FanGraphs Splits 3: Back to the Minors

Four days into the splits debut at FanGraphs, I think we understand the need to exercise caution with using these small sample numbers. The left/right splits, for instance, need to be regressed to the league averages until a high number of data has been collected. However, if we trust Minor League numbers enough to help us project players at the Major League level, I believe we can use splits at the Minor League level to help in our understanding of a player. Obviously, studies need to be done to find the correlation of splits at the Minor League and Major League level before I can say this with any certainty. Luckily, we now have the data to do so, thanks to the splits here at FanGraphs, and the wealth of information at Jeff Sackmann’s I’m not here to do that kind of heavy lifting today, but to show the direction of analysis I believe we can achieve with the numbers currently available.

We are taught to essentially ignore the splits posted by rookies, because we would need to regress those numbers so much to have any certainty in them. But if a player had 1,200 more plate appearances in the Minor Leagues, shouldn’t we trust those splits more than a broader league average? Today, I took the seven rookies that had at least 300 PAs in 2009 and 700 more in the Minor Leagues, and I want to compare the splits congruent between this site and Minor League Splits: batted ball info by handedness. I believe this will inform our comprehension of each player’s Major League splits, and our projection of their performance going forward. (LD-L is a player’s line drive percentage versus left-handed pitchers. You can figure out the rest.)

Chris Coghlan, 2B, Florida Marlins

Level   AB    LD-L%   GB-L%   FB-L%   LD-R%   GB-R%   FB-R%
MLB     504   18.9    47.2    34.0    23.7    47.8    28.5
MiLB    1117  17.4    45.5    36.7    18.2    43.9    37.7

What Changed: Many fewer fly balls vs. RHPs.

Going Forward: Coghlan’s batted ball profile against LHPs barely changed from the Minor Leagues to the Majors, and I see little reason he can’t continue to be successful against southpaws. Against right-handers, I don’t believe he’s a hitter that will routinely post a 20% line drive rate, so I think you’ll see some regression in BABIP this season, and I’ll think we’ll have that to point to.

Elvis Andrus, SS, Texas Rangers

Level   AB    LD-L%   GB-L%   FB-L%   LD-R%   GB-R%   FB-R%
MLB     480   27.2    48.5    24.3    20.1    57.4    22.5
MiLB    1600  17.2    58.8    23.7    14.4    58.3    27.0

What Changed: Traded 10% of groundballs for line drives vs. LHPs.

Going Forward: I would like to believe Andrus’ aging and development led to the change, but he’s just not a guy that can sustain a 27% line drive rate against any kind of pitcher. He should maintain his performance against right-handed pitching, if not improve, to compensate a regression vs. LHPs

Colby Rasmus, CF, St. Louis Cardinals

Level   AB    LD-L%   GB-L%   FB-L%   LD-R%   GB-R%   FB-R%
MLB     474   14.1    41.0    44.9    21.1    33.0    45.9
MiLB    1569  15.1    34.8    49.8    18.0    33.8    47.7

What Changed: More groundballs, less elevation vs. LHPs.

Going Forward: Scouts have criticized Rasmus’ approach against southpaws for years, and a change in process didn’t do much to his results in 2009. The Cardinals should see if Rasmus can start making harder contact against lefties early in the season, but if contention stands in the way of Rasmus’ development, he should be platooned quickly.

Andrew McCutchen, CF, Pittsburgh Pirates

Level   AB    LD-L%   GB-L%   FB-L%   LD-R%   GB-R%   FB-R%
MLB     433   20.5    37.5    42.0    17.9    43.9    38.2
MiLB    1964  15.8    40.8    43.4    15.5    49.1    34.9 

What Changed: More line drives.

Going Forward: This is the most difficult case, because I don’t see a real reason for regression. McCutchen has been a scouts favorite for a long time, dating back to some of his struggles in the lower levels. Nothing about this batted ball profile is out of line — he’ll be a great case for personal development vs. regression to Minor League standard.

Gerardo Parra, RF, Arizona Diamondbacks

Level   AB    LD-L%   GB-L%   FB-L%   LD-R%   GB-R%   FB-R%
MLB     455   11.7    64.9    23.4    20.1    49.7    30.3
MiLB    1402   8.7    61.4    29.9    12.7    54.3    32.9

What Changed: More Power vs. RHP, not enough vs. LHP.

Going Forward: Parra should be platooned from Day One of Spring Training, no regression to the league, or further chances at development needed. The stigma against platoon players is such that people will think I “hate” Parra, but it’s not true — he’s becoming a legitimate threat against right-handed pitching. You could do worse than a Jacque Jones or Michael Tucker career.

Everth Cabrera, SS, San Diego Padres

Level   AB    LD-L%   GB-L%   FB-L%   LD-R%   GB-R%   FB-R%
MLB     377   10.1    72.2    17.7    16.7    58.9    24.5
MiLB    877   13.6    67.1    18.8    12.1    59.8    27.6

What Changed: Even more groundballs vs. LHP.

