True Alpha – How the CLENOW portfolio makes a fool out of the S&P 500

Beating the index isn’t really all that hard. I’m going to show you a great way to beat the market while having some fun in the process. This is a brand new way of managing portfolios, and I’m going to explain every detail of it. The True Alpha method is easy to trade and is unique in offering personalized vanity portfolios. There is a considerable amount of research behind this approach and now it’s being made available to the general public.

Yes, dear reader, this is one of my more frivolous posts. The previous paragraph is both true and not true at the same time. Don’t worry though. Even when I’m joking, I’m dead serious. Or if it’s the other way around. I forget. Still, you may actually end up learning something from this post.

The basis of my brand new True Alpha(tm) strategy is a concept I’ve proudly stolen from a brief post by Joel Dickson at Vanguard a couple of years ago. It’s really a bit of an odd article. At first, he brings up a valid problem, only to quickly dismiss it and fall back into a general marketing pitch. I like the first part of his post, and I’m going to pick that ball right up again.

Achieving True Alpha(tm)

True Alpha(tm) is a revolutionary portfolio construction algorithm, selecting stocks on the sole basis of the first letter in the stock ticker. What, what? Yes, yes, calm down and listen.

To show you that we’re not just messing around here, let’s throw some simulations at the wall and see what sticks. As a baseline universe, we’ll use the S&P 500 Index. The simulations take historical index membership into account of course, including the graveyard. If it didn’t, it would be kinda useless as a simulation, wouldn’t it. Dividends and the usual stuff is naturally handled properly as well. For now, we’ll allocate equal cash to all positions.

The portfolio is rebalanced monthly to ensure that we maintain approximate cash parity. If not, the weights would go seriously out of whack with time.

First we’ll try the obvious. Let’s run a simulation that buys only stocks starting with a certain letter. Have you ever considered what initial ticker letter is the most beneficial for long term price development? Why not? You probably considered what month is the best, or what weekday.

The answer is of course Y. If you had bought all the stocks starting with Y, you would have more than doubled the return of the S&P 500 for the past 15 years. Other good letters include B, H, P, T and W, all of which would have yielded double digit annualized returns since 2000. However, you should obviously stay away from Q and Z, where the annualized return for the same period was negative.

The drawdowns are fairly evenly distributed, but again the Z stocks show their weakness. Clearly you should never buy stocks starting with Z.

Still, you don’t really get a good feeling for long term price development without a proper time series chart. Below you see the S&P 500 total return index as a thick black line, and each individual True Alpha(tm) strategy separately. As you see, almost all of our new strategies showed significant outperformance.

But, wait. It gets better! If we can do strategies on individual letters, why not on full words? What if we bought all the stocks starting with the letters in a particular word? Surely you must have already considered which word would be the best one. Well, let’s settle that once and for all.

You’re probably curious about how the current CLENOW portfolio would be composed. I know I am. As it turns out, the CLENOW portfolio is highly diversified. All in all, there are about 130 stocks in the current portfolio. The sector allocation is spread nicely on all 10 sectors.

Come on, stop it! This is nonsense!

Ok, fine. I’m just having a bit of fun.

All the results presented above are real. The simulations are done properly. It’s not a trick. Well, the trick is in the interpretations and conclusions. The Vanguard article ended with simply dismissing the results as unrealistic backtests and examples of misleading investment approaches. I was a little disappointed that Joel didn’t touch on the reasons why we see these kind of results. There is after all a clear reason why we see outperformance by simply buying all G stocks or the amazing CLENOW(tm) strategy. I’m convinced that the author is well aware of the issue I’m getting at and he probably had good reasons not to go into the underlying problem.

The elephant in the room isn’t the superior stock picking model shown above. It’s the highly inferior weighting system of the S&P 500, and most other equity indexes. Market cap weighting is the issue.

Market cap weighting means that the top few stocks totally control the index. For the S&P 500, Apple has a weight of 3.7%. That’s about the same as if you combine the bottom 100 stocks in the same index. The top 10 stocks in the index, Apple, Microsoft, Exxon, GE, J&J, Amazon, Wells Fargo, Berkshire and JPM together have a weight of 16.5%. That’s similar to the weight of the bottom 300 stocks all together.

Market cap indexes are not diversified. We all pretend there are 500 stocks in there, but only a handful actually matter. To make things worse, the stocks that do matter are the largest companies in the world. How much more upside is there once you’ve got a company with a market cap equivalent of the GDP of Switzerland? Is it more likely that those giants will continue up, or that some other company will gain in value? Which has more possible upside?

The truth is that you can pick stocks at random and have a very high probability of outperforming the market. As long as you pick enough stocks to gain a decent diversification. Yes, I’ve done the research on that, and yes I will post an article on that soon too.

But what about the CLENOW portfolio?

As much as I would like to claim that the magic is in the stocks starting with the letters C, L, E, N, O and W, the reason why this works has a different explanation. There are enough stocks starting with those letters to gain sufficiently broad diversification. Buying those stocks will result in a broad participation in the overall market, and simply by not using market cap weightings we get instant outperformance. In the simulations above, I used cash parity for simplicity. You’ll see more or less the same (though slightly better) results with vola parity.

And the underperformance of Z stocks? Well, how many stocks start with Z? Not very many, is it. If you buy and hold only Z stocks, you’ll usually get a portfolio of 2-3 stocks. Therefore, the result will be pure luck. In this case, bad luck.

But still, I take pride in knowing that you will see a strong outperformance simply by purchasing stocks based on my name. Surely that means something. After all, it’s not like you can take any ridiculous word and come up with similar results.