A better way to visualize hitters’ stats.
[click to enlarge]
(I included all A’s with at least 100 at bats, and Antonio Perez, just to show how awful a season he had…)
The blue bar represents batting average; the green bar on-base percentage; the red bar slugging, and the brown bar OPS.
The tick marks indicate the number of home runs hit.
The vertical black bars represent American League average numbers for AVG/OBP/SLG/OPS, in that order.
So, for instance, you can see that while Nick Swisher had a below average batting average, his OBP was higher than the league median, as were his slugging and OPS.
I’m still working on refining it, but I think this has a lot of potential.
Also, I’d love to be able to automate making it. Spending two hours getting up close and personal with Microsoft Paint wasn’t much fun. Anyone with some programming skills want to help out?
October 19th, 2006 at 12:17 am
cool way to look at offense comparisons, I like it. i linked to it at my site (conditionoakland.blogspot.com)
Chavez’s numbers were a pleasant surprise, compared to the league average.
December 14th, 2006 at 7:12 am
[…] I think I’ve settled on calling these graphics “Stat-meters”, but if a better name comes along, I’ll go with it. As I explained when I originally came up with this idea, I like that the ability to quickly glance at something and be able to tell just what kind of hitter someone is. I’ve been looking over rows of numerical stats for long enough that I think I can get an idea by just the numbers. But I’m sure not everyone can. Therefore, I came up with these graphics. […]
January 17th, 2007 at 3:35 pm
Personally, I’d recommend something else. What you’ve done is essentially a stacked bar chart. It’s undesirable IMO, because we’re not actually that good at judging the width of bars when they are not aligned at one end. Without looking at the numbers, could you tell me whether Milton Bradley or Bobby Kielty had a higher OPS? Could you tell me whether Mark Ellis’ OBP is above or below average?
It’s also debatable whether the chart is a wise use of space. All of the information above could have been presented in a table, in far less the amount of space that you’ve used. Moreover, I would actually even argue that it would be easier to read the table than it is to read your chart.
Don’t mean to give you a hard time. All of this of course, is just my opinion. Academically-trained statisticians (such as myself) tend to be fairly picky about the use and composition of statistical graphics; certainly standards and expectations are different in other fields (though of course, they are different in ways we tend to disapprove of).
January 17th, 2007 at 4:34 pm
In response to “a statistician”, yes, there were problems with this graphic. It was just a prototype. I’ve developed it further since then, as seen here:
http://thepastime.net/2006/12/14/stat-meters/