Let's talk about measuring performance.

I'm not a fan of Morningstar performance ratings. I don't rely on a fund's reported performance numbers or the Wall Street Journal's numbers either.

But if the prevailing methods of measuring performance are so poor, why does everyone use them, and what's a better method?

Once upon a time, the SEC standardized performance reporting by requiring all funds to report their returns based on the same time periods: one, three, five, and ten years, or since inception, calculated through the last day of each calendar quarter.

The idea was that if all funds reported performance the same way, investors could compare them more easily and fairly.

The problem is, the standardized periods used to measure the returns are fixed in a single span of time, based on calendar quarters. That's not how you, or I, or most investors invest, however.

Let me put it another way. How many times have you made your entire investment in a fund at the very beginning or end of a calendar quarter?

Do you always invest on March 31st, June 30th, September 30th, and December 31st? I don't.

Because these calendar, quarter-end periods are fixed, mutual fund performance can vary dramatically from one reporting period to the next.

For example, take a look at Vanguard Capital Value. At the end of 2007's third quarter the fund's 3-year return was 14.6%, but at the end of the fourth quarter its 3-year return was just 4.6%. And at the end of 2008's first quarter Capital Value's 3-year return was a negative 0.3%.

You heard me right…