Track record investing is the third and last detrimental investment trap I’ll discuss here. Like stock picking and market timing track record investing is as it sounds, using track records or past history to determine whether or not a money manager or specific fund is a good investment.

Track record investing is tricky because we evaluate track records all the time, they’re a normal part of our lives. I use Apple computers instead of windows computer for a few reasons, not least among them is the track record of Apple devices to outlast their windows counterparts. I buy specific products on Amazon only after reading far too many reviews to determine the track record of said product. Track records are not bad, they’re actually super helpful. But, when it comes to investing, it’s dangerous to rely on the same mechanism we use to evaluate Amazon products to decide whether or not a money manager or mutual fund is a good investment. Here are a few reasons why:

The number one disclaimer in the world of investing is “past performance is not indicative of future results.” Why is that phrase posted everywhere? Well, first, because it’s the law. But more importantly, it’s because we so badly want to use track records to determine which mutual funds to invest in, and it doesn’t work. Like we learned about stock picking, people can’t consistently predict the market. If a mutual fund manager does well one year it does not indicate that the manager has figured something out, it means he or she got lucky. People can make guesses about the future, and they do, but no one knows the future, and the market will move based on things that will happen in the future.

A fun example of this past performance issue is Morningstar’s famous five star rating system. Morningstar has long rated funds year after year with one to five star ratings, and it’s a big deal to earn the coveted five stars. However, since 2010 Morningstar has stated that the cost of a fund is a better predictor of its future success than Morningstar’s star rating, essentially admitting that the rating system is useless for evaluating future performance. Track records cannot help when it comes to choosing specific investments.

Another issue with this past performance stuff has to do with the data consumers actually see. Investment companies are notorious for practicing something called selection bias. Selection bias means that if a mutual fund does really poorly, the investment company will kill it and expunge its data from their records so that it has no effect on the overall statistics consumers see. Unfortunately those dead mutual funds often have a drastic effect on returns for real people. The worst 200 funds that were killed between 1923 and 2016 averaged a cumulative return total of -81%. Not only does past performance have nothing to do with future results, but the data consumers see only muddies the water.

So should we shun all data when it comes to investing? Definitely not. It’s important to make a distinction between past performance data, which relates to specific funds or managers or even whole firms, and long term market data, which teaches us about how investing works and where returns come from. I use historical market data all the time, not as if those are returns my firm has achieved since the 1920’s, but to help people understand what to expect from the market, and to show them why diversification is so important. So just remember, past performance really isn’t indicative of future results. Look for an advisor who understands the academics of investing, not one who shovels you past performance data.