Going Forward: Cabrera will be this type of player forever — bashing the ball into the ground — more when batting left-handed than the opposite. With game-changing speed, Cabrera just needs to right his defensive woes to become a very manageable shortstop. And he’ll need it, because you’re not going to win long-term with this offensive strategy.

Nolan Reimold, OF, Baltimore Orioles

Level   AB    LD-L%   GB-L%   FB-L%   LD-R%   GB-R%   FB-R%
MLB     358   16.3    51.0    32.7    13.3    46.7    40.0
MiLB    1512  14.6    34.8    50.6    11.1    40.4    48.4

What Changed: Drastic fluctuations in flyball-to-groundball ratios against both pitchers.

Going Forward: Essentially, it looks like Reimold translates a whole lot of infield flies (career 18.0% in the minors) into groundballs against left-handed pitchers. This actually is a step forward, and as a result, his numbers on fly balls (.453 wOBA) were very good now that he wasn’t mixing in an infield flies a fifth of the time. Still, a high IFFB% against right-handed pitchers, so that split bears watching in 2010.

Note: I generally ignored the percents difference between fly balls and line drives between the Minor League and Major League level. It seems that so-called “fliners” are classified different, and I believe are marked as line drives less in the Minor Leagues. Something else to study, I suppose.

Tomorrow: Pitchers!

Print This Post

9 Responses to “FanGraphs Splits 3: Back to the Minors”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Obligatory Sox Fan says:

    Super interesting! I look forward to tomorrow’s post.

    Vote -1 Vote +1

  2. BenJ says:

    Bryan- be very, very careful when comparing MLB and MiLB splits.

    Presumably the MLB splits are from BIS, and the MiLB splits are from elsewhere.
    First, the different data sources might have different definitions of LD, FB, etc. One source might have 15% LDs, another at 20% or 25%. The GB/LD distinction often gets overlooked, but it’s not always clear cut either. BIS has a specific definitions for when a GB becomes a LD, and it might be a different definition from the minor league data. I don’t know how true this is, but you should be careful.

    Secondly, most (if not all?) MiLB data is collected by park-specific stringers. The nice thing about BIS data is how there are 15+ different scorers rotated among different games every night. With park-specific stringers, you will get some stringers who are have specific LD vs. FB tendencies, or to a lesser extent LD vs. GB tendencies. These scorer tendencies would carry over into the player data.

    With these two possibilities, you should be very careful citing splits as you do here.

    Vote -1 Vote +1

    • Bryan Smith says:

      Yes. Did you see the note at the end of the article? I tried to never comment on any changes in LD vs. FB, because I knew that going into the article. I do have some faith that GB vs. “elevation”, and that’s where most of my comments were centered. This article also isn’t what I would call hard-hitting analysis … I’m writing to see what comes of it, and to show what we might be able to do in the future with more surefire data.

      Carefulness has been exercised.

      Vote -1 Vote +1

  3. BenJ says:

    Yes, I saw your note. Since you presented an entire article with charts that compare LD and FB data, I felt that you hadn’t been careful enough. Additionally, I wanted to add some insight to the potential LD% differences you mentioned.

    And as I said, the LD/GB distinction you focused on is more complicated as well. When does a GB become a LD? If it hits the ground “in front of” the fielder, is it automatically a GB? What if the fielder is playing deeper or shallower than usual? Then the definition is dependent on the fielder’s positioning. Or if the ball is hit equidistant between the drawn-in 3B and deep SS? Then you’ve got the scorer guessing. Park-specific scorers could be doing any number of things. Hopefully, this isn’t too big of an issue, but I don’t know. There’s definitely going to be some variation.

    Lastly, I don’t know the Minor League Splits data source, but I don’t think anybody but BIS records “fliners”.

    Vote -1 Vote +1

    • Bryan Smith says:

      I didn’t mean actual fliners, as I meant the distinction is probably different between those in-between flyballs and line drives. That’s not written real clearly. My fault.

      I didn’t give any thought to the LD/GB distinction. That’s a very good point.

      Vote -1 Vote +1

  4. tim says:

    I think your underselling Everth Cabrera by stating “he’ll be this type of player forever”. He jumped straight from A ball to the major leagues, and his LD% increased every month through august. In sep/oct he hit a wall and sort of fell apart. I wouldn’t be suprised to see him develop his offensive game into something more along the lines of Rafael Furcal.

    Vote -1 Vote +1

    • Bryan Smith says:

      He might not always hit 70% of balls into the ground, but it’s not a stretch to say his game will always be rooted in groundballs, no?

      Vote -1 Vote +1

      • tim says:

        He probably will always hit a lot of ground balls. But if he continues to take a walk, and the 7% LD % he posted in sept/oct doesn’t reflect his true ability, then

        “you’re not going to win long-term with this offensive strategy”

        doesn’t seem appropriate.

        Vote -1 Vote +1

  5. Bryan Smith says:

    Readers: The follow-up to this won’t be until next week. I apologize, but good news: I’m prepping for my podcast debut!

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>