies, but the most important point is about position sizing. In index terminology, it’s called weighting. Most modern indexes are market cap weighted, meaning that the higher the total value of a company, the larger the weight. Makes sense, right?
No, not really. What this system means in practice is that Apple has a weight of 3.9% while Diamond Offshore Drilling has a weight of 0.010105%. If both shares make a 3% price move today, AAPL will give a positive index contribution of 0.12%, while DO will contribute with 0.0003%. For DO to contribute with 12 basis points, as AAPL just did, it would have to show a gain of 1,200%. Why do we even bother to pretend that there are 500 stocks in the S&P 500? In reality, we’re just buying a hand full of the world’s largest companies.
Before you say that DO must be a really tiny little company that doesn’t deserve more weight, remember that it’s still a part of the S&P 500 large cap index. It’s a four billion dollar company.
The method of using market cap weighting holds back performance dramatically. If a company is worth four billion dollars, it can still double. Several times. But if a company is worth 750 billion dollars, more than double of the world’s second largest company and more than any other company in the history of companies, how likely is it to double?
Point being, the largest companies are not the ones likely to show the best gains.
I promised you a randomizer that beats the markets, didn’t I? So let’s get back to that.Here’s our market beating trading system:
Well, that just ridiculous, isn’t it? Buying 50 random stocks each month? Sound stupid.
Obviously, running such a strategy once is pointless. Anything can happen. It is after all random. So let’s run it lots of times. Let’s run it 500 times. I made a dummy input variable in the simulation software and used the optimizer to try 500 iterations of this unused variable. Finally a valid use for the optimizer functionality!
The chart below shows a subset of 50 of these runs. Throwing 500 lines in there wouldn’t add anything. I’ve included both the best run and the worst and made sure that the 50 iterations here are representative to illustrate the results of this exercise.
In case you’re curious about those drawdowns and how they relate to the long term annualized return, I’ll throw in another image here below. It shows you a scatter with drawdown and annaulized returns and it should speak loudly all by itself.
Obviously, I’m not actually suggesting that you go and buy 50 random stocks every month. All I want to show is that even such an absurd method has an overwhelming chance of beating the market. What you want to do, is to find a method to ensure that your results are similar to those top performing iterations above. If you can do that, you’ll outperform almost everyone in the world.
Now let me ask you again: Do you really want to invest in the market index?