Earlier in my series on using sabermetrics for fantasy baseball, we saw that fantasy owners generally prefer batters to hit the ball into the air in order to have a chance at a home run. Yet all fly balls are not equal for this purpose. A player can maximize his power production by pulling the ball into the air.

Today we'll look at how to utilize pull % for fantasy baseball analysis and, ultimately, to win your 2017 fantasy baseball leagues.

Editor's note: Be sure to check out all our strategy articles on how to win your fantasy baseball leagues: Our Sabermetrics series - Part 1: BABIP for Hitters, Part 2: HR/FB%, Part 3: Batted Ball Distribution, Part 4: Plate Discipline; Points League Primer; Using SIERA to Win Your League; How Your Brain Messes with Your Drafts; and Why You Shouldn't Overpay for Saves.

How to Interpret Pull%

In 2016, roughly 64 percent of all home runs were to the batter's pull side. Only 12 percent of homers went to the opposite field, with the remaining 24 percent going out in center. Last year saw a HR/FB explosion, so I looked into 2015 data to see if these rates were affected. They were not, as 63 percent were pulled, 12 percent went opposite field, and 25 percent left centerfield. These trends seem to be reliable each year.

In a way, this makes intuitive sense. Pulled baseballs tend to be hit with the highest exit velocity, making it easier for them to leave the stadium. The power alleys next to the foul poles on either side of the ballpark also present the shortest distance to the cheap seats. If a player's HR/FB dramatically improves, a change in approach involving more pulled baseballs could help explain why.

Boston's Xander Bogaerts provides a good 2016 illustration of this kind of change. In 2015, he pulled only 16.7 percent of his fly balls, producing a HR/FB of 5.3 percent and a total of seven dingers. He significantly upped his power game by pulling 28.1 percent of his flies last year, leading to a much higher 11.4 percent HR/FB and 21 bombs on the campaign. The increased power was not exclusively the result of the Pull% spike, as he upped his FB% as well (25.8 percent in 2015, 34.9 percent last year). It helps to validate his HR/FB increase, though.

This does not mean that you should ever look at raw Pull%. Of all pulled baseballs in 2016, 60 percent were grounders. Pulled grounders might have a higher exit velocity than other ground balls, but the shift still eats them up with minimal difficulty. They will never turn into home runs. By contrast, only 20.4 percent of pulled baseballs were classified as fly balls. Ideally, fantasy owners want their hitters to pull fly balls while limiting how often they roll grounders to their pull side.

This is much easier said than done, as all players pull many more grounders than flies. Let's consider Mark Trumbo as an example. His raw Pull% of 42 percent was only marginally higher than the league average 39.7 percent rate, and he pulled 54.9 percent of his grounders compared to 27.2 percent of his flies. At first glance, you might think that Trumbo was making himself vulnerable to the shift without significantly boosting his power potential.

That assumption would be wrong. The shift was designed for batters who pull much more than 54.9 percent of their ground balls, so Trumbo was able to torch it for a .319 average last season. Many batters fail to pull even 20 percent of their flies, so Trumbo rates as above average in that regard as well. Trumbo is the rare hitter who can go deep to any field, but 26 of his bombs were still pulled last year. Pulling more grounders than flies is far from a death sentence.

Conclusion

To sum up, pulled fly balls tend to perform better than other fly balls. This means that pulling more flies can produce an increased HR/FB, but you should never use raw Pull% to determine this. Most pulled balls are hit on the ground, where all of the exit velocity in the world cannot turn them into home runs. Therefore, you should filter a player's Pull% by batted ball type to produce the most reliable results. The next part of this series will look at lineup slot as a predictor of counting stats such as RBI and runs scored.