Let the Bloodhound System Sniff Out Profits

It's hard to make a strategy that will beat the S&P 500. Even if you do make one, there's no guarantee it will outperform in the future. But with the right tools, finding new winning strategies is a whole lot easier.

Later in this article, I'll talk about one strategy that could outperform the broader market in the future. But first, I want to show you a tool called the Bloodhound System that gives you the ability to try out and evaluate your own strategies.

Going back to the future
Bloodhound's two creators met while working at supercomputer-maker Cray Research, and both have many years of experience in the fields of supercomputing and computer-based simulation, including the Bloodhound missile system, from which their product takes its name.

That kind of experience shows through in the product. Like RobotDough, Bloodhound features a powerful stock screener, but its key feature is an extremely robust backtester, which most screeners leave out.

It's easy to come up with a great-sounding strategy and run it through a screener to find the right stocks. It is not so easy to come up with one that is actually profitable over time. Joel Greenblatt didn't just pull the Magic Formula out of nowhere and decide it was a good idea. He used a database of stock market information and tested the Formula to see how it would perform over a 17-year period.

Bloodhound has such a database, going back to 1987, containing only stocks that have complete information during that time. That means recent IPOs are included, but OTC stocks that might skip a few financial reporting periods aren't. Using that database, Bloodhound will let you set up a screen, add in trading parameters like minimum holding period and broker commissions, and see how your strategy would have performed. It will then give you just about every statistic about the strategy that you could ask for, like average drawdown, Sharpe ratio, and win-loss ratio, compared against similar statistics for the S&P 500.

See how you measure up
In my attempt to come up with a market-beating strategy, I built this screen based on sound principles and tinkered with it for several hours until it worked.

• Market cap over $300 million, to weed out any crazy penny stocks.
• TTM cash flow at least the same as a year ago, to weed out the losers.
• Enterprise value to free cash flow ratio less than 12, to focus on cheap companies.
• Dividend yield at least 2%, to find companies that return that cash flow to shareholders.

The backtest for this strategy assumed that you bought the top 10 stocks from this screen based on 10-year growth in cash flow per share and then held them for at least one year. Over the last 20 years, the strategy would have returned 17.8% annually, versus the S&P 500's 9%, with only a small amount of additional downside risk.

Of course, just because the strategy has worked until now doesn't mean that it will in the future. But if you wanted to use it today, you'd end up buying these stocks:

Source: Bloodhound, author computations.

The bottom line
Bloodhound is a powerful tool with some great features. It's more mechanical in nature than RobotDough, focusing its strength on statistical analysis, while the latter offers tools for deeper fundamental research, like forensic accounting notes. Bloodhound is also fairly expensive -- $1,295 annually is the cheapest plan that includes most of the strategy features. This might be too steep for a lot of new investors, but for those who can afford it, Bloodhound is a good tool to have.