They aren't really doing a fair, apples-to-apples comparison.
First, they say that to "win" (i.e. to beat the market) each fund had to be in the top 25% over a given 12-month period. While I really wish they had used 12-month periods that exactly corresponded to calendar year (i.e. 12-31-20xx to 12-31-20xx) as it would make things a lot easier, the comparison can still be shown to be unfairly biased in favor of "the market index". If we take VIIIX (the super-low expense ratio institutional version of the S&P 500 index fund) as representing the market index, we can see that it was nowhere near the top 25% in the years 2009, 2010, 2012, and 2014 (it came in at the 52nd, 28th, 34th, and 41st percentiles respectively for those years).
Two, they are comparing apples and oranges by including smallcap funds. The proper comparison to the cap-weighted S&P 500 (or even to the cap-weighted total market) would be US large blend, US large value, and US large growth funds....maybe include US midcap funds that verge on being largecap (like RSP and PEY) as well.
Three, they neglected to compare over a full bull/bear cycle...or even several bull bear cycles. Start at, say, late 1994 or early 1995 (that way you get a bull market from 1995 to early 2000, a bear market from March 2000 to late 2002, a bull market from 2003 to October 2007, a bear market from late 2007 to March 2009, and a bull market from March 9 2009 to the present) and see if any funds beat the S&P or total market since then (and to avoid hindsight bias, look at only funds that returned more than the index overall from 1995-2002--both to get a decent cycle of both bull and bear market performance and to pick funds that were already "beating the market" and see how they did over the next bull/bear cycle from 2003 to the present--and then see how they did vs the index up until 3-16-2015).
Finally, I think the authors of that article were a little confused about what "beating the market" over the long haul actually means (or rather, what definition of "beating the market" actually matters over the long haul).
NO mutual fund or ETF (or closed end fund, or UIT, etc) will beat the market (or any other chosen benchmark)
EVERY year in a row...with very few exceptions, the funds that will tend to kick the market index's butt during bull runs will tend to lag during bear ones...just comparing various non-S&P 500 index large-cap funds, take a look at TWEIX or SEQUX vs RYOCX or SPECX for a good example of what I'm talking about; the former two lagged during the 1990s bull market but did comparatively (vs the S&P 500 or the cap weighted total stock market) very well from 2000 to 2009; the latter two crushed the market from 1995-1999 and 2003-2007 and March 2009 to the present but sucked during the 2000-2002 and late 2007-March 2009 bear market...but for the total period of 1-1-1995 to the present--which includes multiple bull/bear cycles--each of these four funds (or their lower expense ETF equivalents when they became available) creamed the S&P 500 or the cap-weighted total US stock market.