Joseph Javier Perla

I bought and held QID for a few weeks until the NASDAQ dropped. QID is a twice leveraged short of the NASDAQ. The NASDAQ was overpriced, I made an easy eleven percent.

But I want to ensure that I’m actually doing well, and that ostensible success is not due just to availability bias. I want to calculate my annualized rate of return. This is not trivial, though, because I added money to my Scottrade account three different times. I didn’t find anything which can help me solve my problem immediately, so I put together an Excel spreadsheet.

I learned a number of interesting things. First, my returns are very good. 24.7% annualized returns (including commissions and trading costs but sans taxes). I trade Apple, Google, Amazon, and some market ETF’s: QID, SPY, IWO. Here is a graph of my returns since I started investing in January 2006:

The bump there isn’t due to market swings or anything like that. My return stream is uncorrelated with the market. (Most of the time I am out of the market, and I have been short the market about as much as I have been long it.) I did so well at first, I got very cocky and started investing in Ford and AT&T, companies about which I know very little. Subtract out those two months of overzealousness, and my trades become both much higher and all positive (0 bad trades).

The Barclay Hedge Fund Index, http://barclayhedge.com/, notes that hedge funds have gained, year to date, 12.1% while I have gained, year to date, 29.2%.

I earned 50.4% since I started investing in January 2006. As a comparison, your bank probably gave you 1%. A money market could have tossed you 10%. A mutual fund might have left you 16%. The normal benchmark, the S&P 500, earned just 15%. In the same period, the NASDAQ earned just 17%. XLK, a tech sector index fund, earned only 25%. I note the tech sector because I’m more likely to invest in tech stocks since I know more about them. A blanket strategy of buying and holding technology stocks would not have created the returns that I have. My excess returns are alpha.

Not only are my excess returns alpha, I’m pretty sure that most or all of my returns (and moreso if you include making up for stupidity in buying Ford) are alpha. Each data point on the graph above is a trade. I’ve made fewer than a couple dozen trades in 22 months. Less than one a month. I held each one for a maximum of a couple of weeks, usually for a period of just 3 business days (enough time for my funds to settle.) That means that I am almost never exposed to the market, so I can’t have any beta. To factor out much of the beta I scooped up as I dip into trades, I have also been short the market, twice leveraged, for a few weeks. For the most part, these returns are pure alpha.

Now, I could have gotten much better returns by, for example, buying and holding Google, Amazon, and Apple. Google doubled, Amazon doubled, and Apple tripled in value. But, first, I am not that patient, and, second, holding the stocks would have left me open to lots of market beta (like the recent tech correction), and also the up and down swings of everyday rumblings. I prefer more consistent, less negative jumps. I don’t know when the next bubble is, or when the next bubble burst is, I just know when a stock will jump by 10% tomorrow.

Half of my trades have been fantastic moves (7%+). Almost the rest of the other half didn’t do much (+/- 2%). The only two non-trivial bad trades were during my cocky period and were disastrous. See a scatter plot of my trade returns below, the trades are plotted over time from left to right.

As you can see, there are clearly three clusters. The clusters are unbelievably well-defined and well-separated. It gives me the impression that I might be able to figure out, before I trade, if the trade will turn out to be a fantastic 7%+ trade, or a ho-hum +/- 2% trade. I continue to maintain that the bottom group is a fluke and that they will never happen again. What’s nice to see here is that I have consistently achieved fantastic trades at all times throughout the past two years. There is no lucky streak. Moreover, the only very bad trades are clustered around a time period when I specifically know I started investing unwisely. I’m surprised by the very, very low noise.

Another interesting note about the graph directly above is that I seem to be trending toward riskier investments, although this might just be my mind searching for patterns in random noise. If you look at the profitable cluster, it is trending up over time. I make higher returns for my good trades. This would be a good thing, except that my ho-hum cluster seems to trend down over time. This divergence implies I might be moving towards riskier trades and not just increasing skill at choosing better trades. Right now, when I am wrong about a trade, it usually just does nothing around 0, but I wouldn’t want to take a slight negative penalty every time. I’ll have to watch out for that and study my trades more.

I would calculate a Sharpe ratio or information ratio for my returns, but that would be inappropriate. To do so would be a rape of statistics. My returns are definitely not normally distributed. The distribution of returns for my trades are two-humped and highly left-skewed. You can clearly tell from the histogram below:

Frankly, it would make no sense to measure my returns over risk, since my risk is nearly entirely on the upside. Moreover, return over risk when risk is all upside would penalize me for having even more positive returns.

Now that I see how my trades stack up, I’m convinced that my returns are real and due to skill. The results are so clear that I don’t even have to run any statistical tests. I can improve, however. Warren Buffet thinks he can earn 50% annually if he had less than a million dollars. I think that’s right. I was sliding 50% annually for the first 10 months until I screwed up. I violated Warren Buffet’s first rule: don’t lose money. I also violated his second rule: don’t lose money. There are a few ways for me to make more money:

First, I can see how to identify the better trades in the ways I described above by analyzing the best ones. Second, I can open a margin account to double my returns or just so that I don’t have to wait until funds settle so I can take my money in and out faster. Finally, I will open a Roth-IRA to take care of those pesky taxes.