As the old adage goes, if you give a man a fish, he will eat for a day, but if you teach him how to fish, he will eat for a lifetime.
In the same vein, fantasy owners must know which players have learned to fish. They must be able to judge whether change is due to luck or skill.
A player's batting average is determined both by the number and type of balls he puts in play - whether they are line drives, ground balls or fly balls. Using statistics from The Hardball Times, we can access a player's balls-in-play data. A hitter's line-drive percentage (LD percentage) correlates highly with his batting average on balls in play (BABIP). In other words, the more balls a batter hits hard, the more hits he gets.
To find an individual hitter's expected BABIP, add .120 to his LD percentage. If the player's actual BABIP is higher or lower than expected, you can assume that he's probably been lucky or unlucky, and will eventually regress to the mean.
For example, in 2006 Gary Matthews Jr.'s strikeout and walk rates did not change, but his batting average jumped from his career mark of .266 to.313. But his LD percentage was not higher than normal - he didn't suddenly start the hitting the ball really hard. Rather, his actual BABIP of .349 was much higher than his expected BABIP of .318, meaning much of his success was the product of luck. Had his actual BABIP been in line with his expected BABIP, his batting average would have been .280 - a little higher than his career average, but much more in line with what we would expect, considering that he hit .275 two years earlier.
Sometimes a player can be lucky or unlucky as far as the number of fly balls that turn into homers, measured by home runs per fly ball (HR/F).
Different types of hitters should compile different HR/Fs: David Ortiz's is going to be a lot higher than David Eckstein's, for example. Therefore, to see if there is any good or bad luck in a player's HR/F rate, compare this season to the player's past. If other statistics, such as plate discipline, have remained stable, yet his HR/F is higher or lower than it has been in the past, one can assume that the player has been lucky or unlucky and will regress to the mean.
For example, in 2006, Jermaine Dye's fly-ball percentage remained almost exactly what it had been during the previous two seasons, but the percentage of fly balls leaving the park shot up to 23.3 percent. This is still a far cry from guys like David Ortiz (26.6 percent) or Travis Hafner (30.7 percent), but in order to appropriately utilize this statistic, players should only be compared with themselves. Thus, given that Dye is old enough to assume that he has not increased his skill, his home run total from 2006 was driven largely by luck and will likely regress to the mean in the future.
The more fly balls a batter hits, the more home runs he will hit. If a batter starts to hit more fly balls than he had previously, it can be assumed that he has developed a new skill.
For example, in 2005 Grady Sizemore hit 31.2 percent of his balls in play as fly balls. His fly balls left the park at a relatively normal rate of 15.1 percent; however, in 2006 Sizemore's fly ball percentage shot up to 46.9 percent, while the amount of fly balls leaving the park actually decreased back down to 12.0 percent.
Sizemore, it seems, has developed more of a fly-ball swing, which should lead to more home runs. The trend is continuing in the small sample size of the 2007 season: A whopping 61.9 percent of Sizemore's balls in play have been fly balls. While this number will certainly come down, it suggests that Sizemore's 2006 increase is legitimate and that an increased number of home runs in 2007 is highly plausible.
