Hello fellow RotoBallers! Sabermetrics have become an integral tool for fantasy baseball draft prep, but a concise resource for understanding the basics can be difficult to find.

This series attempts to define and explain all of the metrics fantasy owners may find useful, citing examples of how to use them in the process. Twenty advanced degrees in applied mathematics are not required to use advanced metrics effectively, and this will be a no math zone. We also won't bring in many of the metrics that are synonymous with advanced stats, most notably the fantasy useless WAR, or Wins Above Replacement.

Instead, the focus will be on sabermetric statistics and ideas that are useful for predicting the standard stats the vast majority of leagues care about, like batting average.

BABIP for Hitters

The most accessible of the fantasy relevant advanced stats is BABIP, or Batting Average on Balls In Play. It simply measures a player's batting average on balls in play, with outcomes such as strikeouts and home runs removed from consideration. In general, the league average hovers around .300, a nice round number to remember. Many know BABIP as an approximation of luck, with either a very high or very low number indicative of a major batting average regression in the future. That is partially correct--the stat can be used to predict batting average fluctuations. However, a player's skills may allow him to run a better than average BABIP, or doom him to a consistently below average figure. One example of this is Jonathan Villar.

Villar never looked like a hitter who could approach a .290 BA over 600 plate appearances before 2016. Last year's improvement can be found in Villar's .373 BABIP, a number significantly higher than the league average .300 figure. If we assume that Villar's BABIP will regress to .300, Villar's batting average and fantasy value plummet. That would be a hasty assumption though.

Villar is an elite speedster--this is why he was picked up as a steals guy in the first place. It makes sense that someone with Villar's wheels could beat out more base hits than other players, while most catchers would lag in this regard. Therefore, an established player's baseline BABIP should not be the league average .300, but whatever that specific player's career BABIP is. Villar's career BABIP is .329, clearly indicating a sustainable ability to beat the league average .300. Of course, .329 is still a lot lower than the .373 figure from last year. If we assume Villar can beat the average BABIP, how do we know if he was in fact lucky?

The answer is to look at BABIP by batted ball type. Villar gets his speedster hits exclusively on grounders, as running really fast does nothing to prevent a fielder from catching a ball in the air. While the league averages around a .250 BABIP on grounders, Villar posted a .313 mark on them last year. Villar also had a BABIP of .300 on his grounders in 2015. Therefore, we can conclude that Villar will outperform the league average BABIP on ground balls as long as he has his speed, and that this ability was not the cause of his 2016 batting average spike because he did it in 2015 too.

Comparing BABIPs by batted ball type year over year, Villar's average spike was due to his performance on fly balls and line drives. In 2015, Villar posted just a .111 BABIP on flies. In 2016, this number jumped to .165, over 50 points better. His career rate is .195, suggesting that his 2016 production should at least repeat this year. His liners were a little more luck driven, posting a .728 BABIP against a .694 career rate. Their regression to the mean will probably prevent a repeat of last season's .373 BABIP overall, but Villar still projects for .330+, significantly above the league average.

The same trend is possible in a negative way. For example, Baltimore's slugging first baseman, Chris Davis, is well known for being an all or nothing slugger that pulls the ball at every opportunity. This makes him susceptible to the shift, as the infield defense knows where the ball is likely to go and can set up accordingly. He also lacks the speed to beat out infield hits most other major leaguers can. These factors figure to hurt his BABIP on grounders, and Davis's .157 last year indicates that it did. In 2015, it was .162, and Davis's career average is just .190.

Clearly, projecting regression toward the league average would be wrong, as his pull tendencies and subpar speed allow the defense to consistently perform better than average against him. Davis's overall BABIP was .279 last year,  a number that should be expected moving forward due to his poor production on ground balls.

To conclude, BABIP can be used to indirectly measure a player's batting average luck by comparing it not to the league average of .300 but to an established player's career number. Foot speed, batted ball authority, LD%, and defensive positioning all give players some ability to manipulate BABIP. Players with these skills may still overachieve, and this regression can be predicted by examining BABIP by batted ball type. Younger players without an established baseline are generally regressed to the league average, but these predictions are less reliable than those based on a player's personal history. Next time, we'll look at a stat called the BABIP of power, HR/FB.