Ultimate Zone Rating Infield Components: 2006-2010

by Justin Bopp on Apr 1, 2011 8:30 AM EDT in Visuals Above Replacement

Enlarges.

This is more of a graphic dump, an attempt to get this out of my system, than anything. I started almost two years ago trying to develop a better visual for defensive metrics, but many lessons since teach me that jacking around with data presentation that doesn't necessarily improve understanding is time wasted.

So, over time my goal changed from "change the world with a defensive visual!" to "show combined team defense" which eventually lead to the "TargetView," which I won't link back to for reasons of embarrassment. Basically, I really liked rainbow skittles colors. The design kinda worked but wasn't terribly useful.

Which brings us to today's visual, which is the last gasp of what I'll probably do with UZR for some time. What you're looking at is a data capture from all of 2006 to last fall, 2010. All players included have a minimum of 100 innings at the queried position, and positions were divided out between infield and outfield (forthcoming).

I then totaled the various components up and created a pie chart showing the general component percentage, and then attempted to create a scale to show the basic minimums and maximums for each. But honestly, the logic behind doing so now escapes me, other than originally thinking it would look pretty cool if I could remove non-reached sections for a specific player view. I'm really unhappy with the scale, but perhaps you guys might appreciate it more than I.

The result is somewhat revelatory in that it shows where infielders get the majority of their defensive value, and somewhat instructive for visual primer purposes--though I'm not certain either were accomplished terribly well.

Here's how this might be applied to a specific player:

Like I said, I'm not sure how effective the scale is. I think I dislike it because it's just inelegant. Anyway, I'd love your feedback. Maybe we can turn this into something valuable with some insight that I'm overlooking.

Tweet Comment 16 comments | 0 recs |

Do you like this story?

Comments

.

My gut reaction is that it might help to gray out the outer area of the pie chart that falls beyond the player’s rating.

Twitter | KC Baseball | I70 Baseball

by KCAaron on Apr 1, 2011 10:22 AM EDT reply actions

Agreed, with another suggestion

This may be beyond the capabilities of the graphics package you’re using, but if you could somehow make the highlighted areas appear larger (like they’re coming out of the overall circle) I think that’d help.

by jwiscarson on Apr 1, 2011 10:56 AM EDT up reply actions

Now that should be easy! I can do something really neat to make that happen.

See Data Differently: Beyond the Box Score | @justinbopp
My designs are now available to print! Check them out.

by Justin Bopp on Apr 1, 2011 11:09 AM EDT up reply actions

Aaron's idea is good, and I've tried various ways of approaching exactly that.

The issue, I think, is that it leaves little real sense of scale — not that this particular pie chart is doing that well to begin with.

See Data Differently: Beyond the Box Score | @justinbopp
My designs are now available to print! Check them out.

by Justin Bopp on Apr 1, 2011 11:10 AM EDT up reply actions

Speaking of which, Bill Petti accurately points out that

the 3 different scales for each ‘axis’ (if you will) is neither helpful nor necessarily accurate.

Man I hate pie charts. Seems like the only real way to make them accurate is keep exactly one variable. Trying to show both percentage value (why use the average infielder’s percentage to show a single players? what’s the point?) AND the actual value for each metric is just a complete mess.

I realize I posted this, but I really do hate it — and not for visual reasons at all, but for data presentation and accuracy purposes. Bill suggested a fan chart might be more effective, but I think we’ll still see the same incongruency between scales.

I do love the feedback, so keep it coming. Maybe we can salvage this project or inspire somone to create something better.

See Data Differently: Beyond the Box Score | @justinbopp
My designs are now available to print! Check them out.

by Justin Bopp on Apr 1, 2011 11:15 AM EDT up reply actions

I think a fan chart like the ones used for ESPN's division preview in the magazine would work

If you use a fan and scale the radius of each component the same I think it will work.

Go to page 65 and read from there .

Columnist at Beyond the Box Score

by Bill Petti on Apr 1, 2011 12:05 PM EDT up reply actions

Don't know if this really relates to your visual at all, but it just popped into my head:

Can’t you present UZR data as a percentage above/below the league mean for each rating? So, let’s say the average SS accrues 5 runs in 500 innings in range, and Player X accrued 6, he’s 20% better than the positional league mean for that metric. Then continue on with the other stuff like error runs and double play runs.

Don’t know the best way to visualize that, though.

Me Elsewhere:

Bless You Boys (Tigers Blog) / Beyond the Boxscore / My Michigan State (and Big Ten) Baseball Blog. / Western Michigan University Hoops / my Last.fm account.

by Mike Rogers on Apr 1, 2011 2:31 PM EDT reply actions

For what it's worth…

The bad: I can’t really tell what this graphic means. Well, I get the first one (I think), but the numbers on the Rollins one don’t feel right. I feel like I’m confusing myself.

The good: Visually, this might be one of the prettiest infographics I’ve ever seen.

On Twitter: @baseballtwit

by adarowski on Apr 1, 2011 4:02 PM EDT reply actions

Completely agreed, though your last sentence is a bit strong.

See Data Differently: Beyond the Box Score | @justinbopp
My designs are now available to print! Check them out.

by Justin Bopp on Apr 1, 2011 4:26 PM EDT up reply actions

Seriously, Adam, I thought you read my posts?

I kid, I kid!

Columnist at Beyond the Box Score

by Bill Petti on Apr 1, 2011 5:07 PM EDT up reply actions

maybe yours HANDSOME,

bet you didn’t think of that.

See Data Differently: Beyond the Box Score | @justinbopp
My designs are now available to print! Check them out.

by Justin Bopp on Apr 1, 2011 5:10 PM EDT up reply actions

*are

See Data Differently: Beyond the Box Score | @justinbopp
My designs are now available to print! Check them out.

by Justin Bopp on Apr 1, 2011 5:12 PM EDT up reply actions

Excellent point

Columnist at Beyond the Box Score

by Bill Petti on Apr 1, 2011 5:24 PM EDT up reply actions

Bill, yours are "reliable pretty". You know, marriage material.

With Justin—you never know what you’re gonna get. Sometimes you get a perfect combination of style and substance. Sometimes you get something like this that is a bit messed up in the head, but it’s so damn hot you try to tell yourself you can “fix it”.

;) (I wink, I wink)

On Twitter: @baseballtwit

by adarowski on Apr 2, 2011 12:09 AM EDT up reply actions