Visual Baseball:  Introducing Bat Slicer

Happy Friday. Here’s something new that I’ve been working on. It’s called Bat Slicer and the idea is to visually represent a hitters’ bat. I’m hoping the visuals are fairly self explanatory, but just in case: We start with strikeouts (represented by a hole in the bat, and then show walks, singles, doubles & triples, and home runs in successive rings. These stats are based on production per plate appearance, with a minimum of 200 PA’s, and then converted into percentile rankings, with 100% being the best, 0% the worst. The legend on the bottom right of the visual gives you a sense of scale. Click here to see a slide show of lots more players.

Let’s start with a player who’s bat we’d expect to look pretty fearsome.

image

In contrast, here’s a guy who carries a pretty light stick.

image

Here’s a guy who is doing some serious damage, albeit with a lot of strikeouts.

image

And here’s an “all or nothing guy.” Lots of home runs and strikeouts. He also enjoys a nice walk down to first base.

image

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: And That Happened
Next: If you give Roger Clemens a steroid »

Comments

  1. Jon Weisman said...

    This reminds me in a funny way of playing All-Star Baseball as a kid in the ‘70s – the playing card were circles.

  2. Dave Studeman said...

    Very neat, Kevin.  I can’t decide if I like the percentile approach—perhaps a linear weighted approach would be better.

  3. dfan said...

    I have a few issues with this but the biggest one is that the area occupied by home runs (say), which is what someone’s eye will pay attention to, has only a passing correlation to the number of home runs the player actually hit. It’s proportional to (XP+HRP)^2 – XP^2 = HRP^2 + 2 * XP * HRP, where HRP = homer percentile and XP = all other percentiles added up. So not only are you 1) showing people a percentile rather than a real number (this by itself is not so terrible and could even be useful), 2) you’re squaring it to produce a number with completely meaningless units, and then 3) adding it to a number that depends on all his other stats.

    Look at it this way; draw a Bat Slicer for someone who is 100th percentile for HR and BB and 0th for every other stat. Now draw one for the same guy but reverse the order of the stats so that BB is on the outside. Which one looks like he’s a better home run hitter?  One of them has 3 times as much HR area as BB area, and the other has 3 times as much BB area as HR area, but they’re the same batter.

  4. Kevin Dame said...

    I played with weighted data but found that the visuals didn’t look different enough.  That’s the main reason why I go with percentile rankings so often…

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>

Current day month ye@r *